SCIENTIFIC PRINCIPLES FOR ENGINEERING AUTOMATIC CONTROL EQUIPMENT

• Document Type: CREST
• Collection: General CIA Records
• Document Number (FOIA) /ESDN (CREST): CIA-RDP81-01043R004500020003-3
• Release Decision: RIPPUB
• Original Classification: K
• Document Page Count: 202
• Document Creation Date: December 23, 2016
• Document Release Date: May 13, 2014
• Sequence Number: 3
• Publication Date: November 1, 1959
• Content Type: REPORT

Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
STAT
TRM1SLOT1011
SCIENTIFIC PRINCIPLES FOR ENGINEERING
AUTOMATIC CONTROL EQUIPMENT
By Various Authors
November 1959
195 Pages
1
STAT
PREPARED BY
LIAISON OFFICE
TECHNICAL INFORMATION CENTER
MCLTD
WRIGHT-PATTERSON AIR FORCE BASE. OHIO
?
1
-t-
STAT
- Declassified in Part - Sanitized Copy Approved for Release
50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
STAT
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Wrzft
Nauchnyye Osnovy Postroyeniya Tekhnicheskikh Sredstv Avtomatiki
Izdatelfstvo Akademii Nauk SSSR
Moskva ? 1957
STAT
Declassified in Part - Sanitized Copy Approved for Release 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
TABLE OF CONTENTS
Page
Foreword
B. S. Sotskov. Main trends in the development of the theory and principles
for the design and construction of automatic and remote control elements 1
N. N. Shumilovskiy and B. S. Sinitsin. The basic problems of the theory of
automatic measurements 14
B. T. Kolomiyets. Semiconductors and their application in automation ? ? ? ? 35
B. M. Tecukevich. Comment 54
N. P. Udalov. Address in article by B. T. Kolomiyets 57
D. V. Zernov, M. I. Yelinson, and A. Y. Kharchenko. Prospects for the develop-
ment of new types of electronic equipment for automatic and telemechanical
installations 59
L. I. Gutenmakher. Prospects for the application of magnetic and capacitive
units and elements in the automatic systems 84
B. I. Verkhovskiy, G. G. Iordan, and N. N. Shumilovskiy. Outlook for the util-
ization of radioactive isotopes in the automation of industrial processes 115
Ye. A. Nekhayevskiy. Statement 130
A. M. Prokhorov. Radio-frequency spectroscopy and the possibility of its use
in the field of analysis 133
L.-D. Rozenberg. Acoustical methods for measuring nonacoustical quantities 138
V. I. Taltroze. Mass-spectrometer as an instrument for the control of indus-
trial processes 153
N. N. Shumilovskiy and R. I. Stakhovskiy. Automatic gas analyzers of the mass-
spectrometer type 176
P. Ye. Rybchinskiy. Address 194
R. I. Stakhovskiy. Address 194
STAT
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
?
FOREWORD
A special session of the Academy of Sciences of the U.S.S.R. dealing with
scientific problems of industrial automation was held October 15 - 20, 1956.
The Trudy sessii /Proceedings of the Session' were published by the Publishing
Hogse of the Soviet Academy of Sciences. Altogether, seven volumes containing
the materials of the plenary session and the work of the various section have been
published as follows:
1. Plenary session (general problems).
2. Basio problems in automatic regulation and control.
3. Scientific principles for engineering automatic control equipment.
4. Scientific problems in remote control of industrial prooesses.
5. Scientific and teohnical problems in automated electric drives.
6. Theory and design calculation methods for automatic machines and auto-
matio lines.
7. Complete automation of industrial processes.
The present (third) volume of Trudy contains the reports and discussions of
the section on the scientific principles for engineering automatic control equip-
ment.
The section considered ten reports on the most significant and promising trends
in the development of the latest types of automatic control devices.
Tne reports made by B. S. Sotskov, N. N. Snumilovskiy and B. S. Sinitsin
dealt with the general problems of developing the technical means of automatic con-
trol. The reports of L. I. Gutenmakher, B. I. Kolonlyits, V. M. Tuchkevich, D. V.
2ernov, M. I. Yelinson and A. M. Kharcnenko concerned tne outlook for developing new
types of electronic, semiconductor, magnetic, and dielectric noncontact devices.
The present state and prospects for the application of radioactivity for con-
trolling industrial processes were discussed by B. I. Verkhovskiy, N. N. Shumilovskiy
and G. G. Iordan. The report by L. D. Rozenberg was devoted to acoustic and ultra-
sonic dev STAT
11
Declassified in Part - Sanitized Copy Approved for Release 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Problems in the industrial application of radio-frequency and mass cpectros-
copy were analyzed by A. M. Prokhorov, V. L. Taltroze, N. N. Shumilovskiy and
R. I. Szaknovskiy. The problems posed in the various reports were thoroughly dis-
cussed.
These reports and the discussions on them indicate the basic scientific tasks
in developing the latest types of automatic control devices.
111
STAT
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
MAIN TRENDS IN THE DEVELOPYENT.OF THE THEORY AND PRINCIPLES
FOR THE DESIGN AND CONSTRUCTION OF AUTOMATIC AND REMOTE CONTROL ELEMENTS
By
B. S. Sotskov
The present report seeks to examine the most significant and general
problems in developing the theory and principles for engineering the elements
and systems of automatic control, guidance, protection and regulation.
The work of designing and Perfecting new types of devices and systems
of automatic and remote control is proceeding at present on a wide scale.
Still, many problems of industrial automation cannot yet be solved because
of the absence of elements necessary for controlling a number of important
technological parameters. Higher parameters -- mechanical (pressures,
velocities), thermal (temperatures and thermal capacities), and electric
(voltages, power)-- in new technological processes, and their rapid rate
of change call for new types of pickups, amplifiers, and final control
elements, which must possess quick response and greater limits and precision
of sensing and transformation.
The importance and complexity of operations performed by automatic
systems demand a high degree of stability of characteristics and performance
reliability in these systems and their components.
The requirements for longer life and efficiency of individual components
and complete systems are constantly on the rise. The ...,-reat need of automatic
and remote control elements requires the standardization of the basic types
of such elements so as to make their mass production possible. Finally, the
necessity of selecting components and devices for systems of automatic con?
trol, guidance and regulation, requires the adoption of criteria for the
evaluation of their-properties, performance and parameters.
The above factors will determine the main trends and tasks of research
and development in this field for the next 10 to 15 years.
The basic theoretical task is the elaboration of a theory of transformations
STAT
Declassified in Part - Sanitized Copy Approved for Release 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
in elements and devices of automatic control, guidance, and regulation, and a
theory of performance reliability of elements considered singly and in systems.
The development of a general theory of transformations must comprise a sys-
tematic examination of every possible physical transformation, whether direct or
indirect (intermediary), so as to determine the possible relationships between
varied input and output values. At the same time, the effect of all other values
and factors must be shown so as to establish the constant value and possible error
of the basic transformation selected for building the element.
The mathematical aspect of the theory of transformations should enable us to
determine (by mathematical analysis) or derive (by mathematical synthesis) the re-
quired functional correlations between the values of input and output on the basis
of known correlations of the individual elements and their components (parts). The
theory of transformation accuracy should also be developed, taking into account the
effect of internal and external factors on the characteristics and parameters of
the transducer. It is necessary to work out both the static and dynamic problems
of transformation precision of elements, devices, circuits and systems. This is an
immense task in itself and is of basic importance to measurement technology as a
whole. The study of aynamic accuracy is linkee closely with the analysis of the
dynamics of complex systems of elements having a different physical nature and a
complex relationship.
Finally, we must solve mathematically the problem of obtaining constant value
(autonomicity) of the required transformation by setting up a suitable structural
scheme of the transducer -- differential, logometric, or compensatory.
We still find ourselves in the stage of individual (sometimes quite interest-
ing) studies aealing with particular problems of the general theory of transforma-
tions. Preliminary work is now being done toward formulating the basic elements.
It is necessary to accelerate this maturing process because of the immense importance
of this basic theoretical foundation for the further developnent of the means of
automatic and remote control.
2
STAT
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
?
The development of the theory of performance reliability of components and
systems of automatic and remote control is another very important problem. Here
we must investigate the effect of external factors and specific operational, mechan-
ical, thermal and electric loads on the stability of the performance and lift of
the elements. We must clarify and define the problems of the reliability of per-
formance of components with continuous transformation (failure, excessive ceviation
of cnAracteristics) and with discrete transformation (failure, faulty action). Of
paramount importance is the development of methods of determining tne general (re-
sultant) reliability of complex systems and devices under conditions of series,
parallel and compound connections of the various elements in the functional chain,
and when the reliability of these elements differs. In evaluating the general per-
formance reliability of a system consisting of a whole series of functional chains
varying in their bearing on the operation of the entire system, we must introduce
the necessary corrective factors which make it possible to evaluate ;he real effect
of the failure of an element in each chain on the operation of the system as a
whole. Finally, we must develop the methods of selecting elements (or subassemblies
of the device) and their operating conditions for a given degree of performance
reliability of the elements and of the system as a whole.
At present we have at our disposal only scattered studies dealing with specific
problems; there has been no programmed and intensive work laying the theoretical
foundations for determining the performance reliability of automatic and remote con-
trol equipment.
Equal in importance to the foregoing general questions of theory is the re-
search and development of new principles for building the elements and systems used
in automatic control, guidance, and regulation, as well as the further development
of analysis and design methods for known types of elements and systems. Several
large groups of problems that call for research might be mentioned at this point.
The first group consists of problems growing out of the utilization of the
physical properties of conducting, semiconducting and dielectric materials needed
3
STAT
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
for building the components of automatic and remote control elements.
Utilization of these properties makes it possible to obtain elements with a
direct functional connection oetween changes in the various actuuting p.-..ysics.1 quanti-
ties anc tne electric and magnetic properties and parameters of the element. Stu-
dies on this group of problems should include the following:
The investigation of changes in the electric and magnetic properties and param-
eters of metals and their alloys, combinations and compositions resulting from the
action of various physical (electrical, mk;netic, luminous, mechanical, thermal,
etc.) and physico-chemical factors;
The
(whether
physical
The
investigation of changes in the electric
one-, two-, or multicomponent types)
factors;
investigation of changes of electric
properties
resulting from
of semiconductors
the action
of various
and magnetic properties of dielectrics
brought about by the action of various physical factors;
The development of automatic and remote control elements and systems by the
utilization of changes in the electric and magnetic properties of metals, semicon-
ductors and dielectrics resulting frcm the action of various physical factors.
The second group of problems relates to research and development in the field
of elements witn a number of intermediary transformations. Use must be made of
such elerrents when the elements of the preceding group - those with a direct func-
tional transformation - are not available or cannot be used (for reasons of constant
value, range of sensitivity or the degree of transformation accuracy).
This group of elements was the first historically and is the most widespread.
However, modern requirements with reuect ..c) precision, working range, sensitivity
and quick action, call for further researcn in this field. The emergence of new
materials witn superior mechanical, magnetic, electric, thermal and chemical proper-
ties is substantially modifying the design of known elements and calls for new
design and analysis methods.
Recently there has been a number of new ideas with regard to engineering
4
STAT
Declassified in Part - Sanitized Copy Approved for Release 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
????
elements of this group; for instance, the replacement of gyroscopes by gyrotrons
in measuring angular displacements in space, angular velocities and accelerations;
further, using gyroscopic devices to measure the velocity of fluids; and finally,
electromagnetic flowmeters, etc.
Research in this field should include:
The development of general methods of analyzing and designing mechanical and
electromechanical systems with given terminal static and dynamic parameters and
characteristics;
The development of more precise and suitable methods for computing complex
magnetic and electromagnetic systems for constant, impulsive, and alternating field
conditions and the development of new structural arrangements for the devices;
The development of more precise and suitable methods for computing complex
thermal systems of various types and the development of new structural arrangements
for the devices;
The development of precise and suitable methods for computing mechanical mea-
suring systems used in determining mechanical parameters and the parameters of
gaseous and fluid media, and the development of new structural arrangements for the
devices.
The diversity of operating conditions, range of measurement and the required
precision and speed of operation of the components of automatic and remote control
equipment make it necessary to seek more and more new principles for their construc-
tion. This is particularly true in the case of the primary elements of automatic
systems, the pickups, which detect changes in the controlled quantities.
Here, the development of new types of components makes use, among other things,
of various auxiliary physical processes and chemical reactions for purposes of
processing and control.
Progress has been reported recently in a number of lines in the application of
auxiliary physical procuses, mainly in connection with the utilization of the ab-
norntion and reflection properties of various radiations: acoustic, optical and
5
STAT
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3 ?
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
?
electromagnetic, radioactive, etc. Positive results have been achieved in these
lines and should substantially influence further developments there.
Here, work must be done in the following areas:
In the field of acoustic radiations as related to the design of automatic con-
trol and regulation equipment, it will be necessary:
To investigate selective and integral reflection and absorption of media, mix-
tures, and compounds for the various ranges of acoustic oscillations;
To study the characteristics of the radiators and receivers upon changes in
the composition and physical properties of the surrounding medium;
To develop new design principles and arrangements for devices for the control
of the composition or physical properties of media by utilizing changes in the
characteristics of raaiation, reflection and absorption.
In the field of optical and electromagnetic raciations, it will be necessary:
To investigate the absorptive and reflective capacities of various media and
of their mixtures and compounds with respect to visible and invisible electromag-
netic radiations under conditions of changed physical quantities (temperature,
humidity, pressure, etc.);
To develop new design principles and arrangements for devices with direct or
indirect utilization of visible or invisible electromagnetic radiations for con-
trolling the composition of media or physical quantities;
To develop new indicators of visible or invisible electromagnetic radiations
utilizing changes (during irradiation) of the properties and parameters of conduct-
ing, semiconducting and dielectric materials; further, to develop and improve ex-
isting principles of engineering radiation indicators.
Radioactivity has found wide application in recent years in the engineering of
various devices of automatic control and regulation. In order to further develop
equipment utilizing ot.-
,
and neutron radiation, it will be necessary:
To investigate selective reflection, absorption and transformation of 01:- ,
and neutron radiations by various media and tneir mixtures and
6
STAT
Declassified in Part - Sanitized Copy Approved for Release 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
STAT
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
.4
compounds (in order to develop new methods of analyzing the composition or the
changes of physical parameters of a substance);
To elaborate new principles for the direct or indirect utilization of ct- ,
and neutron radiations in building devices for controlling changes
in the composition of a substance or its physical parameters;
To develop new indicators of radioactivity utilizing changes in the electric
parameters of conductors, semiconductors or dielectrics producad by CI-
and neutron radiations.
A group closely related to the preceding groups consists of devices using
principles of spectroscopy, radio-frequency spectroscopy, gamma- and neutronoscopy,
and mass-spectroscopy. In order to develop these principles for building control
devices, it is necessary:
To investigate the spectral characteristics of various media, their mixtures
and compounds;
To develop new engineering principles and arrangements for automatic devices
with direct or indirect utilization of the principles of spectroscopy, radio-
frequency spectroscopy, gamma- and neutronoscopy and mass-spectroscopy to control
the composition of complex media;
To develop indicators and computing elements for various spectroscopic devices
for automatio recording, processing of results and conversion of results to the kind
of material suitable for further use.
A special group of elements and systems of automatic control and regulation
consists of devices which make use of auxiliary chemical reactions resulting in
modified optical, electrical or other physical properties of the medium being an-
alyzed or of special indicators (whether solid, fluid or gaseous). Modification of
the indicated ehysical properties is utilized by the device for further functional
transformation and analysis.
The elaboration of the above principles for building devices requires:
The investigation of the effect of the chemical composition of various
7
Declassified in Part - Sanitized Copy Approved for Release 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
STAT
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
substances on the physical (optical, electrical, etc.) properties of the substances
themselves and on special indicator substances;
The development of new principles and arrangements for automatic devices for
controlling the composition of substances utilizing changes in the physical proper-
ties of the controlled media and special indicator substances.
While the elements (pickups) used to detect the changes in the parameters of
the controlled phenomenon or medium, must by their design and purpose be quite di-
verse, the reverse is true of the elements of automatic systems that are next in
sequence - the intermediary and final control elements. It is preferable that these
elements used to further amplify, distribute and transform the value and form of the
impulse detected by the pickup be, whenever possible, partly or fully standardized.
Depending on the type of energy used for internal transformations in the auto-
matic system, there are electrical, pneumatic, hydraulic and mechanical intermediary
and final control elements.
The most developed electrical group of intermediary and final control elements
has undergone in the course of recent decades a transition from contact and electro-
mechanical types of elements to electronic and noncontact magnetic, semiconductor
and dielectric elements (amplifiers, stabilizers, distributors, storage and comput-
ing devices,etc.). This does not mean, naturally, that contact and electromechani-
cal elements have become less important and are no longer in use. Far from it.
What has happened is that their field of application has been somewhat narrowed,
while the most efficient of these devices are still being used in the transformation
of electrical energy into mechanical energy.
However, the mushrooming development cf electronic elements - amplifiers,,
stabilizers, distributors, computing devices, and the recent noncontact magnetic,
semiconauctor and dielectric elements - has changed radically the engineering of
intermediary and final control elements of automatic systems.
10 are witnessing at present a clearly-crystallized trend toward the creation
of automatic systems by a schematic combination of standard electronic or, whenever
8
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
?
called for, magnetic, semiconductor or dielectric components. This method has be-
come prevalent in the development of modern systems, and only in oases where this
method results in the construction of extremely complex and consequently not fully
dependable systems or fails for some reason to bring the expected results,has it
been necessary to build special electronic or electromechanical devices. Suffice it
to say that in modern automatic control systems for large bombers tne number of
electron tubes reaches 7 or 8 thousand and in the case of electronic computing
machines the number ranges between 500 and 18,000. With such great numbers of elec-
tron tubes in use, their main defects become readily apparent, namely their short
life and undependability. This compels us to search for other solutions of the
problem based on special electronic and electromechanical devices.
Magnetic elements of control systems are being developed along two major lines:
magnetic components devised to utilize changes in the magnetic permeability of fer-
romagnetic materials due to an alternating magnetic field in the presence of a
magnetizing constant magnetic field of a different magnitude, and magnetic compon-
ents devised to utilize a large residual induction and a sharply defined differen-
tial in the magnetic permeabilities of separate segments of the curve of hysteresis
cycle. .The first principle is embodied in amplifiers, stabilizers, regulators,
distributors and generators. The second principle is used in distributors and im-
pulse storage devices.
Analogous elements (amplifiers, current stabilizers of voltage and power, dis-
tributors, impulse storage devices) may be built of materials with nonlinear resis-
tance (the varistor type) or with nonlinear dielectric properties. Both of these
groups of elements are finding application at the present time. Each of these
(with nonlinear p , A and 6 ) has its special cnaracteristics as regards pro-
perties and application, and hence they co=plemen each other well as control sys-
tem elements.
The following are the principal tasks confronting the engineers of the above
types of intermediary elements-
9
STAT
neclassified in Part - Sanitized Copy Approved for Release 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
STAT
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
1?.
As regards electronic elements of automatic control, it will be necessary:
To develop long-lived (witn service of 30,000 - 60,000 hours) and stable
electronic (and ionic) tubes, particularly multi-electrode and multiple-unit tubes,
and also electron tubes with radioactive cathodes;
To develop special types of electronic and ionic devices: distributors (com-
mutators), functional trarsducers, counting tubes, electronic and ionic converters
of magnetic, electric, mechanical and other quantities.
In the field of nonlinear magnetic, semiconductor and dielectric elements, it
will be necessary:
To develop new magnetic, semiconductor and dielectric materials with sharply
pronounced nonlinear characteristics, and to investigate their properties;
To develop types of elements and methods .for their design and analysis; to
analyze their static ana dynamic cne.racteristics and parameters; to develop methods
for analyzing complex systems containing nonlinear controlled elements;
To develop rethods for engineering simple and complex systems with nonlinear
elements for given terminal static and dynamic characteristics.
In addition to the ceve.lopment of noncontact components mentioned above, the
search fcr new types of elements is continuing.
This search is directed especially toward the following types:
Elements capable of utilizing the Hall effect. Application of various semi-
conductors such as InSb, InAs, F.E;Se makes it possible to obtain electromotive forces
of the order of tens of millivolts for values of flux density up to 10,000 gauss.
This will permit a wide use of such semi-conductors for the construction of ampli-
fying and computing devices;
Elements eased on the galvanomagnetic effect, i.e., the effect of the change
of electric resistance in a magnetic field. This effect, observable in bismuth, was
formerly applied only in tne com,truction of cievices for measuring ragnetic fields.
In a number of new semiconauctors, InSb, InAs, etc., tnis effect is very pronounced,
Ara TI1 n1 i fi erc rninnuti rc, Ft 1 prnerts and pickups may be built widen utilize the
10
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
STAT
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
o?
galvanamagnetic effect of these semiconductors;
Elements based on the utilization of the phenomenon of superconductivity (i.e,
the stepwise drop of the electric resistance of a whole group of materials when
temperature is lowered almost to absolute zero, or more precisely, to a few degrees
on the Kelvin scale) and on the effect of the external magnetic field on the tem-
perature at which the phenomenon occurs. Components of tnis type may be used as
current and power amplifiers and as components of distributing and computing de-
vices.
The above are but isolated examples of the great,still unused opportunities
latent in the achievements of our physicists and chemists.
In analyzing output (final control) elements one should mention first of all
the appearance of a whole series of high-powered controlled electronic devices (of
the ignitron type or similar to it) and of controlled saturable reactors (high-
powered magnetic amplifiers) for the purpose of controlling the flux of electric
energy.
Numerous types of final control units are being devised for the control of
?
mechanical displacements and speed. Tnese may be divided into two basic groups.
The units of the first group have controlled electric motors or electromagnetic cir-
cuits for alternating or direct current. Ti'e should note at this point the recent
development Of various types of two-phase electric motors for increased alternating
current frequencies and stepwise electromagnetic final control units controlled by
discrete inputs of current.
The second group is comprised of final control units with controlled clutches.
The largest group of controlled clutches consists of electromagnetic clutches:
friction, magnetic-coupling and asynchronous types. Here, particular mention should
be made of electromagnetic clutches with a ferromagnetic filler which are new both
in design and principle of operation and in their characteristics and the possiuilities
of using them for the construction of final control units. For the proper develop-
ment of these clutches, it will be necessary to carry out a series of serious
11
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
l?
investigations of the physical and cf,emical properties of ferromagnetic fillers.
Mention should also be made of clutches with magnetic coupling. Uncier condi-
tions of controlled magnetic flux they make possible final control units with con-
trolled starting and stopping. In constant magnetic flux (e.g. that around a
permanent magnet), these clutches represent a ccrvenient ccrponent for various de-
vices whenever it is necessary to transmit force or moment through some immobile
nonmagnetic partition.
A significant role is played at present and will continue to be played in the
future by pneumatic and hydraulic equipment of automatic control and regulation.
It is not generally known that pneumatic and hydraulic devices have in the past 10
to 15 years kept abreast of electronic devices in their rate of development and
practical applications.
Significant changes have occurred in the last few years in the design of these
devices; the compensation principle for simple and samplex systems of pneumatic and
hydraulic equipment has been applied on a large scale. And a number of problems
have come to the fore in the field of hydraulics. Unless solved, these problems
will impede further progress in the development of this class of equipment.
The following are the main problems facing us in the field of pneumatic and
hydraulic mechanisms:
The investigation of problems of the flow of fluids and gases through narrow
apertures of v!ric.'us lengths 'IA forms; t.e inve-tigation of arouhd various
surfaces, etc.;
The investigation of the dynamics of operation of complex systems of pneumatic
and hydraulic eq.:irment and the developrL:t of methods for increasing their speed
of operation;
The development of retnods of synthesizing ccmplex pneumatic and hydraulic sys-
tems for given conditions and characteristics;
The search for new methods for controlling the flow of fluids in hydraulic
equipment for control and regulation.
12
STAT
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
t?
Here, attention should be drawn to recent efforts to find new working fluids
for hydraulic systems, in particular, fluids with high electric or magnetic conduc-
tivity. This would permit tne application of electromagnetic pumps for regulating
tne flow of liquids; by cnanging the current or the magnetic flux within these
pumps, it is possible to vary the working pressure within the hydraulic system.
These, then, are the principal problems facing us in engineering the basic in-
termediary and final control elements of automatic systems.
Reference has already been made in the first part of tne report to the princi-
pal and basic scientific problems bearing on the engineering systems of automatic
control, guidance and regulation. We shall now turn our attention to still other
imoortant technical problems.
The principal problem facing us here is tne problem of standardizing the con-
struction of automatic and remote control systems. This question ninges on tne de-
velopment of system arrangements which would permit the combination of various in-
pt elements with standardized intermediary and final control elements. Sucn
standardization will require extensive preliminary work in adapting the cnaracteris-
tics and parameters of the various input elements to standardized intermediary and
final control elements, but the effort will be repaid by the great gain stemming
from mass production of tne technical means of automatic control.
The second problem has to do with setting up criteria for evaluating the prop-
erties, characteristics and parameters of elements and complete systems of automatic
control, guidance and regulation. Unfortunately, this highly important problem has
not, so far, attracted the attention it deserves; as a result, we lack the required
criteria for a correct evaluation of the different variants of devices and systems
needed for automatic installations.
We have outlined the m.in prol,lems in the develocmer.t of elerents of automatic
and remote control mecnanIcs; tne solution of tnese proble-in wil: re-.Lire a great,
concerted effort on tne part of matnemaicians, physicists, cnem:sts, and special-
ists in tne tecnmical means of automatic and remote control. It is hoped that suclkia
cooperative and concerted effort will result from the present meeting.
13
Declassified in Part - Sanitized Copy Approved for Release 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Cop Approved for Release ? 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
N. N. Shumilovskiy and B. S. Sinitsin
THE BASIC PROBLE.S OF T1E THEORY OF AUTOMATIC LE,ISURE:.21; TS
Introduction
The rapid development of modern production, the intensification of technological
processes, and tle Df procuction objectives are inconceiva:le without the
increasing application of measuring devices.
Those automated measuring devices referred to as automtic control devices, are
finding particularly wide use in varying degrees. It must ce observed that the term
autocontrol" is often used in a larger sense. In using this word, we shall, however,
confine ourselves to that part of its definition which implies the idea of autocontrol
and automated measurements.
The use of autocontrol devices and even of entire "systems" gave rise to many
important problems which are still far from solved. Thus, for example, great diffi-
culties arise from the excessive increase in size of panelboards, which carry hundreds
of pieces of devices. It is felt that indicating instruments should be replaced wher-
ever possible with recording instruments. Attempts are Leing made to use devices for
automatic control ard computing installations, etc., concurrently LY, 2, 3, 4, 5, 6,
7, tg.
In the present communication, we are examining in particular a group of questions
concerned with the error theory of automated measurements and the methods for obtain-
ing greater accuracy in automated measuring installations.
During the long years of development of accurate laboratory devices for manual
control, sufficiently thorough methods of :-..--esurement were developed and a theory of
errors for such measurements was worked out and solidly grounded.
The characteristic features of accurate measurement methods were in most cases
the assumption of the stability of the value of the measured quantity and of the exter-
nal conditions of the measurements. It has alwa:s been the aim to insure these con-
ditions and to eliminate at times unavoidable variations.
STAT
The industrial indicating devices, representing strictly speaking the first step
14
nariaccifipci in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
I.
in the transition to the modern systems of control, were so crude in the original
period of their application that no special measurement methods were required in
these cases.
However, the developing technology demanded, on one hand, greater accuracy in
the measurements of continuously varying quantities under operating conditions; on
the other hand, laboratory devices for manual control more mad more often did not
satisfy the experimenters, from the physicists mad astronomers to the engineers in
the plant laboratories.
By their very nature these devices were not suitable for the
measurement of varying quantities; at the same time, however, such measurements were
urgently needed. The solution was found in the automation of the measuring processes
and the development of devices for autocontrol.
Their introduction and wide use in production and in the laboratory were launched
not so long ago and are still continuing.
Of course, the old "classical" measurement methods, proceeding in most cases
from the "static" conditions and assuming the presence of a relatively small number
of devices, were found, to say the least, to be inadequate for the new conditions
in the presence of modern measuring installations. In fact, with the conversion to
the automated measuring arrangements, the field of physical phenomena utilized for
measurements grew tremendously, the layout aid the construction of measuring instal-
lations became more complicated, and new sources of static errors arose; finally, the
systems of autocontrol, having automated the measuring process and made possible the
measurement of varia:de quantitites, required the consideration of dynamic errors
which developed as a result of the variation of the variable quantity itself and as
a result of the variation of the external influence, i.e., the interferences.
It must also be pointed out that the simultaneous use of a larger number of
devices, their use in combination with the systems of automatic regulation mad control,
particularly in connection with calculating and telemechanical arrangements, gave rise
to a number of additional requirements for the operation of the autocontrol appar-
atuses and, above all, again required improvement of their accuracy.
15
STAT
'?' ' D,r+ t-Cf ri (nrv Aooroved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
41.
Though somewhat late in regard to the measuring practice, the theory of auto-
matic measurements originated and began to develop as a result of these conditions.
The objectives of tYis theory must obviously be the development of the theory
for independent concrete measurements, performed with the aid of autocontrol systems,
as well as the establishment of the general theoretical foundations for automatic
measurements.
The most important part of the theory of automatic measurements must be the
theory of errors.
Now the question naturally arises as to whether this new trend is an isolated
development or whether there are similar trends in other fields of knowledge. We
accept the latter possibility; this, however, does not exclude the presence of many
original independent questions, as well as the possibility of an indirect influence
being exerted by the theor:? of autonatic measurements on the related disciplines.
As an example for the latter case, we can point to the "classical" theory of
errors and the mathematical treatment of the results of measurements; as an example
of the direct influence on the development of the theory of automatic measurements
we may cite the general theory of communications, the theory of automatic regulation,
and the theory of calculating devices.
The aims and the problems of automatic measurements are closely related to and
are the result of the development of pro,dems previously solved with the aid of
precise laboratory instruments. The novel aspect of all this, the things taking place
under the operating conditions of the autocortrol systems and, above all, the presence
of the continuously varying measure-3 quantities?which create actions and interferences-
-are all very closely related to operating conditions in the field of communications
as well as in the system of the automatic regulation.
Questions of Terminology
The complication of measuring layouts and the increase in their new functions
require a more accurate and complete terminology. We shall not attempt, in this sSTAT
16
Darf qnniti7Pd Coov Aooroved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
paper, to examine this rather large number of terms; we shall only touch upon the
basic definition of measurement and the closely related terms "measured quantity"
and "result of measurement."
The definition given by M. F. Malikov states: "We call measurement the percep-
tual process consisting in the comparison, by means of a physical experiment, of a
given quantity with its certain magnitude taken as the unity of comparison" vg 7,
page 18).
This definition does not stress the admissiaility of continuously varying quan-
tity, which is, as we have indicated, one of the most important operating require-
ments in the automated measuring installations. It is true that, by interpreting
the concept of "quantity," M. F. Malikov on the same page 18 6i7writes: "In the
abstract meaning, the word 'quantity' in general denotes a 'property' to which can
be applied the concept of more or less..."; this in itself apparently permits the
examination of the measurements of variable "quantities." However, further referring
to the result of the measurement (M, page 30), he states that "...the result of
4-hg measurement is always a concrete number" (underlined by M. F. Malikav). This
last statement apparently presupposes the stability of the measured quantity during
the measurement.
In automated measurements of continuously varying quantities, the knowledge of
even an entire series of values of the varying quantity cannot be considered as the
result of the measurement. It is necessary for us to be able, either directly or
through some intermediary variable, to correlate the values of the measured "quantity"
with the values of time.
Under term of the results of measurement of the continuously varying measured
"quantity" we must therefore understand the definition of its values as a simple or
cuiplex function of time. The uasic definition of measurement can be formulated as
follows.
We call measurement the perceptual process, consisting in the comparison (by
means of a physical experiment) of magnitudes of a given quantity (which in general STAT
17
ropy Approved for Release @ 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13 CIA-RDP81-01043R004500020003-3
is a function of time) with its certain magnitude, taken as the unit of comparison.
We can use the previous formulation too, but we will have to define more accur-
ately what we mean by the measured "quantity" and the result of measurement Z107.
General Questions of tne Investigation of Static
Errors in the Systems of Automatic Control
The special features of the automatic control systems--as compared to the man-
ual control devices and, in particular, as compared to the laboratory decade-type
Aic:7 compensators and bridges--are the great complexities of their structural ar-
rangements, including, at times, a considerable number of links and the presence of
a numoer of additional sources of errors, among them, the external interferences.
In examining the static errors, it is expedient and allowable in the majority
of cases to divide them into two groups, depending upon whether or not the magnitudes
of error are functions of the measured quantity /W. For the systems of autocontrol,
comprising a large number of sections with lumped parameters, it is expedient, on
the basis of the method of superposition, to work out expressions for the total error
of the entire system, caused by the variations of the parameters of separate sections..
Such complete expressions permit the evaluation of the error for the entire
measuring system and the determination of the degree of influence of the variations
of the separate parameters upon the magnitude of the error, the dependence of this
error upon the "coordinates" of the point of application of the disturbing interfer-
ence, as well as the degree of the mutual compensation of the separate errors. Be-
sides, they permit the establishment of the relationship between the errors and the
sensitivities, with respect to the reasLred quantity as well as with respect to the
disturbing action. All this, in its final result, considerably facilitates the in-
vestigations of the automatic control systems.
For sone systems, whose structural arrangements differ only as to the number of
sections in this or that branch, it is expedient to make up the expression for the
error by assuming the presence of an arbitrary but finite number of sections in each
branch. This allows tile formulation of sufficiently general rules, characterizing STAT
18
nna-Inecifiarl in Part - Sanitized CODV Approved for Release @ 50-Yr 2014/05/13 CIA-RDP81-01043R004500020003-3
the given group of the system. We obtained such expressions for some of the most
often encountered groups earlier 272,
In connection with the requirements for greater accuracy in autocontrol systems,
the investigation of static error of a system having not only sections with lumped
but also with distributed parameters is quite interesting. On the basis of the ex-
amination of certain particular cases, we have estaolished sufficient conditions
whose presence permits the consideration of the general error of the autocontrol
system as an additive function of a multitude of points of the system.
Let us stress that tha requirements set forth for the static accuracy of the
autoratic control systems are often quite rigid; this means that it is necessary to
consider a large number of influences which could appear to be of secondary signif-
icance. There is no reason to consider the influence of their variations when in-
vestigating the dynamics of the automatic control systems; however, their signifi-
cance must be carefully taken into account in the statics.
Because of this, the need for common examination of statics or dynamics in
various degrees does not mean that the first is only a special case of the second.
Let us note that many questions of the statics of automatic control systems are
still awaiting solution; here we have, for example, the problems of synthesis, which
have not been solved; the examination of the questions of statics does not sufficient-
ly tie in with the problems of dynamics, and so forth.
Dynamic Errors of Systems of Automatic Control
Passing now to the examination of the questions of dynamics, we must mention
that the questions of stability have a lesser significance here than in automatic
regulation 2:147. The problems of quality and particularly of dynamics errors occupy
an important place in the theory of automatic measurements.
Among many characteristic features of the operation of these measuring devices,
the proolem of the accuracy of the measurements of continuously varying quantities
forms the most important group of questions, which differentiate the theory of the
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
19
STAT
narlaccifiinri in Part - Sanitized Com/ Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
I.
of the automated measuring arrangements from the "classical" theory of measurements
by means of devices with manual ccntrol.
As a matter of fact, the automation of measuring installations allowed the con-
version to the measurement of continuously varying quantities, and because of that,
this new, this most important property of automatic devices brought aoout its own,
perhaps its most important and most difficult proolem, i.e., the problem of accuracy.
It may appear at first glance that the investigation of the dynamic systems of
automatic control cannot produce anything new in comparison with the theory of auto-
matic regulation. However, this is not the case. Let us point out, first of all,
that there exists a range of physical phenomena, suoject to control, but not requir-
ing or even not susceptible to automatic regulation. Therefore, questions naturally
arise in the theory of automatic measurements which do not arise in the theory of
automatic regulation.
Second, considerably greater requiremerts are set forth for the static as well
as the dynamic accuracy of the systems of automatic control, than for the systems of
automatic regulation. (As a characteristic example, let us point to the newest auto-
matic bridges with multiple scanning the rnale for temperature measurement in the
assembly with ti:e platinum thermoneter whose static error consists of several thous-
andths of one percent; in this way, the accuracy of these instru.Tents approaches the,
accuracy of the meteorological instruments /15, 167.)
It is natural that these quite rigid requirements for the accuracy of automatic
control require a corresponding theoretical generalization.
Third, let us point to the fc:2o1-1-f- i-portant circumstance. In thu design of
the automatic regulation systems, it is to a great extent necessary to take into
account the properties of tl-e objective of the regulation, which to a consi derable
degree determine the construction of the system.
In the measuring supervision systems (if thu supervision principle is at the
basis of the automatic arrangement), the objective consists in the movable system of
the device--the registering member and the indicator which supervises the variatic? TAT
20
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
STAT
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
1.
?
of the measured quantity. It is hardly necessary to prove that the variations in
the characteristics of these sections of the autocontrol system can be achieved a
lot easier than the variation of the characteristics of the objective in the auto-
matic regulation systems. On the contrary, the inertia of the pickup (primary con-
verter), which in the majority of cases has a small influence upon the dynamics of
the automatic regulation systems, has primary significance in this operation of
automatic control systems. And, finally, let us point out that the autocontrol systems
are in general open systems )147.
Many essential problems of the theory of automatic measurements have already
been solved. Let us mention here the problem of finding the errors of devices with
arbitrary variations of measured quantity Z177; the methods for calculations of
automatic compensators Lib, 19, 207, which made possible the construction
of apparatuses with an indicator scanning the scale in about 0.3 sec; the general
theory of the linear amplifiers '317; the extensive treatment of questions of the dy-
namics of magnetoelectrical devices /527; and, finally, the construction of measuring
devices of scanning converters A-37.
Certain studies on the theory of automatic regulation are of great significance
to automatic control; these studies deal with the operation of automatic regulation
.3ystems which appear to be particular cases of automatic regulation in many systers
LW'
In investigating the dynamics of systems of automatic control, it is possible in
many cases to limit the examination to the questions of quality. This is true, for
example, with respect to multipoint devices when they are used for the measurement
of comparatively slowly varying quantities. However, in an even larger number of
cases?when the problems of the measuring installations consist in more exact measure-
ments of continuously varying quantities--it is necessary to proceed to the investi-
gation of the dynamic errors with random variations of measured quantities.
There are many ways to increase the dynamic accuracy of the automatic measuring
installations.
21
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Sometimes the errors can be reduced considerably only by a radical improvement
in the properties of the elements in the autocontrol systems through the use of the
latest inertialess methods of measurements.
It appears that great possibilities exist for the application of the principle
of the scanning converter in the construction of high-speed instruments.
It is very desirable to design and to manufacture a limited series of such
instruments so that they may be tested thoroughly under laboratory and plant con-
ditions.
However, it is far from possible always to use any kind of inertialess methods
of measurements under certain conditions. In automatic control systems, it is nec-
essary to use measuring elements possessing inertia. In such a case--when elements
possessing inertia are present--a great role must be played by the methods of the
correction of dynamic properties for the entire system. It must be noted that the
correction methods developed by the theory of automatic regulation are very insuffi-
ciently used in the measuring practice.
The basis for the theoretical investigation of the questions of the dynamic
accuracy of automatic systems is, as we know, the mathematical theory of probability,
and, in particular, the theory of random functions. (We shall not dwell in this
work on several other methods for the solution of these problems 2.72-5, Chapter XXI177.)
Under these conditions the measured quantity (master action) is most often considered
as a fixed function of time /2-6, 2/. Such an examination presupposes, first of all,
very small dimensions of the sensing element in practice and the absence of distortion
of the field of the measured quantity in the vicinity of the sensing element and,
second, a given fixed position of the sensirg element in space.
The specifics of the automatic measurements and, above all, the great accuracy
required in automatic measuring installations as compered to the automatic regulation
systems, in certain cases demand a somewhat different statement of the problem.
Thus, it becomes necessary at times to take into account the dimensions of the
Declassified in Part - Sanitized Cop Approved for Release ? 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
22
STAT
Dmri? niti7PCi rODV Approved for Release @ 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
?
sensing elements (for example, in measurements dealing with the investigations of
turbulence jel, 27).
No lesser interest, in the case of autocontrol of the parameters of continuous
processes, is presented by the possibility of displacement of the sensing element.
Let us examine this question in greater detail. The measured "quantity" must be
considered in this case as a function of two independent variables: time and space
coordinates of the sensing element (Figure 1).
Considering the presence of the fund-
amental, predominant direction of propa-
gation of the perturbation in continuous
industrial processes, it is possi.;le to
assert that there exists a correlation
Figure 1. Simplified diagram of the
system of the automatic control with
one sensing element.
between the values of the measured "quantity" at the given moment of time at a cer-
tain point in space, with the coordinate I and the value of the same measured quantity
at the point with the coordinate/4-41 at a certain future moment of time. The cal-
culation of such a relationship permits the supplementation of the usual methods of
"time" correction, used in the theory of automatic regulation, with the additional
"space-time" correction. To realize this, it appears necessary to supply tre system
with several sensing elements. One of them--the primary one--must be located at a
point in the process at Which we want to measure the parameter we are interested in,
while the others--the corrective ones--must be placed ahead of the main sensing
element (i.e., forward toward the propagation of the perturbation of the measured
quantity). It must be noted that such forward location of a single correcting sensing
element is already employed in automatic regulation, not to increase the accuracy of
the measurement, but for the correction of the dynamic properties of the system as a
whole, including the object itself LP, 317. However, the theory of errors of
similar automatic regulation systems (taking into account the forward location of the
corrective sensing element) has not been developed as far as we know.
We have not encountered te application of a system with t?e corrective sensinsTAT
23
Declassified in Part - Sanitized Copy Approved for Release 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
?
elements scattered in space to increase the accuracy of the measurement in the
dynamic regime.
Let us cite some of the results of the experimental investigation of autocontrol
systems with the corrective sensing elements located in space away from the primary
elerent.
The measured quantity is the temperatu-e of the air stream within tne pipe of
an experimental setup.
The temperature was varied by means of changes of current in heaters, located
at the origin of the pipe.
Figure 2. Easic diagram for t:e determination of errors of
tne system of temperature autocontrcl with a correcting thermometer.
Resistance thermometers: TCo - reference (law inertia); TC1 - primary
(inertia); parts of t'ne correcting tnermometer: TC2 - inertia part;
TC3 - high-speed part.
Legend: 1) dynode
The simplified basic diagram is shown in Figure 2. The graph of the temperature
variation in the region of the primary inertia thermometer TC1 and the reference
thermometer TC0 located next to it is s'rown in Figure 3. In producing such a var-
iation of the measured quantity, we strcv ts obtain the characteristics of the var-
iation.-lose to the share of a jurp. The granh of the momentary values of errors
is shown in Figure 4.
As we can see, the forward location of the correcting ther-
mometer insures a substantial decrease in error.
A similar measuring arrangement, but with a different array of sensing elements,
WAS tried with the variation of the measured quantity, which approximated a short
24
STAT
Declassified in Part - Sanitized Copy Approved for Release 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Legend: a) t, sec.
The system with the correcting .sensing devices--even if it is linear oy itself--
cannot be considered strictly speaking as linear in the presence of random correlat-
ing couplings. Final judgment concerning such systems can evidently be made on the
basis of their tests in the presence of random variatiohs of the measured "quantity."
STAT
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
??
&FT). f?co
Impulse. The graph showing the variations
of the measured quantity is shown in
Figure 5 and the curve of the variations
in often momentary values of errors is
shown in Figure 6. We notice that in
this case also the forward location of
Figure 3. Graph of the variation of
the measured temperature the correcting sensing device insures the
Legend: 1) t, sec. decrease in the error. Many experiments
conducted under similar conditions showed that the forward location of the correct-
ing sensing device results in a reduction in the dynamic error of the systems.
Experiments also shamed that the introduction of two additional correcting
sensing assemblies produces additional decrease in error.
11(1,T
Figure 4. Graph showing tha error for the system of
temperature autocontrols
In case of: 1 - single TO1; 2 - single TC2; 3 - all three thermometers,
TC', TC2, Te3, but without the forward location of correcting thermometer;
4 - all three tharmometers and with the forward location of the corrective
thermometer at 9 in; 5 - same as 4, but with a forward location at 12 in.
25
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
dr.
We tested some of the autocontrol systems similar to the arrangement in Figure
2, with random variations of temperature. For this purpose, additional resistances
were introduced into the heater circuit, which produced random variations in the
heater paver around a constant average value, according to a normal law. To obtain
random variations, a special generator was designed and built for the random oscil-
lations, based on the use of the known demonstration apparatus of Galton 3_2.1.
awe
ft
10
2
0
20
JO
JO {sec.
Figure 5. Curve of the impulse variation of the measured temperature.
Figure 6. Error curves of the system of autocontrol with impulse
variation of measured quantity.
In case of: 1 - single TC; 2 - single TC2; 3 - all three thermometers, TC1,
TC2, TC3 but without the forward-located thermometer; 4 - same as 3, but withSTAT
forward distance of 1.5 m.
26
npnlassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release
50-Yr 2014/05/13
CIA-RDP81-01043R004500020003-3
The results of the investigation of the system with random variation of the
measured quantity also confirmed that the forward location of the correcting sensing
device insures a decrease in dynamic errors.
The variation of the stream velocity in general has an influence on the quality
of the correction. At the present time work is xring conducted on building a system
involving the introduction of the correction for the speed of propagation.
Figure 7. Three-dimensional graph of the correlating function of the
homogeneous random field.
To investigate theoretically the questions concerning the accuracy of the auto-
control system with forward-locatea correcting sensing devices in the presence of
random variation of the measured "quantity," it is necessary to know its average
value M(P) and its correlating function M(P) C (PI).
The correlating function can be represented graphically as a surface
R 6-0 = R(c,A),
With this information and the knowledge of the characteristics of the system,
it is possible to find the values of the root-mean-square of the error. The solution
of the problems of systhesis is also possible.
In conclusion, let us say that the methods for the decrease of dynamic errors
described above do not exclude each other by any means but rather complement each
other and are often used simultaneously.
The improvement in the dynamic characteristics of systems often finds itself
in contradiction to the requirements set forth for their accuracy in the static methoesTAT
of operation.
? A -,,,migari for Release
50-Yr 2014/05/13 CIA-RDP81-01043R004500020003-3
STAT
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
This circumstance must without doubt be taken into account in the design of the
automated measuring installations.
Some Modern Development Trends in the Field of
Automatic Measuring Apparatus
The diversity of measurement methods and of problems involved in automatic
control gave rise to many arrangements and constructions of measuring apparatus. In
addition to the dial-type apparatuses with increased rotation moment, autometic com-
pensators, and automatic bridges, we can begin to nate the wide use of the coordinate
type of registers (with polar or right-angle system of coordinates), of the previously
mentioned instruments with scanning converters, and of various versions of digital
devices in autocontrol systems built on the aggregate principle, etc.
Some of these products are not yet being manufactured by the domestic industry
or they are being manufactured in such small quantities that it is difficult to
arrive at any conclusion about their operating qualities.
Let us consider in this connection some of the development trends in automatic
measuring installations and in problems arising in this connection in the theory of
the automatic measurements.
The expansion of the field of measured quantities, the ever increasing use of
methods of electric measurements for nonelectric quantities, and the expansion of
the limits of ryaasurements require the development of new types of sensing devices
and all kinds of improvements in the existing ones. In many cases, good results were
obtained in t're design of sensing devices using semiconductors, radioactive isotopes,
etc. The problem of the improvement of sensing devices is one of the most important
in the field of automatic control.
Of great importance also is the improvement in the subassemblies of the devices
used in automatic control, since many of them have weak spots. Thus, for example,
the slide mires of the automatic compensators and the bridges introduce errors be-
cause of their individuality, the nonuniformity of the resistance winding along its
length, sliding wear, etc.
28
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
STAT
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
The sources of working current of the autocampensators of direct current vary
? ? their emf with time and with the ambient temperature. The sources for feeding the
compensators of alternating current also are lacking in stability. The standard
elements require particularly careful handling; due to their imperfection, the errors
of the apparatuses are also increased.
We do not have reliable information about the ch-xacteristics of the automatic
comnensators without sliding wires and of standard and dry cells manufactured by
some foreign firms, while the development, conducted by local design and scientific
organizations, is progressing very slowly in this direction.
Much can be done to improve assemblies not only in the field of automatic com-
nanqAtors and bridges but also in any other region of automatic control.
One must particularly call attention to the need for the improvement of the
process of recording measurement results, especially, better paper for photorecording
in daylight.
Let us now examine SOMB general tendencies in the development of matomatic meas-
uring apparatus and above all in the problem of digital devices.
Digital devices have for a long time been used along with continuous-operation
instruments in measuring practice. Decade-type laboratory bridges and compensators
are examples of this, though they are completely nonautomated.
The following factors have been responsible for this in recent years: first,
in many cases the measurement itself has a different, separate character and con-
sequently the digital devices, due to their very nature, are better suited for making
the measurements, than are the devices with continuous operation; second, the digital
devices allow the direct feeding of the data into the computers and thus become one
of the important elements of the next, more developed stage of automation; third,
digital devices are more convenient for telecommunication of t're measured results and,
finally, these devices are devoid of certain errors that are inherent in continuous-
operation devices (reading errors caused by the parallax, the nonuniformity of the
scale, and so on).
29
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13 CIA-RDP81-01043R004500020003-3
The rapid deve:opment of digital devices wrs made possible after the intro-
duction of electronics into measuring practice; hov.ever, even now the solutions for
the construction and arrangement, which are at the basis of these devices, are quite
diverse. There exist electromechanical and purely electronic variations.
?
We consider thst one of tne most important pro.le s of the near future is the
development, mastery, End manufacture, by the local instrument plants, of digital
indicating and recording devices, while the scientific research organizations must
develop the thecr:- of digital devices, taking into account their application in the
measurements of varying quantities.
It must be noted that, because of certain drawoacks of the digital devices and
above all because of the nongraphic aspect of digital recording, there is no basis
to think that thcse devices will completely displace the devices with continuous
action.
In our opinion, there is a great future for the multiple scanning of the scale,
which represents an original combination of digital and continuous-action devices
/n, 16, 337. Combining the high accuracy of the digital devices with the graphic
recording obtained from the devices with continuous registration, these devices are
destined to replace to a great extent the accurate laboratory measuring arrangement
using manual setting during experimental investigations. The use of the multiple
scanning of the scale also makes it possible to somewhat increase, by quite simple
means, the accuracy of the existing instruments manufactured by the local plants. It
appears desiraole to us to manufacture in the nearest future a moderate number of
instruments with ma.ltiple scanning of the scale.
Measuring practice is in many cases grid zly facilitated by the so-called coordin-
ate-type recorders which make it possible to recorc one quantity as a function of
another. However, too few of these devices are ein produced and they are not avail-
able in sufficiently different models for use in the laJoratory and in the plant; it
is therefore practically impossible to use them extensively. Their theory is also
very insufficiently developed.
?.?
30
STAT
('rirvI Annroved tor
@ 50-Yr 2014/05/13? CIA-RDP81-01043R004500020003-3
STAT
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
.?
Recently a demand arose for the construction of automatic compensators and
bridges for alternating current. In balancing such measuring devices, two inter-
dependent parameters must be made to vary automatically. The theory of such devices
is not yet sufficiently developed.
Another important problem consists in the creation of automatic measuring de-
vices for indirect measurements, i.e., the building of groups of devices and computer-
installations which would automatically measure several pc..rameters and would auto-
matically compute the measurement results on the basis of certain given direct measure-
ments. Not so long ago, only isolated examples existed of such complex measuring in-
stallations. But, the rapid development of computer installations has stimulated
considerable progress in this field. Let us note that, in addition to their use for
the mathematical treatment of the measurement results, the computer installations
can be used for the design of tFe measuring installations, which automatically vary
their awn characteristics, when variations occur affecting any of the conditions of
measurements. Such measuring installations can be considered as selfadjusting mea-
suring systems. The important questions of the theory of such devices are also in-
sufficiently developed.
Finally, let us examine the question of checking the automated measuring in-
stallations. We will hardly be wrong in stating that little has been done in this
field. In fact, the existing system for the checking of these devices provides
mainly for the testing of these devices under static conditions. Proper attention is
not being devoted to the problem of determining the dynamic errors; the methods for
their determination are quite imperfect, while the checking facilities?reference
standards and instruments?are often lacking.
Much work remains to be done in the creation of facilities for the checking of
automated measuring installations under the conditions of measurements of variable
quantities and for the development of checking methods.
Particularly great difficulties are encountered, of course, in working out
the standard measurement methods.
31
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
a
???
???
In conclusion let us state that the successful solution of the enumerated
problems, as well as of those we could not mention in this short report, is only
possible under the proper combination of conditions of the development of new techni-
cal means for automatic control and of the solution of problems of the theory of
automated measuring installations.
Many special features and, above all, the consideration of the continuous vari-
ations of measured quantities make the theory of automatic
from the error theory and from the calculation methods for
measurements different
accurate laboratory devices.
The specifics of the automated measuring installations and, in particular, their high
accuracy?which at times, under static conditions, approaches meteorological accur-
acy--excludes the possibility of considering this new field of knowledge as the
simple application of tin theory of autoratic regulation to the concrete problems of
measuring practice.
BIBLIOGRAPHY
1. Trapeznikov, V. A., "Basic problems of
processes," Avtom. i telemekh. Autom.
2. Berezovets, G. T., "New developments in
i telemokh., 17, No. 1, 1956, pg. 78.
the complex automation of industrial
and Telemech.7% 15, No. 5, 1954, pg. 384.
industrial pneumatic automation," Avtom.
3. Simkin, M. M., "Application of elements of digital computing technique to
systems of automatic controlling a-d regulation," Avtom. i telemekh., 17,
1956, pg. 180.
the
No. 2,
4. Trapeznikov, V. A., "Basic problems in the development of technical automatic
facilities for industrial use," Priborostroyeniye Linstrument?buildinil, No. 1,
1956. .
5. Rakovskiy, M. Ye., "Basic principles of the construction of systems for the auto-
mation of continuous technological processes," Priborostroyeniye, No. 6, 1956,
pg. 1.
6. Keinat, G. Veilfach-messung und Registrierunc zur Betriebsiiberwachung. Archly
fir Technische Lessungen. J032-4, L.-200, September 1952, s. 199.
7. Leo, ii. The value of instrument chart records. Gas World, 139, N 3637, 1954,
pg. 1205.
8. Electro-Instruments--An a. c. digital voltmeter. Control Engineering 2, N 10,
1955, pg. 94.
32
STAT
if ori ('rirw Aroroved for Release @ 50-Yr 2014/05/13 CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Cop Approved for Release ? 50-Yr 2014/05/13 CIA-RDP81-01043R004500020003-3
????
9. Malikov, M. F., Osnovy metrologii grinciples of metrologi7, Publishing House of
the Committee on Metrology and Measuring Devices attached to the Council of
Ministers USSR, Moscow, 1949.
10. Sinitsin, B. S., The problem of terminology of automatic control, Dokl. Livorskogs
politekhn. in-ta. ATeports of the Lvov Polytechnic Institut?, Vol. 1, No. 2,
1955, pg. 142. ?
11. Kirpatovskiy, S. I., Nekotoriye voprosy teorii elektricheskikh dutchikov jgome
theoretical questions on electric sensing devicesj, candidate thesis, Lvov, 1951.
12. Sinitsyn, B. S., "Dependence of errors of elpectromeasuring devices upon their
sensit.ivity. Problems of automated and measuring practice," Nauch. zap. Inst.
mashinov. i avtomat. AN USSR A7cientific Notes of the Institute of Machinery and
Automation, Academy of Selena-6's Ukrainian SSR7, Vol. II, No. 2, 1953, pg. 41.
13. Sinitsyn, B. S., "Analysis of static errors in autocontrol systems," Nauchn.
zap. Ltvcvskogo pclitekh. in-ta, scientific notes of the Lvov Polytechnic Insti-
tute7, Electric instrument-building series, No. 1, 1956.
14. Shumilovskiy, N. N., Osnovy metodov avtomaticheskikh izmereniy 2Trincip1es of
automatic measurement methods7, report delivered at the Conference on Electric
Measurements and Instrument-building (in print), Lvov, 1956.
15. Stull, D. R., "An automatic recorder for resistance thermometry," Review of the
Scientific Instruments, 16, 1945, pg. 318.
16. Williams, A. J., "Electric recorder with range and precision adequate for the
platinum resistance," Trans. AIEE, 71, 1952, pg. 289.
17. Baranchuk, Ye. I., "Dynamic errors of devices designed for the measurement of
rapidly changing quantities," Avtom. i. telemekh., 11, No. 3, 1950 pg. 170.
18. Ivakhnenko, A. G., "Automatic measuring compensators, investigated from the point
of view of quality and stability," S. Inst. elektrotekhn. AN USSR jEollection
of the Electrotechnical Institute, Academy of Sciences Ukrainian SSR7 No. 5, 1950;
No. 7, 1951; Nos. 8, 9, 1952.
19. Lerner, A. Ya., Ulusheniye dinamicheskikh svoystv avtomaticheskikh kompensatorov
pri pomoshchi nelineynykh svyazey ffmprovement of dynamics properties of automatic
compensators by means of nonlinear links7, thesis for the degree of doctor tech-
nical sciences, Moscow, 1952.
20. Minina, O. M., Privod avtomaticheskikh elektronnykh potentsiometrov /rive for
automatic electronic potentiometers7, candidate thesis, IAT, Academy of
Sciences USSR, 1954.
21. Strelkov, S. P., "General theory of linear amplifiers," Part I, Avtom. i telemekh.,
9, No. 3, 1948, pg. 233; Part II, ibid., 10, No. 4, 1949, pg. 274.
22. Kharchenko, R. R., "Correlation of dynamic characteristics of electric measuring
devices and converters," Priborostroyeniye, No. 2, pg. 21, 1956.
23. Temnikov, F. Ye., Avtamaticheskiye registriruyushchiye pribory jutomatic record-
ing devices7, Mashgiz, Moscow, 1951.
33
STAT
in Dmr+ rliti7Pd Cony Approved for Release @ 50-Yr 2014/05/13 CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
24. Felidbaum, A. A., Metody issledovaniya perekhodnykh protsessov v samonstanavliva-
yushchikhsya sistemakh Zinvestigation methods for transitional processes in
selfadjusting systemg, doctoral thesis, Moscow, 1948.
25. Solodovnikov, V. V. (editor), Osnovy avtomaticheskogo regulirovaniya foundations
of automatic regulation7, Mashgiz, Moscow, 1954.
26. Solodovnikov, V. V., Vvedeniye v statisticheskuyn dinamiki sistem avtomaticheskogo
upravleniya /Introduction to statistical dynamics of systems of automatic controg
GTTI, Moscow-Leningrad, 1952.
27. Yaglom, A. M., "Introduction to the theory of stationary random functions," Usp.
mat. nauk /Advances in mathematical sciences7, Vol. VII, No. 5, 1952, pg. 1.
28. Yaglom, A. M., "On the calculation of inertia of meteorological devices when taking
measurements in turbulent atmosphere," Tr. Geofiz. in-ta. AN SSSR /Works of
Geophysical Institute, Academy of Sciences uss17, No. 24, (151), pg. 112.
29. Pearson, C. E., "Corrections in hot-wire correlation measurements," Quart. Ippl.
Math., 12, No. 3, 1954, pg. 235.
30. Gerasimov, S. G. (editor) Avtomaticheskoye regulirovaniye kotel'nykh ustanovok
Automatic regulation of boiler installationg, Gosenergoizdat, Moscow-Leningrad,
1950.
31. Gordov, A. N. (editor), Metody izmereniya temperatur v promyshlennosti 6Tethods
for temperature measurement in industrY7, Metallurgizdat, Moscow, 1952.
32. Khalid, A., Matematicheskaya statistika s tekhnicheskimi prilozheniyami 6Tath-
ematical statistics with technical applicationq, IL, Moscow, 1956.
33. Sinitsin, B. S., "Automatic compensators and bridges with multiple scanning of
scale," Dokl. Lvovskogo politekhn. in-ta., Vol. 1, No. 2, 1955, pg. 147.
34
STAT
npnlassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
STAT
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
?
B. T. Kolomiyets
SEMICONDUCTORS AND THEIR APPLICATION IN AUTOMATION
As is known, semiconductors are distinguished by the ability to change their
electric conductivity under the influence of many external factors. These include
light, temperature, magnitude of electric and magnetic fields, direction of current,
pressure, etc. Previous experiments in the field of electric properties of semi-
conductors made it possible to find materials in which one of the enumerated factors
assumes the major role in controlling the conductivity and to determine ways for
control by means of some other properties in many semiconductor materials. Both of
these circumstances resulted in the creation of new electric circuit and cells, with
the parameters and the characteristics differing by the second or third order from
the previously known ones. By way of example, we can mention such general character-
istics as sensitivity, power dissipation, length of service, weight, volume, etc.;
each of these characteristics differs from the characteristics of previously existing
devices by 100 to 1,000 times and more.
The abovementioned features of the semiconductor devices and many other properties
are interesting for industry in general and for automation and telemechanics in par-
ticular. On their basis we can make simply-constructed and reliably operating equip-
ment for the control, signaling, end regulation of technological processes in many
branches of industry.
The prospects for the application of semiconductor equipment in automation and
telemechanics are very great. Such equipment can be used and is being used as sensing
elements and as intermediate elements in various automatic and telemechanical in-
stallations. In all known cases, the semiconductor products proved themselves to be
reliable elements, completely satisfying the strict requirements of industrial oper-
ations. However, the many existing semiconductor devices are not yet oeing used to
a great extent in automation practice.
The present paper is intended as a review of the present state of the art in
the creation and utilization of semiconductor equipment in the Soviet Union and aoroad
35
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
? and as an examination of the prospects of their development. Below are described
.
? groups of semiconductor devices whose application considerably widens the potential-
ities of automation and telemechanics.
Photoelectric Devices
The extensive research on photoelectric porperties of semiconductors in the
Soviet Union resulted in the creation of many original photocells. Here we find, first
of all, silver sulfide photocells produced by the Pksics Institute of the Academy of
Sciences Ukrainian SSR. The present photocells, which were developed as far back as
1938, possess a sensitivity as high as 10,000 ,rza per lumen as compared with 500/46a
for the selenium photocells. It must oe stressed that these photocells are being
used extensively in Soviet industxy but are not known abroad.
lie find even greater sensitivity in (also completely original) photoresistors
made from powdered cadmium sulfide developed by LFTI of the Academy of Sciences USSR
and NII MRTP. These photoresistors, manufactured industrially since 1954, have a
sensitivity reaching 3,000,000Aa per lumen.
The appearance of semiconductor photoresistors led to a new stage in the develop-
ment of industrial photoelectric automation: the creation of simple, reliable, and
long-lasting equipment. The example of the use of photoresistors in automation can
be considered as characteristic for all semiconductor products. Because of this, we
shall examine them somewhat in greater detail. Let us also note that in the field
of photoresistars our achievements exceed thosaabroad.
Among many advantages brought to automation by semiconductor photoresistors, as
compared with the photocells when activated by en external photoelectric effect,
we can mention the following.
1. High sensitivity. As we said before, the sensitivity of the photoresistors
attains 3 amp per lumen, that is, the sensitivity of the photoresistors is higher
than the sensitivity of ordinary antimony-cesium vacuum photocells by 30,000 times.
2. High allowable power dissipation. Under illumination, photoresistors can
SI-AT
control electric circuits requiring several watts of power; this cannot be achieved
36
in Part Sanitized COON/ Approved for Release @ 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
with photoelectric devices built on the basis of the external photoeffect.
3. Small dimensions and weight. The photoresistors do not require for their
operation a vacuum or gas medium; this makes it possible for them to have very small
dimensions. For use in the laboratory, we have photoresistors with a light-sensitive
area from 1 to 400 mm2 in the shape of squares, rectangles, disks, sectors and washers.
The possibility of giving the light receivers any shape considerably expands the
range of their application in automation.
The operating stability of photoresistors is very high. There are industrial
installations where the photoresistors have operated without interruption and replace- c
ment for more than two years.
Among the diortcomings of the photoresistors must be mentioned their sluggish-
ness, which limits their application to problems where their operating time is of
the same order as that of the electromagnetic relays. Table 1 shows the time constants
for all of the photoresistors manufactured by industry. It shows that the lead sul-
fide photoresistor of the type FS-Al has a very small time constant.
Table 1
Type of Photoresistor
Time Constant, sec.
FS-Al
FS-B2
FS-K1
FS-K2
4?10-5
1.10-3
25'10-3
30.10-3
For applications where a rapid response from the light is required, it is nec-
essary to use the new type of semiconductor photocells: germanium photodiode de-
veloped by LFTI of the Academy of Sciences USSR. This type of photocell combines to
a remarkable degree the best properties of vacuum photocells and photoresistors. From
the former it borrows the rapid response and the proportionality between the photo-
current and the light beam and from the latter the high sensitivity,
Table 2 compares the main characteristics of photocells, photoresistors, and
photodi odes.
37
STAT
neclassified in Part - Sanitized Copy Approved for Release 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
STAT
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
????
The construction of the photoresistors manufactured by industry and their
dimensions are shown in Figure 1, those of the photodiodes developed by LFTI are
shown in Figure 2. These figures show that the dimensions of bath types of products
are small but this is their great advantage in the construction of automated instal-
lations.
b)cte-,o,
c) oc-
e)9V-KZ
IfT0471
( )
? '
Figure 1. Construction of main types of resistors
(dimensions in millimeters).
Legend: a) FS-Al
b) FS-A4
c) FS-B2
d) FS-Kl
e) FS-K2
The semiconductor photocells--that is, rectifier photocells, photodiodes, and
photoresistors taken as a whole--embrace a very large span of radiations, from gamma
rays to the medium infrared wave lengths. This makes it possible to design control
and automatic instrumentation in those regions of radiation which increasingly enter
into engineering practice. Figure 3 shows the spectrum sensitivity of germanium
photodiodes and photoresistors manufactured by industry. It must be also mentioned
that the photoresistors made from the multicrystals of cadmium sulfide, manufactured
by the Physics Institute of the Academy of Sciences Ukrainian SSR, have sensitivities
in the ultraviolet region of the spectrum and in the x-ray region.
38
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
Table 2
Types of Photocells
Type
Designation
Sensitivity
icta per
lumen
Maximum
Photocurrent
Vacuum antimony-cesium
STsV-3
100
0.1
Gas-filled cesium
TsG-4
500
0.1
Germanium photodiode
FD-1
30,000
3.0
Lead-sulfide photoresistor
FS-Al
10,000
0.2
Cadmium sulfide photoresistor
FS-Kl
3,000,000
50
Photoelectronic multipliers
FEU-19
10,000,000
. 0.1
?
Figure 2. Outline dimensions of LFTI germanium photodiodes.
Legend: a) FD-2
b) FD-1
The existence of an extensive variety of photoresistors led to the production
of many types of controlling and automatic instrumentation of industrial value. There
are already a considerable number of examples illustrating that the use of various
installations incorporating the new techniques employing semiconductor photoresistors
saves hundreds of thousands of rubles in industry per year and frees manual labor by
replacing the labor of man with the work of automatic machines. These examples have
to do with the metallurgical industry (Uralmetallurgavtomatika), printing industry
(Moscow, Leningrad), food industry (Karpov plant and K. Tsetkin tobacco factory in
Leningrad), and others. However, the acove-mentioned facts are only isolated appli-
cations and are not widespread. The reason for this, in our opinion, is the insuf-
ficient level of production of photoresistors, insufficient information, and, above
SI-AT
all, the absence of a specialized production center for the development and manufactu.-
39
Onr+ Ca .t. ri ('cnv A ID o ed for Release @ 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
STAT
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
?
ing of photoelectric automatic equipment to satisfy the numerous requirements of all
branches of industry.
V
J?J
1
gc
b) 4numa gmmu, MX
Figure 3. Spectrum sensitivity:
1 - of germanium diode; of photoresistor: 2 - FS-Al, 3 - FS-B2,
4 - FS-K1, 5 - FS-K2.
Legend: a) photocurrent
b) wavelength, ft
One of the factors hindering the wide use of photoresistors in industry is the small
amount of work in tl-e creation of new types and designs. The Ministry of the Radio-
technical Industry has only one small laooratory working along these lines and is
therefore not capable of satisfying all requirements. There is a series of important
proolems whose solution is impossible because of the small size of the laboratory at
NII of the Ministry of the Radiotechnical Industry.
The main problem in the field of photoelectric automation is the creation of a
strong developmental and production foundation. It is necessary to organize the
production of automatic equipment which usoz. semiconductor photocells, as well as
considerably to expand tke production of photoresistors and increase the amount of
scientific and developmental work in the investigation of naw types and designs. These
measures must e undertaken as soon as possible, not only because it is necessary to
supply our industry with the new technique, .out also cecause the investigations now
being conducted on the photoelectric properties of the semiconductors promise to give
40
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part- Sanitized Copy Approved for Release ? 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
t4
industry new and more valuable types of photocells. The new properties of photo-
cells naturally will lead to the creation of new instruments which then must be intro-
duced into industry on a large scale. In order to shorten this process, a physical
base is required in the form of strong branch laboratories and industrial production.
In the nearest future, our industry must produce a new kind of photoresistor
from cadmium selenide. According to the preliminary results, these photoresistors,
whose spectrum sensitivity is shown in Figure 4, have a sensitivity 10 times larger
than the photoresistors of cadmium sulfide, with considera.:ly less sluggishness.
These properties, combined with their spectrum sensitivity will undoubtedly make it
possible to develop new devices and new automatic equipment for industry.
wv
10
CAU
..f
a
ft
\\.......,...........
I
W
4' 4' 47 4/ 4!
rb ),,Zsowt atmokii bozo', mi?
Figure 4. Frequency characteristic of photoresistors of cadmium
selenide and cadmium sulfide.
Legend: a) photocurrent, relative units, %
b) wavelength of light, /4W
In addition to the photoresistors of cadmium selenide, one must expect in about
two years the appearance of lead selenide photoresistors. As we can see from Figure
5, these photoresistors are particularly interesting because of their high sensitivity
in the infrared region of the. spectrum which extends to 6 /46.
As we said before, the germanium photodiodes are a very interesting kind of
semiconductor photocell. It is expected that they will soon be produced industrially.
However, it is necessary to call popular attention to this fact because, from 1954
to the present time, the organization of such producti= has not been successful. SI-AT
41
Darf qnniti7PCI Cony Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13 CIA-RDP81-01043R004500020003-3
It is necessary to ask the Ministry of the Radiotechnical Industry to organize the
production of germanium diodes in order to guarantee progress in many technical trends.
including automation.
The group of photodiodes is a very promising kind of semiconductor photocell.
Their spectrum sensitivity extends from the gamma rays to the far infrared region of
the spectrum (10-12iU). There are already indications in the literature on the sub-
ject to the effect that it is possible to make photodiodes from a series of new semi-
b) ,fnuma 11. lenge/ Alk
Figure 5. Spectrum distribution of the internal photoeffect in PbSe.
Legend: a) relative sensitivity, %;
b) wavelength, IC4'
conductor materials; some of their first new properties are described there too. Be-
cause of the low sluggishness and high sensitivity, this kind of photocell merits
serious consideration for use in automation. To push the development in this direction
it is necessary to create the required conditions, first of all, by securing the
requisite pure chemical products.
To satisfy some of the requirements of automation, considerable interest is pre-
sented oy the rectifier semiconductor photc-ells of the silver sulfide type, mentioned
aoove, and of the selenium type. Both of these photocells found in our country are
used extensively. However, foreign industry is apparently making wide use of recti-
fier photocells, made of single-crystal germanium, silicon, or cadmium sulfide, which
are called solar batteries. All of these photocells have a very high sensitivity
and their coefficients of conversion of light energy into electricity, according tosTAT
42
in Darl? qnniti7Pri rODV Approved for Release @ 50-Yr 2014/05/13 CIA-RDP81-01043R004500020003-3
the published data, is 10% and more.
In the manufacturing and design of photoresistors, our industry is ahead of
foreign industry, but in the field of rectifier photocells just mentioned, we are
apparently considerably behind. Information about them began to appear in foreign
liturature two years ago. In our country, however, we did not uegin to receive the
first specimens with very modest pf.rameters until this year.
The rectifier photocells are bound to be of consideraJle interest in automation
because they can be used as sensing elements of the automatic instrumentation and
as sources of energy. The volume and level of theoretical experimentation in the
field of photoelectric phenomena in the Soviet Union does not lag in its scope
behind efforts abroad. It is only necessary to consolidate the material-technical
bases.
Thermal Resistors
The semiconductors have a negative temperature coefficient of ohmic resistance.
The conductivity of the semiconductors varies greatly with the change in temperature,
according to the exponential law d= Ae----T - which applies to a very large number
k
of semiconductors. The type of variations for metals and semiconductors is shown
in Figure 6.
a
d) Temflepamoll, T
Figure 6. Type of relationship between resistance and temperature.
Legend: a) resistance, ohms
b) metals
c) semiconductors
d) temperature, ?C
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
43
STAT
npnlassified in Part - Sanitized Copy Approved for Release
50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
?
S.
The great dependence of conductivity of semiconductors upon temperature has
been used and at the present time, on the basis of this, industry developed and is
now manufacturing various types of semiconductor-type thermal resistors (thermistors).
The semiconduator thermal resistors in our country are used extensively in the
solution of such important problems as the automation and control of industrial pro-
cesses in the most varied branches of industry. V.ith the aid of the thermal resistors
we can measure control, and regulate temperature, stabilize the potential, temperature
compensation, fire alarm, and other types of signals, and measure humidity, pressure
velocity of gas streams, and many others. The majority of these problems are success-
fully solved by means of thermal resistors manufactured by the plants. The thermal
resistors of NII of the Ministry of the Radiotechnical Industry, developed together
with LFTI of the Academy of Sciences USSR, are shown in Figures 7 and 8.
The interest in semiconductor thermal
resistors is caused by the general features
of the semiconductor devices: their small
weight and volume, stability in operation,
and excellent electric properties.. As an
example showing the advantages of the semi-
conductor resistance thermometer over the
metallic thermometer, we can cite the
following facts. The thermal resistor
of the type KMT-5 used as resistance
thermometer has a volume hundreds of times
smaller than that of the metallic re-
sistance thermometers, and for the same
te-aeratLre difference produces a change
of resistance many thousands of times
larger.
1)A640
p*M7-1
2) MI47-,
tiAN1-#
Figure 7. Construction details of
thermal resistors PET-1, KMT-1, MMT-4,
and MIT-4.
a - oody of thermal resistor; b - contact
caps; c - layer of enamel paint; d - pro-
tective metallic case; e - glass in-
sulator; f - metallic foil; g - layer of
tin.
Legend: 1) LT-1 and 1MT-1
2) MMT-4 and KMT-4
It can be asserted that in the field of semiconductor thermal resistors and
44
STAT
narlaccifiPri in Part - Sanitized Com Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13 CIA-RDP81-01043R004500020003-3
??
their application, the development work in the Soviet Union is an the same level
with the work abroad. However, the general use of these resistors cannot oe con-
sidered satisfactory because not even the current requirements of industry are being
met due to the low volume of production.
The appearance of thermal resistors led to the conception in industry of a
series of interesting ideas, which require the development of new designs of thermal
resistors with new electric parameters. Among them can be cited resistors for auto-
matic starting of electric motors without rheostats, resistors for control and reg-
ulation of cooling of automobile engines, charging of batteries, and so on. These
problems and other similar ones, in spite of their great significance to the national
economy, cannot be solved due to the low production volume of devices for experi-
mental-design and development purposes.
Figure 8. Thermal resistors for temperature compensation.
A drastic increase is required in the volume of work on semiconductor thermal
resistors and the necessary materials must be made available for this effort.
The investigations dealing with the temperature effects of the semiconductors
are proceeding in two directions. One has for its aim the creation of new designs
for thermal resistors. During the current year, our industry will produce thermal
resistors of the type KIAT-l0 (Figure 9), intended for the thermal control equipment
developed by the Electrotechnical Institute of the Academy of Sciences Ukrainian SSR.
These thermal resistors, together with the electromagnetic relays,mske possible the STAT
45
nna-Inecifiarl in Part - Sanitized CODV Approved for Release @ 50-Yr 2014/05/13 CIA-RDP81-01043R004500020003-3
STAT
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
?4
construction of automatic equipment which signals the increase above the preset
temperature within the limits of +20 to +100?C. The accuracy of operation at any
temperature within the above interval is within one degree.
Next year, our industry will receive another one of the new types of the semi-
conductor devices which are of considerable interest for automation; they are called
bolometers. Semiconductor bolometers represent one of the variations of the thermal
resistors, made in the shape of a thin film of semiconductor which changes its ohmic
resistance as a result of absorption of heat radiations.
Figure 9. Thermal resistors of type KMT-10 for heat control.
The second investigation trend is aimed at the discoveryof new semiconductor
?
materials, combining high temperature dependence with high conductivity and stability
when operating at high temperatures. In the conduct of these experiments, a series
of interesting results were obtained which ensure further progress in the development
of this group of semiconductor devices.
Amplifiers and Rectifiers
The investigations of the phenomena at the boundary between metals and semi-
conductors led to the aevelopment of rectifiers from alternating to direct current.
Similar investigations with semiconductors of different carrier polarity and with
high mobility very recently were crowned with remarkable success in our century,
namely, the development of semiconductor amplifiers. The absence of a vacuum, the
small dimensions, and other qualities of these semiconductor devices have brought
about a revolution in the radio industry and those branches of industry which use it.
Automation is also included in this, since it utilizes radio devices extensively.
46
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
STAT
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
We are not going to examine the details of construction and the characteristics
of germanium and silicon diodes, triodes, and other devices, since they are well
known. Also, their value to automation is well understood, due to the long life, re-
liability of operation, small volume, and weight, etc. In examining the uses of
these devices in automation, we must recognize first of all the *small scale of
practical use, which is due to the general unsatisfactory conditions within the group
of semiconductor devices under examination.
In earlier semiconductor devices, our thermal and photoresistors were on the
same level with foreign technology and even surpassed it; in the field of germanium
and, particularly, silicon triodes and diodes, we are considerably behind.
_ .is lag exists in the field of the creation of industrial models and the engin-
eering of their industrial production, as well as in the volume of production and
in the application of types already developed. In our opinion, the first situation
is linked to the insufficient amount of activity in all stages of experimental work,
from tke Academy of Sciences to the plant laboratories. The second situation is
due to the lack of industry's serious consideration of the production of germanium
1
diodes and triodes. Only recently, after three years of the design work on the first
models, WAS the production of germanium devices entrusted to a competent plant.
The absence of an organized and sufficient volume of production of semi-
conductor devices made of germanium and silicon impedes the progress of industry
at the present time and threatens the development of industry in the future. At the
present time, interesting research work is being done on new semiconductors with
large drift mobility. This research, in which Soviet science is not lagging behind
foreign science, will undoubtedly soon result in the discovery of new, more perfect
semiconductor devices. In industry, in general, and in automation, in particular,
the lack of production or a small volume of production will naturally hinder the
rapid transmission, to industry, of the expected results of scientific research.
A great event in the field of semiconductor rectifiers is the development of
powerful germanium rectifiers by the group of scientific workers of LFTI of the
47
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
? V/
aft
Academy of Sciences USSR. The new contribution brought to industry by these recti-
fiers nay be judged by the data shown in Table 3.
Table 3
Type of Rectifier
Selenium
Germanium
Allowable
current density
ma/cm2
50
100,000
Value of Inverse
voltage, v
26
200
The table shows that the current density in the new rectifiers is 2,000 times
larger than that of the best existing selenium rectifiers. This circumstance and
the large value of inverse voltage make it possible to design rectifiers which are
hundreds and thousands of tines smaller in volume but have the same value of rec-
tified current.
When large-capacity germanium rectifiers came into existence, we were not lag-
ging behind foreign technology. However, judging by the degree of their use in
practice, the gap widened with time to our disadvantage. The reason for this lies
again in tle absence of organized production of semiconductor equipment.
The outlook in the field of semiconductor rectifiers is excellent. Thus,
technology will in the nearest future receive large-capacity silicon rectifiers.
According to the available information, industrial production of silicon rec-
tifiers has been started abroad, with parameters two times larger than those of
germanium rectifiers.
The investigations of electric properties of new semiconductors with large
drift mobility must also open the way for the development of new rectifiers that will
be even more perfect with respect to sone of their parameters.
It is appropriate to say that, concurrently with the work on the new types of
rectifier, intensive research with considerable results is being conducted in our
country to improve the properties and the preparation techniques of the existing
copper oxide and selenium rectifiers. In view of its economic importance, this work
is still overshadowing the work being done on the newest rectifiers. This, however,
STAT
48
narlaccifiinri in Part - Sanitized Com/ Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
does not bespeak the happy state of affairs in the field of selenium rectifiers. The
volume of their production and their quality is below the requirements of this country
? and the standards of the foreign samples. This is one of the causes for insufficient
development of automatic instrumentation.
?
we
Varistors
Among many remarkable features of semiconductors, there is one which has not
found sufficient application in Soviet industry End which is bound to be of consid-
erable interest to automation and telemechanics. Many semiconductors exhibit a con-
siderable deviation from the Ohmic Law; in silicon carbide, for instance, the con-
ductivity is proportional to the third and even fourth power of the applied poten-
tial (J Vn). Such a relationship between conductivity and potential makes it
possible to create nonlinear electric resistances that are very valuable to industry.
Abroad, such resistances are called varistors and, according to the foreign press,
they are manufactured by a number of firms for the most diverse purposes and for
varied potentials. Such resistances are made in the form of discs with diameters
raniing from one to several centimeters; their thickness varies from 1 to 10 mm and
even more.
This type of semiconductor devices practically does not exist in the Soviet
Union. A report on the last work done on them was published in the magazine
Avtomatika i telemekhanika Automation and Telemechanicg in the year 1951.
Recently some progress has been noticed in the work with this group of semi-
conductor devices. Thus, LETI is doing very successful work on the study of non-
linear relationship between the semiconductors and the strength of the electromagnetic
field and on the development of industrial models. However, the volume of this work
corresponds neither to the importance of this research trend to industry, nor to the
possible large-scale use of varistors. It is important to resume and expand this
research in the/Acadany of Sciences (Semiconductor Institute) as well as in the branch
institutes. In the first place, it is important to reestablish the development vrork STAT
49
rIfr4in Part Sanitized Coov Approved for Release @ 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
on the industrial types of varistors and on the technology of their industrial pro-
duction at VEI, where work in the past has been on a very high level; at the same
time, developmental work on these automation elements should be included in the activ-
ities of the Institute of Refractory Materials of the Ministry of Ferrous Metallurgy,
New Semiconductor Devices
The number of semiconductor devices described above, which are extensively
used in industry and particularly in automation and telemechanics, was recently in-
creased by the addition of new original devices. The large values of the thirmo-
electromotive force of semiconductors made possible the development of new types of
sources of electric energy, e.g., thermogenerators or, as they are also called,
thermal batteries. The honor for the development of the theory and of the industrial
models of the thermal battery belongs to the Semiconductor Institute of the Academy
of Sciences. This. institute also developed ad electric refrigerator using semicon-
ductors. Both of these types of semiconductor devices will be manufactured in the
near future in sufficiently large industrial quantities to allow them to find their
proper place in automatic apparatus, first of all, as power sources, and to act as
thermostats for automation elements.
The conductance of semiconductors depends also on the magnetic field and the
pressure. The first property recently led to the development of another new semi-
conductor device, which is very useful for rapid and accurate measurement of a mag-
netic field. This device, which as yet has not received an official name, was con-
structed for the first time in one of the laboratories of the Semiconductor Institute
of the Academy of Sciences.
The second property of semiconductors--the change in conductance with pressure--
is at present time in the preliminary stages of investigation. Using this feature
of semiconductors, it is possible, in principle, to develop elements which can be
utilized to measure pressure.
Of interest among the new semiconductor devices is an electric light modulator
50
STAT
im,,,i,ecifinri in Part cnniti7ed CODV Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Cop Approved for Release ? 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
4
which consists of a thin plate of germanium, through which passes a light beam whose
value changes depending upon the strength of the current flowing in the plate.
Outlook for the Development of Semiconductor Devices
The number of semiconductor devices described above is based upon the utilization
of many physical properties of previously known semiconductors. The great successes
achieved in amplifiers, rectifiers, and photocells are related to the qualitatively
new developmental stages in the field of solid states--the study of electric proper-
ties of the purest substances and the manner in which the addition of individual
elements of the periodic Chart affects these properties. This investigative approach
led to the discovery of new qualitative properties of materials (large values of
drift mobility, changes of sign, etc.) and to a conscious control of these properties.
On this basis are constructed the above-described diodes, triodes, photodiodes, solar
batteries, and powerful rectifiers. A new illustration of the production and use of
the qualitatively new properties of semiconductors is the recently produced LFTI
semiconductor device which, by virtue of its characteristics, is analogous to a gas-
filled rectifier tube.
The study of materials of maximum purity is now the leading trend in semi-
conductor physics. This trend, which in many ways depends upon our chemical industry,
without any doubt is the reason for the improvement of the existing semiconductor de-
vices and for the development of new ones.
In addition to this trend, which ensures scientific and industrial progress in
the field of semiconductors, two new areas of research are noticeable at present.
The first of these arose as the result of endeavors to expand the number of sub-
stances having large drift mobility in order to supplement the group comprising only
two substances: silicon and germanium. On the basis of the crystallin-chemical re-
semblance, LFTI forecast the resemblance of the physical properties of silicon and
germanium to the binary combinations of elements of the third and the fifth group of
the periodic chart of elements. This was confirmed experimentally by obtaining large
STAT
51
npriaccifipri in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
drift mobilities in the new group of substances. As an example, we can cite the
following figures.
cm2
The drift mobility in germanium is 3,800 V. sec; however, in one of the new
. ' substances, InSb, the drift mobility, according to the latest data, is as high as
80,000 . The difference between these numbers illustrates perfectly well
v. sec
the possibilities for the development of such semiconductor devices as amplifiers,
rectifiers, photocells, and photodiodes.
Investigations of the semiconductor group with large drift mobility are now being
conducted here, in the Soviet Union, as well as abroad. The practical results of
these efforts will depend to a considerable degree upon the chemical industry be-
cause the problem of purity and admixtures is here decisive.
Within this new group of semiconductor substances, which has received in liter-
ature the name of the metallic group AMBIT the most interesting at the present time
are InSb, InAs, In?, and GaAs. Information appears in literature concerning pro-
duction, using the above substances, of first models of photocells, photodiodes, and
equipment for magnetic field measurement. The transistor effect lying at the basis
of semiconductor amplifiers also has been observed.
The second area of investigation directed toward the production of new semi-
conductor substances is developing in the direction of the synthesis of complex com-
pounds formed by three or more elements, and the study of their electric, physical,
and general physicochemical properties. This area of research has as its goal the
determination of general laws that bind electrical properties of a semiconductor to
its chemical properties and structure and has begun to produce useful results.
Thus, for example, several new tricompcnent chemical compounds were obtained (in
the systems Te-Sb-Se and Te-Sb-S); means of uninterrupted and significant variation
of conductance were established, etc.
Triple and more complex compounds in the systems of the group AMBITopen the
way for securing a range of semiconductor substances with smoothly varying properties
and of materials with even larger drift mobility. STAT
52
Dni-F rnpv Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
? An original result of the investigation of complex compounds was the discovery
of a new and large group of semiconductor substances, i.e., vitreous semiconductors.
?
Due to their electrical properties, these substances--obtained by fusing such elements
as arsenic, selenium, tellurium, sulfur, tallium, antimony, and others--are typical
semiconductors. Because of a number of their physical and chemical properties, they
can be considered as ordinary glass.
The current detailed investigation of vitreous semiconductors will in time de-
termine their place and signficance among semiconductor devices. However, judging
from the results of optical investigations,they can find their application even now
as optical filters for the infrared portion of the spectrum.
It follows from the above that the outlook of our semiconductor industry is
? great.
Conclusion
The above review of the present state and future prospects for tre development
of semiconductor devices, as well as a general acquaintance with the conditions of
affairs concerning their immediate utilization in the automation and telemechanics,
leads to the following deductions.
1. Semiconductor devices open wide vistas for the development of industrial
automation and telemechanics. However, at the present time there is a considerable
gap between the potential for using semiconductor devices and the extent of their
practical application.
The following semiconductor devices are being developed for industry: diodes,
triodes, photodiodes, large-capacity rectifiers, photoresistors, thermal resistors,
magnetic field measuring instruments (magnetometers), thermal batteries, and refrig-
erators.
None of these types of semiconductor has as yet found sufficiently large appli-
cation in practice.
The basic reasons for this state of affairs are the very insufficient level of
53
STAT
narlaccifiinri in Part - Sanitized Com/ Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Cop Approved for Release ? 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
industrial production of semiconductor devices and the lack of specialized organ-
ization for the development and manufacturing of automatic instrumentation for in-
dustry (photorelay, thermal relay, and others).
2. We are lagging considerably behind foreign industry in some semiconductor
devices, for example, varistors, germanium and silicon photocells, and some others.
The theoretical level of work on semiconductors in our country is basically
the same as abroad.
The lag behind foreign industry is due to the insufficient amount of scientific
investigation, particularly in experimental and construction work. It must be
stressed that our chemical industry is falling far short of meeting our research
needs in the field of semiconductors.
To ensure the most rapid increase in the use of semiconductor devices and fur-
ther progress in their development, it is necessary:
a) To intensify the volume of scientific research and experimental and
construction work on semiconductors at the Academy institutes and, in particular,
in plant laboratories and branch institutes, by creating the necessary conditions.
b) To require the chemical industry to organize induptrial production of
superpure elements. It is recommended that an Institute of Pure Elements be organ-
ized for this purpose.
c) To develop in every -way the work in the field of semiconductor mater-
ials, aimed at Obtaining new semiconductors with the best properties for industry.
c) To expand the production of semiconductor devices, and in particular
of diodes, triodes, large-capacity rectifiers, photoresistors, and thermal resistors.
To organize the production of new devices, such as photodiodes, photocells, etc.
e) To organize a numl)er of centralized production centers for manufactur-
ing the simplest automatic instrumentation utilizing semiconductor devices.
Comment
by Doctor of Physical and Mathematical Sciences V. M. Tuchkevich
The list of semiconductor devices which can now be found in the laboratories cSTAT
54
Dmri? niti7PCi rODV Approved for Release @ 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
_
be increased over the number mentioned in the report of B. T. Kolomiyets.
It is necessary to approach the problem of semiconductor equipment for automation
from a somewhat different point of view: not simply by describing the propertius of
certain pieces of equipment, but also by asking the question as to which instruments
are necessary for the automation and what requirements must be made of this equipment.
Semiconductor devices can be divided into two major groups: devices utilizing
properties of materials and devices utilizing contact properties. Each group of
these devices finds its own field of application.
Semiconductor devices at the present time have a number of shortcomings.
As we know, the parameters of the semiconductor devices are dependent upon tem-
perature. This property in certain cases is used to build corresponding sensing
elements, but in other devices this dependence upon temperature presents a condition
which can be a great oostacle, considerably limiting their field of application.
Many of the semiconductor devices do not now possess sufficient stability and their
parameters change with time and perhaps are also dependent upon other factors.
The parameters of semiconductor devices cannot as yet be reproduced with suf-
ficient accuracy. The law level of production results in a large scatter of para-
meters in those devices which are manufactured.
These are the main shortcomings of the devices, which require a special approach
their application.
The dependence upon temperature, which is practically absent in certain devices
of the nonsemiconductor type now in use, can be compensated for in the majority of
semiconductor devices at the expense of the reserve sensitivity or some other par-
ameters. It seems to me that the construction of various automatic equipment re-
quires a somewhat different approach on the part of the specialists in the field of
technical facilities for automation, than the simple presentation of definite, con-
crete requirements in these devices. It is necessary in many cases to construct
this equipment on a new basis, on new principles, taking into account such specific
features as are possessed by the semiconductor devices.
55
STAT
1
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Because certain groups of semiconductor devices have been distributed only re-
cently and that the industrial production of these devices is insufficient, not all
of their properties are known.
? The semiconductor devices have new properties different from the properties of
the amplifier vacuum tubes, and the problem consists, not in the replacement of the
vacuum tubes by the semiconductor amplifying devices, but in the development of com-
pletely new schemes which cannot be accomplished with the vacuum tubes. This problem,
it appears to me, has not been sufficiently discussed, probably because of insuf-
ficient knowledge of these semiconductor devices.
On the other hand, a number of semiconductor devices, particularly germanium
and silicon amplifier-rectifiers for possible use in automatic installations, are
alreaay being manufactured by our industry, though in an insufficient quantity.
These devices are produced without regard to the operating conditions and the.
specific requirements of the automatic systems, primarily for use in the radio indus-
try. When they are used in computer installations, it thus becomes necessary in
these machines to take into account entirely special requirements which are not con-
sidered when these devices are used in ordinary radio installations.
As a parameter of this kind, it is possible to point to the magnitude of current
within the closed circuit in a semiconductor device. This parameter is of no special
significance in the radio amplifying installations, but is exceedingly important in
the mathematical machines. If one is to consider the development of equipment for
automatic purposes, it is necessary to state the problem of the development of spec-
ial semiconductor devices for automatic installations, just as the problem is stated
here for the development of vacuum tubes for automatic installations.
It appears to me that, in planning our future work, it is necessary to examine
the semiconductor devices also from this point of view. It is necessary to formu-
late the requirements which must be fulfilled by these new semiconductor devices that
are designed for use in the automatic systems.
The aim of the ?resent SARCi n is not only to state that we have such devices
STAT
56
npnlassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Cop Ap roved for Release ? 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
at the present time and that they are still being produced in an insufficient quan-
tity, but also to outline the aims of our work in the future. I would like to men-
tion the new principles of construction of semiconductor devices which are projected
now and which, it seems, will represent an important area of research in the future.
The construction of the thyratron type devices can be considered as belonging to
this area.
Extensive application in a number of automatic installations will obviously be
found by a device based on a property, i.e., the modulation of a light beam by electric
current.
It appears that, by passing a modulated current through a germanium plate, it
is possible to modulate a light beam. This device will make it possible to obtain a
modulated beam of infrared rays.
Address
of Candidate in Technical Sciences N. P. Udalov
Information concerning the properties of semiconductor elements, the methods
of their design, and the experimental characteristics necessary for the design, is
not sufficient. Often the designers refuse to use semiconductor thermal resistors
only because they are not sufficiently familiar with the properties of these resistors.
The first attempt to utilize these devices in their designs?when not grounded on
proper calculations and experimental data--usually ends in failure; this discredits
the notion and people cease to occupy themselves with these problems for a long time.
Because of this, I have a recommendation: in the resolution of our session to
record the need for the Automation Institute to conduct, at the beginning of 1957, a
conference on ti..e problems of the application of semiconductors in automation, with
the wide participation of the representatives of various scientific developmental
institutions.
Semiconductor thermal resistors have great prospects of employment if such
thermistors can be made to operate on strong current. Unfortunately, our industry,
which makes stable thermal resistors, class not manufacture them in a sufficient
57
STAT
? rl
? ori ('rnv Anoroved for Release @ 50-Yr 2014/05/13 ? CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
a
number of models. Fifteen years ago, excellent thermal resistors, called "urdox,"
were manufactured atroad. They were made to carry currents up to 30 amp. It is
necessary to manufacture heater-type thermal resistors capable of handling the currents
of motors to be used as starting rheostats for electric movers, thermal resistors
with sharply falling characteristics for installations requiring large time delay,
from 1 msec to tens of minutes.
58
STAT
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
D. V. Zernov, M. I. Yelinson, and A. M. Kharchenko
PROSPECTS FOR THE DEVELOMENT OF NEW TYPES OF ELECTRONIC EQUIPMENT
FOR AUTOMATIC AND TELEMECHANICAL INSTALLATIONS
Introduction
Modern automatic and telamechanical installations are unthinkable without the
extensive use of electronic equicment. As we know, large-scala replacement of
mechanical aad relay elements and systems with electronic equipment was started a
long time ago (15-20 years back) and this triggered the considerable progress in
automation and telemechanics because of the sharp increase in the speed, sensitivity,
and accuracy of the automatic installation. As the use of electronic equipment
increased, the number of such devices grew considerably.
The last decade is further characterized by a large increase in the use of
automatic installations of various types of gas discharge equipment (thyratrons of
varcous types, glow discharge tubes, decatrons, plasmatrons, ignitrons, etc.),
Finally, during recent years, the semiconductor elements began to assyme particularly
varied and growing importance because of their advantages.
Naturally, it is impossible to discuss completely in this short presentation
the entire subject of the use of all the above-mentioned equipment in automation
and telemechanics. Because of this, we shall limit ourselves to the analysis of the
present-day status and future prospects for the use of only the vacuum electronic
equipment, restricting ourselves in addition to the special types.
As is well known, the widest application in automation and telemechanics as
well as in the radio industry is enjoyed by the electron tube with the grid control-
ling the intensity (density) of the electron beam.. During the period of vigorous
mastery of the electron tubes, automation and telemechanics accepted them in the form
they were developed for radio needs. It is evident, however, that we need additional
requirements here for the reliability,durability, stability under external influences,
fidelity of reproduction of the electron tubes, as well as for some of the others of
their electrical parameters. Unfortunately, the industry, as a rule, is presently
STAT
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13 CIA-RDP81-01043R004500020003-3
59
Dar!. qnniti7PCI rODV ADDroved for Release @ 50-Yr 2014/05/13 CIA-RDP81-01043R004500020003-3
STAT
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
not manufacturing electron tubes especially made for use in automation and tele-
mechanical installations. This problem must be considered in the near future.
One hears opinions from time. to time that further improvement in the electron
tubes is not warranted since they will soon be replaced by semiconductor devices.
But this, like any other extreme point of view, is a rather one-sided way of looking
at this. In reality, this discussion will concern most probably only the partial
replacement. Very indicative in this respect is the evaluation of the prospective
use of electron tubes and semiconductor devices in U.S. aviation equipment.
According to information from 19 American firms, at the present time an average
of 22% of the tubes in aviation equipment have so far been replaced with semiconductor
devices. It is estimated that in the equipment being developed 38% of the tubes will
be replaced in 1956 and 50% in 1957. In many cases the replacement becomes unprofit-
able because of the increased complexity of the equipment. Thus it is considered
that about 40% of the electron tubes will be replaced with semiconductor devices.
The appearance and the vigorous development of semiconductor devices consider-
ably narrowed the prospects for the use of the electron tubes in many areas of
automation.
Side by side with the electron tubes, a large number of electronic devices has
been developed up to the present time.
The fact is that the electron tubes, first of all, cannot perform all the nec-
essary functions required by the automatic cycle. Thus, for example, electron tubes
in many cases cannot be directly used as sensing elements for various quantities
(displacement, acceleration, pressure, magnetic field, radiation, etc.) and cannot
perform many functional transformations, etc.
Second, the complexity of modern automatic and telemechanical installations
(for example, multiplicity and variety of modulation forms in telemechanical systems
or computing machines) requires a very large number of electron tubes. This threatens
the reliability of the operation of the system.
Apparently, the problem of greater reliability can be solved by two methods:
60
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
d?
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
either we sharply increase the reliability of each element of the system (electron
tubes), or we sharply reduce the number of elements making up the system.
The second method is tied up with the development of special electronic equip-
ment, capable of replacing a whole series of electron tubes and of corresponding
radio engineering arrangements. With this in mind, it is necessary to single out
the typical and most frequently used blocks of tubes and to attempt to find the
operating principles and to construct simple electronic equipment capable of perform-
ing the functions of these blocks. Experience shows that such solutions are found
in many cases and, as we shall show below, that they are very efficient.
The physical basis for the creation of the described special electronic equip-
ment is the great flexibility of the electron beam which, when concentrated by some
means, is capable of changing not only its intensity but also its position in space
under the influence of electric and magnetic fields.
The entire exposition below will be mainly devoted to the analysis of the
present-day status and prospects for future improvement and creation of new special-
ized electronic equipment. The authors of the present paper participated in the
development of many types of such equipment while on the staff of the former IAT
Electronics Laboratory of the Academy of Sciences USSR and, subsequently, on the staff
of IRE of the Academy of Sciences USSR.
Below we shall examine the use of special electronic devices, such as sensing
devices, amplifiers, distributors, converters, functional tubes, and memory elements.
Sensors
Among all the types of electronic sensors let us examine only the radiation
sensors and the magnetic field sensors.
- 1. Radiation Sensors
Photoelectronic methods of radiation indication have found a wide acceptance in
automation and telemechanics; they are extensively used as photocells with external
photo action and as photoresistors and rectifying photocells.
61
STAT
Declassified in Part - Sanitized Copy Approved for Release 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Ap roved for Release ? 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
0,g
Photoresistors and rectifying photocells are very simple devices, possessing
a very large integral sensitivity (up to 10 a/lumen).
Their characteristic deficiencies are the poor threshold properties and large
inertia which narrows their field of applications. It is important to note that the
operating stability of these devices is still unsatisfactory (instability of the
output signal, aging, sensitivity toward temperature, humidity etc.).
The vacuum photocells are made at the present time with several types of photo-
electronic cathodes: antimony-cesium, cesium oxide, bismuth-cesium. So-called multi-
alkaline cathodes are now being developed, i.e., cathodes which contain, in addition
to cesium, other alkaline metals (potassium, sodium, etc.). These types of photo-
cells are designed for the visible and the nearest region of the infrared spectrum.
For the ultraviolet part of the spectrum, photocells are being developed with
pure metallic cathodes (magnesium, beryllium, cesium, etc.); however, the optimum
type of cathode has not been found yet. Magnesium photocells are now being manu-
factured.
The sensitivity of the vacuum photocells does not exceed 40-200 j.Qa/lumen.
Therefore, the use of these photocells for automation naturally requires an additional
amplifying system.
In contrast with photoresistors and rectifying photocells, the vacuum photo-
cells have practically no inertial and are more stable; these properties assure them
of a sufficiently large area of application.
Particularly promising is the use of photomultipliers (and of electronic multi-
pliers in general). As we know, the photomultipliers use the secondary emission for
the amplification of the electronic current; this enables them greatly to improve
their threshold properties. Modern multipliers, without cathode cooling, are capable
of registering light beams up to 10-12 and 10-13 lumens. Photomultipliers are now
manufactured with cesium-oxide, antimony-cesium, and purely metallic photocathodes,
having different types of construction depending upon their purpose.
An antimony-cesium layer, as well as oxygen activated copper-magnesium and STAT
nprlaRRifipn in Part - Sanitized Copy Approved for Release
62
50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Co .y Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
NT,
aluminum-magnesium alloys are used as secondary electron emitters.
Thirteen-stage multipliers requiring a supply of 1,000-4,000 v and few-stage,
miniature, vibration-proof multipliers requiring 600 v are being manufactured. The
integral sensitivity of multipliers is very high; it attains 30-40 a/lumen.
The average output currents of the new types of multipliers using alloy-type
emitters attain several milliamperes; under pulse operation, they attain several
amperes.
With the aid of the electronic multipliers, it is .possible to register not only
electromagnetic radiations within a wide range of wavelengths (from infrared to
T-rays) but also the corpuscular radiations of various types (electrons, GC-particles,
ions, etc.).
There are many uses for photomultipliers. One must mention here, first of all,
the automation of the atomic industry and of atomic power, the automatic installations
of experimental nuclear physics, and also the automatic installations in many branches
of industry using radioactive isotopes.
Photomultipliers are very valuable in automatic control and regulation instal-
lations in the metallurgical, metalworking, chemical, and other branches of industry
and also in space navigation.
Thus, a further improvement and development of new photomultipliers and of
electronic multipliers in general is warranted in order to create specimens of high
stability and sensitivity that have definite kinds of spectrum characteristics. It
is also important to produce multipliers having a large power output, and also
natural pulse duration, i.e., a small spread of electron transit times.
2. Sensors of the Magnetic Field.
In automated and telanechanical installations, a growing place is being assumed
by the magnetic recording of signals with their subsequent reproduction, transformation,
etc.
As we know the reproduction of signals is being accomplished at the present
time with the aid of electromagnetic heads in an imperfect manner because of the STAT
narlaccifiPrl in Part - Sanitized Copy Approved for Release
63
50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
?
?
special frequency characteristic of these heads (the output signal increases with the
increase in frequency) and their law sensitivity.
It is practically impossible to reproduce a time constant signal, while the
frequency correction within the limits of the sound range requires cumbersome
equipment.
The same problem of signal reproduction on the magnetic tape without drawbacks
inherent in the electromagnetic head, can be easily accomplished by means of a very
simple electron-beam device (see Figure la).
The device consists of an electron gun producing an electron beam. This beam
can be deflected by the magnetic field under investigation; for this purpose, the
.7/7
a
Figure 1. Electron-beam device for measuring a magnetic field.
a - schematic diagram; b - output characteristics; 317 electron gun.
Legend: a) output 2;
b) Gauss;
c)
d) output 1.
device is inserted into a magnetic circuit the same as the coil of the electro-
magnetic head. To increase the magnetic induction in the regibn of the deflection of
the beam, permalloy poles are used inside the device. The output arrangement of the
tube consists of two split metallic plates. When there is no signal (deflecting mag-
netic field), the electron beam falls on both plates equally, so that there is no
signal in the differential output of the tube.
The presence of the signal deflects the beam, depending upon its magnitude and
sign, toward one of the output plates, with the result that an output signal appears.
In contrast to the electromagnetic head, the signal is proportional to the magnetic
STAT
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
64
narlaccifiPri in Part - Sanitized Com Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Co
A proved for Release
50-Yr 2014/05/13 CIA-RDP81-UluLior.,,
field itself, and not to the speed of its variations with time. Figure lb shows
the relationship between the magnetic field and the output current for a device de-
veloped by IRE of the Academy of Sciences USSR, in 1955.
The curves show that ti-era is linearity within the limits of t0.3 Gauss, which
considerably exceeds the values of tle signal registered on tape. The sensitivity
of the device with a load-resistor of 100,000 ohms is 10 v/Gauss; this is about 100
times higher than the sensitivity of the electromagnetic heads.
The frequency characteristic is independent of the signal frequency within the
limits of 0 to 104 which in many cases is an invaluable property of the tube.
Apparently, it will be possible in the future, to make similar electronic devices,
capable of even larger output currents, possessing an even higher sensitivity toward
the magnetic field, and suitable for the reproduction of magnetic tape signals as
well as for other applications related to the reaction to the weak magnetic field.
Unfortunately, the sample tubes developed by the laboratories are not yet
sufficiently used by industry. It appears to us that in the future a wider use of
these tubes should be urged upon industry.
Amplifiers
Signal anplifiers are the key part of the autoratic cycle. The electron tube
was until recently the undisputed master in the electronic equipment for the ampli-
fication of weak signals. (Cie naturally do not consider here the magnetic ampli-
fiers.)
The semiconductor amplifier is now becoming a powerful competitor of the
PAlease 2?14105/13 CIA-RDP81-01043R004500020003-3
electron tube.
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
field itself, and not to the speed of its variations with time. Figure lb shows
? the relationship between the magnetic field and the output current for a device de-
veloped by IRE of the Academy of Sciences USSR, in 1955.
The curves show that there is linearity within the limits of I0.3 Gauss, which
considerably exceeds the values of tle signal registered on tape. The sensitivity
of the device with a load-resistor of 100,000 ohms is 10 v/Gauss; this is about 100
times higher than the sensitivity of the electromagnetic heads.
The frequency characteristic is independent of the signal frequency within the
limits of 0 to 104 which in many cases is an invaluable property of the tube.
Apparently, it will be possible in the future, to make similar electronic devices,
capable of even larger output currents, possessing an even higher sensitivity toward
the magnetic field, and suitable for the reproduction of magnetic tape signals as
well as for other applications related to the reaction to the weak magnetic field.
Unfortunately, the sample tubes developed by the laboratories are not yet
sufficiently used by industry. It appears to us that in the future a wider use of
these tubes should be urged upon industry.
Amplifiers
Signal amplifiers are the key part of the autoratic cycle. The electron tube
was until recently the undisputed master in the electronic equipment for the anpli-
fication of weak signals. We naturally do not consider here the magnetic ampli-
fiers.) The semiconductor amplifier is now becoming a powerful competitor of the
electron tube.
But, as we mentioned in the introduction, the electron tubes are still very
important. Interchangeable long-life tubes, unaffected by external conditions, must
be developed. It is desirable to improve the parameters of the electron tubes. Thus,
for example, it is very important to develop tubes with high transconductance. It
is rewarding to note the appearance of tubes, such as 62;h11P with transconductance
of S = 28 ma/v with anode current of 26 ma and anode potential of 150 v (c 13 13 ict./af)
and of the output tube 6E5P with the transconductance of S = 27 ma/v, with anode STAT
65
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Cop Approved for Release ? 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
current of 45 ma and potential of 160 v (c 16
16
There seems to be good prospects for the use of secondary emission to increase
the transconductance, which becomes Possible as the development progresses in the
efficient and current stable secondary emission emitters. The principal advantage
of such tubes lies in the possibility of increasing their transconductance without
an increase in the interelectrode capacity; this favors tleir use as wideband and
impulse amplifiers. This results from the location of the secondary electron emitter
(dynode) behind the control grids (Figure 2a).
Tubes with the secondary emission are suitable for use as triggers for the gen-
eration of short pulses etc.
A number of such tubes was developed abroad (EFP60, VX5038, Valmark tubes, and
others). Thus, the EFP60 tube has a transconductance of S = 25 ma/v with anode current
of 20 ma and input capacitance of 0.004 je,./4,f. The same tube, operating in the regen-
erative circuit, may ensure the steepness of the pulse rise in several .millimicro-
seconds. The resistance of the rising part of the characteristic operating under
the above conditions (with 30 v dynode supply) is smaller than 0.5 ohms (Figure 2b).
Of great interest are tubes with secondary electron emission, which utilize the
beam deflection instead of the grid control. In this case, it is possible to use
the positive characteristic of the anode as well as the negative characteristics of
the dynode (Figure 2c), and thus to make the quiescent current of the tube (in the
absence of a signal) equal to zero. Such tubes (with current rs-10 ma and trans-
conductance of 1 to 2 ma/v) are successfully used for differential amplifiers,
voltage regulators, multivibrators, binary storage elements, coincidence circuits,
etc.
We consider the development of electron tubes with secondary emission very
promising.
Amplifiers for small signals are very important in automation. The amplifier
EMU-2, manufactured by our industry, uses tubes 2E2P and is employed under special
operating conditions, amplifies a current as weak as 10-14.
66
STAT
nariaccifipci in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13 CIA-RDP81-01043R004500020003-3
4 is
vti
0, 4 I
S.
:!) a
Figure 2. 1-a- Cross-section of a tube with grid control
utilizing secondary emission
2 - cathode; 3 - grids; 4 - dynode; 5 - anode; 6 - v; 7 - kilohm;
8-b- Characteristic curves of tube EFP60;
9-c- tube with beam deflection and utilization of
secondary emission:
10 - controlling grid; 11 - accelerating electrode; 12 - focusing system;
13 - deflecting plates; 14 - electron beam; 15 - collector; 16 - dynode;
17 - anode; 18 - restraining grid; 19 - width of beam; 20 - beam deflection.
Characteristic curves: I - anode; II - dynode;
III - total output.
As further development of amplifiers for weak currents, it is desirable to make
special electron tubes with grid currents of 10-16 to 10-17 a (electrometer tube
2E2P has a grid current of about 10-15 a).
67
ISTAT
Darf qnniti7PO CODV Approved for Release @ 50-Yr 2014/05/13 CIA-RDP81-01043R004500020003-3
STAT
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Distributors
In modern automatic practice, and particularly in telemechanics and computer
technology, many different electronic pulse circuits are used in addition to the
usual amplifying systems. These circuits perform the functions of generation, shap-
ing, conversion, distribution, delay, counting of electronic pulses, and so on.
These systems consist in the main of ordinary receiving-amplifying tubes and
semiconductor devices; in many cases the number of tubes and semiconductors in these
installations becomes very large, amounting to hundreds and even thousands of pieces.
the limited life span of electron tubes and the lack of identity in the parameters
of tubes and in particular of semiconductor devices, impedes the full utilization of
installations employing these devices and makes then insufficiently reliable.
In this connection, as we said in the introduction, a need arises for the devel-
opment of special electronic equipment, which will combine the functions of more or
less complex tube blocks and which will considerably decrease the total number of
ehctronic and semiconductor devices, utilized in this or another complex system.
By this means, of course, it will be possible to achieve an improvement in the re-
liability of the entire installation, even though the service life of each device is
relatively of short duration. From this point of view, the electron beam devices
of the commutator type become very promising; they control the current in the output
circuits not only by means of the density of the electron beam, but also by changing
the position of the beemwith respect of the system of output electrodes. It is
relatively easy to combine in these devices a large number of inputs and outputs and
to obtain arbitrary forms of dependence of the output currents cm the imput potentials.
This gives them great flexibility and universality and makes possible the construction
of equinment for the most diverse practical objectives. One of the pioneers in the
development of such types of equipment is the large IAT Electronics Laboratory of
the Academy of Sciences USSR, which produced several new types of electron-beam
commutating devices, some of which have already found application in industry.
A new type of electron-beam commutating device has been developed and has passed
68
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
the tests this year at the IRE of the Academy of Sciences USSR. It is a ten-cell
? ring trochotron, possessing considerable advantages over earlier types of linear
trochotrons, namely: simplicity of construction, large output currents (up to 10-20
ma), and independence of output circuits from control and commutating circuits.
The principle of operation of the trochotron, as we know, consists in that the
electron beam of a given magnitude emitted by a cathode is formed into a ribbon-shaped
beam by the intersecting and approximately homogeneous electric and magnetic fields
and follows the equipotentials of the electric field.
By changing the electrode potential, it is possible to direct the beam in the
desired direction and to lock the position of the beam.
2?, .4 ,
4 4 .,,,/\ifi)( 4
410 1 -
631 fr)1 4 f.-.13;')--4
t_N17
bi yil
7\4 -
x Y
A-17
g
Figure 3.
1 -a- schematic arrangement of construction and switching of the binary
trochotron:
2,3,4- Ll, L2, L3 - switching electrodes, blades; 5,6,7 - 151, P2, P3 - re-
ceiving electrodes, plates; 8,9,10 - 111, U2, U3 - control electrodes;
11 - b - binary trochotron type KL-6;
12 - c - double switch;
13,14 - Pl and P2 electrodes of the first deflecting system; 15,16,17 -
n2, electrodes of the second deflecting system; 17,18,19,20 - Bl to B4 -
69
P,
STAT
in Part Sanitized Coov Approved for Release @ 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
p.
output electrodes, possessing secondary emission capability; 21, 22,23,24 -
b1 to b4 - output (trap) electrodes; 25 - stubs, focusing electrodes.
The cross-section of the ring trochotron with some of the arrangement elements
is shown in Figure 3a. The magnetic field is directed perpendicularly to the drawing.
As the figure shows, around the cathode are symmetrically located 10 cells
formed by 10 switching electrodes (blades) and 10 receiving electrodes (plates). A
control electrode is located in each cell.
If a resistance is inserted in the circuit of each corresponding blade, and
connected to the positive pole of potential source (smaller than the magnetron cutoff
for a given construction); the device will have 10 working, stable beam positions
when the beam is locked in one of the cells, and one stable position when the beam
is "extinguished" (cutoff condition of the magnetron). When a negative pulse is
applied to one of the blades, tie beam is locked in the corresponding cell because
a small portion of the beam current (about I ma) produces a potential arop in the
resistor of the blade circuit, which ensures a law potential of the blade. The
major portion of the beam falls on the plate of the given cell.
If a negative pulse is applied to the control electrode of this cell, the
beam will be transferred to the blade of the next cell:in the direction of switching.
The beam current reaching this blade produces a potential drop in the resistor of
its circuit, and the beam is locked in this next cell.
In this manner, by applying negative pulses to the control electrodes, it is
possible to switch the beam from one cell to the next one consecutively.
To increase the operating reliability of the trochotron and to decrease the
requiremehts imposed upon the counting pulses, the control electrodes are connected
in two interlacing groups, road the negative pulse potentials are consecutively applied
in each group. It is advisable for this purpose to install in front of the troch-
otron a trigger and connect the anodes of its tubes to the controlling electrodes.
A unit containing a trochotron can be used as a counting decade (the resolving
time of this decade is 0.2 it4sec).
70
STAT
Darf qnniti7Pd Coov Aooroved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Cop Approved for Release ? 50-Yr 2014/05/13 CIA-RDP81-01043R004500020003-3
???
If the plates of the trochotron are borught out of the tube envelope, they can
then be used as the large capacity output electrodes with large output current
(10 to 20 ma).
The ring trochotron can be used as a multichannel distributor because it is
possible to connect in series several trochotrons and thus increase the number of
commutating circuits.
The trochotron can operate with mall potential supply (24 v) together with
semiconductor amplifiers.
The new type of trochotron, according to the information from abroad, has found
many uses, e.g., pulse counting, frequency divider, distribution in systems with
modulation of one carrier frequency simultaneously by several independent audio
channels, gate systems, matrix systems, coding, time measurement, etc.
The ring trochotron can also be used as a component part in telemetering, tele-
control, radiolocation, homing systems, automatic control, etc.
In spite of the fact that we started the development of trochotrons earlier than
other countries, the trochotrons are already being massproduced in Sweden and in the
U.S.
According to the foreign information, because of the optimum operating conditions
of the cathode in the ring trochotron, its service life is very long, i.e., 50,000
hours. We find it very worthwhile to organize the widespread application in industry
of the several models of ring trochotrons as fast as possible.
Electronic digital calculating machine represents one of the most accomplished
automatic arrangements. At the present time they are constructed with electron
tubes. It appears that in the near future semiconductor devices will be widely used.
? However, it is worthwhile to examine the problem of utilizing, in the special pur-
pose digital computers of the nonstationary type, specially developed electronic
equipment, capable of replacing an entire system requiring a large number of ordinary
elements. This will greatly simplify the computer and will make it more reliable.
STAT
worthwhile because in portable computers, oper-
71
ir Dmr+ niti7Pd rODV Approved for Release @ 50-Yr 2014/05/13 CIA-RDP81-01043R004500020003-3
STAT
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
? *a
ating under variable ambient conditions, the semiconductor elements are not stable;
their stabilization would require cumbersome arrangements. The economic factor may
also prove to be important.
Foreign literature describes several types of special electron tubes, combining
in one device the functions of several elements of calculating arrangements.
The former IAT Electronics Laboratory of the Academy of Sciences USSR built a
universal digital logical element; in 1955 the IRE of the Academy of Sciences USSR
built a special electron-beam device--a binary switch, operating according to a
logical scheme with two inputs and four outputs (Figure 3c) and with an output current
of about 5 ma at 350 v.of potential supply. Two devices within a single enclosure
Inputs
Outputs
1
2
1
2
3
4
O
o
1
o
o
o
O
1
o
1
0
0
1
0
0
0
1
0
1
1
0
0
0
1
can operate as a complex summation unit. These devices are operating directly from
one another without intermediary elements. The results of the pulse testing of the
binary switch can be summarized as follows: with 100,000 ohms in the load circuit
of the output electrode and a capacity of 30 Aftf, the time for the appearance of
the output voltage is equal to 1.5 to 3 iusec with a potential gradient of 100 to
150 V.
Further improvement and the search for the new types of electronic devices of
the commutating type that perform the distribution function should be undertaken
without a doubt.
Converters
The problem of signal conversion from one kind into another is often encountered
in the automatic and particularly in the telemechanical installations (different
types of modulation of signals coming from sensing elements, conversion of one type
of modulation into another to increase interference-killing features or to increase
72
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
the ease of reception, conversion of a constant signal into a pulsating one and
conversely, etc.). The flexibility of the electron-beam devices, permitting per-
formance of the most diverse types of conversions, is strikingly illustrated in the
solution of these problems. A number of special devices of this type have been
?
developed up to tle present time. Lack of space does not allow us here to describe
all existing types of devices; hence, we shall examine only the coding and electronic
contact tubes.
The problem of coding the transmission, i.e., the conversion of any level of a
constant signal or of a signal varying with time into a series of pulses, set up
according to Some rule, can be realized with the aid of several types of electronic
devices.
,o
Figure 4. Coding tubes.
a - pulse coding tubes
2 - electron gun; deflecting plates; 3-131 - first, 4-P2 - second,
5 - screen with openings; 6 - collector;
b - tube with pulse-time modulation and quantization of amplitude:
7 cathode, 8 - focusing arrangments; 9 - plates for pulse time modulation;
10 - screen with a slot; electron beam; 11 - nonmodulated; 12 - modulated;
13 - input plate; 14 - screen with openings; collectors; 15 - output;
16 - auxiliary. STAT
73
im,,,i,ecifinri in Part cnniti7ed CODV Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
? ?
One of these is an electron-beam tube having two pairs of deflecting plates
with electrode openings and with receiving collector (Figure 4a). The signal to
be coded is applied to the horizontal deflecting plates and the short pulses are
applied to the vertical ones.
The arrangement of openings in the screen is chosen in accordance with the nec-
essary coding requirements. It is evident that the coded combination in the output
of the tube, set up on the output screen by the horizontal pulsating deflection of
the beam, will directly correspond to the given level of the coded signal. The speed
of operation of such a tube is law due to the weak current of the beam, but it can
be greatly increased through the use, for example, of the secondary electron emission.
To narrow dawn the necessary frequency band or to transmit additional information
with a given frequency band it is expeditious to use "quantization," i.e., splitting
into separate levels of signal amplitudes in common with the pulse modulation.
The construction of the tube capable of pulse modulation and quantization of
amplitudes is shown in Figure 4b. The flat-shaped beam passes through the first
deflecting system, which is supplied with an auxiliary variable potential. As a
result, this beam passes through the second pair of deflecting plates, supplied with
the signal, and reaches the output arrangement consisting of a plate with openings
of special Shape and of the collector electrode. If the beam has a constant density
along its width, we will obtain a pulse-modulated quantized signal on the output
collector. The tube gives good results with the frequency of the alternating poten-
tial on the first pair of deflecting plates equal to 5 mc and with a microsecond saw-
tooth signal on the signal deflecting plates.
Extensive use in automated and telemechanical installations, as well as in tele-
phony, can be made of special electron contact tubes, utilizing secondary electron
emission and functioning as converters of a constant signal into an alternating one,
as contactors, switches, etc., and differing by the fact that they possess a very
small internal resistance kilohm). The appearance of these tubes for the first
74
STAT
npnlassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
STAT
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
a
?
Or
.??
?
Figure 5. Contact tubes utilizing secondary emission.
a- unidirectional contact; b- arrangement for transmission of an alter-
nating signal through the contact; c- oscillogram of the modulated
signal; d- two-way contact; e- contact with grid control.
75
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
STAT
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
time made possible the actual replacement of mechanical contacts with electronic
ones in a number of new cases; this will make it possible to increase considerably
the operating speed and the reliability of contacts, and sharply to reduce the paver
required for the contact control.
In these tubes, a flat electron beam is formed by the electron-optical system
and is directed toward the secondary emission contact. The principle of operation
of the contact is illustrated in Figure 5a. The primary electrcns p strike the elec-
trode a which is at the positive potential. If the secondary emission from the
electrode a is absent, the current at this electrode, depending upon its potential,
will vary according to the curve I. However, if the electrode a is an efficient
emitter of secondary electrons and if the nearest electrode b has a higher potential
(Ub Ua) , then the volt-ampere characteristic of electrode a will correspond to
curve II. If, further, a resistance is inserted into the circuit of the electrode a,
then, having supply voltage Ub, the stable equilibrium potential (Ua) will be de-
termined by the point of intersection of the straight line III and the volt-ampere
characteristic II. Potential (Ua) is approximately equal to potential Ub, which
amounts to a closing of contacts between electrodes a and b. It is evident that the
potential of electrode a will follow the variations of the potential of electrode b.
The closing and the opening of contacts can be accomplished by the deflection or the
blanking of the beam. The arrangement for closing of the electronic contacts for the
transmission of alternating potential is shown in Figure 5b.
The modulation of the signal (U = 20 v, f = 80 c) by means of the interruption
of the beam by the modulating potential (U = 60 v, f = 600 c) is shown in Figure 5c.
The described contact is unidirectional, i.e., the potential Uct follows Ub, but
Ua follows U, only within narrow limits.
To obtain a two-way contact, the contact arrangement of the te is made symmet-
rical, i.e., both electrodes (a and b) are efficient secondary-electron emitters
(Figure 5d). In this case, each one of the electrodes can operate either as an anode
or as an emitter.
76
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Several contact arrangements can be installed within one enclosure, and the
cormutation of several circuits can be accomplished through their use.
The electronic contacts of the described simple type can be connected in a series
of no more than 3-5 units. In order to connect a largo number of contacts in series,
it is necessary to separate the functions of the controlling electrode from the elec-
trode collecting the secondary emission electrons, as is shown in Figure 5e.
Input electrode c has the form of a grid and collecting electrode b, with a
high potential, is placed on the side. The main part of the secondary electrodes
strikes the collecting electrode. The current of the control grid can be made approx-
iraately equal to zero, if it is made of a material with the coefficient of secondary
emission d = 1.
Ten contacts can be connected in series with this type of construction.
The upper limit of the signal frequency which can be transmitted through the
contact is determined by the time constant of the contact which, when R = 1,000 ohms
and C = 3 Aiuf, gives us t = 3.10 sec-9.
The tubes developed at IRE of the Academy of Sciences USSR, R = 3,000 ohms,
C = 5 to 10 Alef, t= 1.5 to 3.10-9, have a noise level of 200 to 500 it.v in the
band up to 1 Mc. The minimum input signal at the frequency of 10 kc can possibly
be Lowered to several microvolts.
By analogy with the corrtact tubes, so-called signal tubes can be made, which
at their output produce pulses of a specific amplitude, with input pulses varying in
amplitude within very large specific limits.
The possible development will be electronic contact tubes with low level input
signal and small number of contacts in one enclosure and multichannel tubes with
average level of input signal (10-100/4v).
There is no doubt that the new problems arising in automation and telemechanics
will require the creation of new electronic equipment, performing the functions of
varied signal conversions. As has been pointed out, the special electron-beam de-
vices are the most suitable elements; in many cases, nothing can compete with them,
STAT
77
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
even when they are compared to semiconductor devices.
Multipliers and Functional Tubes
During recent years, special electronic tubes have been being developed for use
in analog computers (computing machines with continuous action).
Let us examine several types of multiplier tubes.
One of these consists of an electron-beam tube, with two pairs of mutually per-
pendicular plates for electrostatic deflection, and of a second pair of plates sur-
rounded with a coil which produces the axial magnetic field (Figure 6a). The poten-
tial proportional to the first factor is applied to the first pair of plates. The
current proportional to the second factor is put ti-rough the coil. It is easy to show
that the resulting deflection of the electrons, after passing through the first pair
of plates and the magnetic coil (due to the acquisition of the transverse component
of velocity), is proportional to the product of the factors.
The receiving system of the tube consists of two plates insulated from each
other. If the beam is located on the separation line, the output signal, in the
differentially connected curcuit is equal to zero. The amplified output signal is
applied to the second pair of the deflecting plates, which in this manner compensate
for tl-e total deflection. The magnitude of this potential constitutes the output
value, proportional to the product. A laboratory model of such a device was at one
time constructed in the former IAT Electronics Laboratory of the Academy of Sciences
USSR. -The accuracy of this multiplier, as demonstrated by theoretical calculations
and experimental investigations, is within 2 to 3%.
In the other type of multiplier, also consisting of an electron-beam tube with
two pairs of deflecting plates, the electron beam with a large circular cross-section
is aimed at a metallic screen, made of four quadrants and insulated from one another
(Figure 6b). If the currents in the first and third quadrants are positive, and in
the second and the fourth negative, it is easy to show that the algebraic summation
of currents will be proportional to the product of deflections x and y, and, in the
STAT
78
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
presence of a linear system of deflection, proportional to the product of the posi-
tive and the negative plate potentials. The accuracy of the multiplication ranges
within 2%; the speed is approximately 100 kc.
The third type of the multiplier which we would like to mention also consists
of an electron-beam tube whose basic feature is R deflection system which produces
a hyperbolic field. The construction of this multiplier is shown in Figure 6c. A
first deflection system (deflecting inthe x direction) is located in front of the
electron gun; a deflection potential U1 is applied to this system. The first system
a
Figure 6. Electron-beam multipliers:
a- with a magnetic coil; b- with a wide beam;
bolic field; 1 - electron gun.
c- with a hyper-
of deflection is followed by the second one, which consists'of four electrodes, pro-
ducing a hyperbolic field in the cross-sectional plane xy. This field is such that
the electric field intensity in the y direction increases linearly with the increase
STAT
79
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
STAT
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
of x, and is negative for x> 0, and positive for x In this manner the beam
deflected by the first system by a value x gets into a hyperbolic field which is
proportional to the deflecting potential ul. It is obvious that in such a field the
electron beam is deflected in the y direction proportionally to 111.122 with the sign,
which takes into account the signs of ul and u2. As in the first tyre of the multi-
plier we examined, a third deflectinc system is provided; it is connected with out-
put plates through a feedback circuit. The potential on the third deflecting system
will be proportional to the product u1.u2.
The accuracy of this arrangement is high, with an error of about 0.5%. The
limiting frequency is approximately 200 kc.
We must consider the long-range development and use of the examined types of
electron-beam multipliers, as well s their improvement, in order to simplify the
construction, to boost the accuracy, and to increase the operating speed noticeably.
There is a very good outlook for the development of electronic devices in which
the output current is any predetermined function of the input quantity, i.e., of
4(mu
Get,41offtlif 6
Figure 7. Quadrature tube--internal construction and
output characteristics.
1 - cathode; 2 - electrcn beam; 3 - deflecting plates; 4 - mask;
5 - collector; 6 - v; 7 Edeflecting plate'
function data units (formatrons). The operating principle of these devices consists
in the fact that the output current is the assigned function of the position of the
flat or round electron beam. The form of the output function is determined by the
shape of the output electrode (mask). It is apparent that only such devices can en-
80
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
??
sure most simply and most accurately and reliably the analog of the required func-
tional dependence.
An example of such a function tube is the quadratron, massproduced in the U.S.
(type QK 256) (Figure 7). The tube has a very simple construction and has the di-
mension of the receiver-amplifier tube. The electron beam, emitted by a cylindrical
cathode, is formed by special electrodes into a disc-shaped beam, which is focused
on the mask. This mask consists of a cylinder with several cutout parabolic openings.
Behind the mask is placed the electron collector. The same figure shows the char-
acteristics of the output current, which approximates very closely the relationship
2
iout m KUin. The total tube current is about 40 ma, which ensures practically any
speed of operation. The accuracy is close to 99,0. The formatrons can find extensive
application as amplifiers, mixers, quadrature detectors, phase and frequency mod-
ulators, limiters, automatic regulators, frequency multipliers, discriminators of
amplitude, modulated waves, etc.
Of considerable interest to calculating technique are the so-called matrix
tubes, which are also functional data units.
Memory Elements
The tendency to complicate the functions performed by the automatic instal-
lations calls for the emergence of new technical means for preserving the applied
signals with the possibility for subsequent reading and erasing, i.e., in prepar-
ation for the new act of retention. To perform this function of memory, a number of
special electron-beam apparatuses can be proposed, all of which use thin dielectric
layers as retention elements, with an electron beam as the recording and reading
agent, capable of producing local positive or negative charges on the surface of the
dielectric layer.
During recent years several such types of apparatuses have been developed (poten-
tial scopes with barrier grid, with dielectric grid, graphectrons, selectrons, tubes
with retention of illumination, etc).
81
STAT
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Memory electron-beam tubes which, with approximately 2,048 elements give about
2,000 rotations at a speed of 4 iusec, were put into massproduction for the electronic
calculating machines here in the USSR in 1956. According to the published inform-
ation, the American firm IBM in 1954 ceveloped a tube with a barrier grid which,
with 3,000 elements, is capable of 2,000 rotations; the same tube with 10/000 elements
is capable of 100 rotations. The Bell System describes a tube (1955) which with
16,000 elements is capable of more than 100 rotations at a speed of 2.izsec.
The current requirements for tubes working in the operating unit of memory
machines consist of 4,000 rotations with 4,000 elements and a speed of 1 A..sec.
It is hoped that these parameters will be attained in the near future.
Conclusion
The material presented in this article shows that a number of problems in the
field of the construction of automation and telemechanical equipment can be success-
fully solved by using special electronic equipment, in particular, through the use
of equipment employing the principle of the control of the output current by the
control of the location of the electron beam with respect to the output electrodes.
Many types of such equipment have already been developed abroad as well as here
in the Soviet Union.
There is no doubt that the ecuipment which has already been developed here or
which is known from publications would find extensive use if it were manufactured by
our industry.
As we know, automatic and telemechanical equipment is being developed by a
large number of scientific research establishments, design bureaus, and industrial
laboratories. An unusual situation resulted in the process of these developments.
In many cases the systems development personnel recognized that the best solution can
be secured through the use of special electronic equipment for which they can formulate
technical specifications. But past experience shows that, if they decide on such an
approach, they will as a rule not be able to complete their development on time. Thus,
STAT
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
82
neclassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
STA
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
they were forced to use existing elements, although this often resulted in com-
plicated and unreliable solutions.
The USSR has a very inadequate variety of special electronic equipment capable
of successfully solving nany problems confronting-automation and telemechanics. The
development and, in particular, the acceptance of such equipment by industry is
progressing very slowly and this undermines the confidcnce of the users of the de-
veloped products. It appears, for example, that there is a need to set up in the
MRTP system a special construction office with a good production base and an exper-
ienced plant Which could manufacture products from development models, adapt them
to industrial needs, and manufacture small quantities of new electronic equipment
as ordered by the users. This equipment must in all respects be usable in current
automatic and telenechanical.systems.
We believe that the creation and its efficient operation of such an establishment
will considerably help the development of industrial automation and telemechanical
facilities in our country.
83
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
STAT
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
L. I. Gutenmakher
PROSPECTS FOR TBE APPLICATION OF MAGNETIC AND CAPACITIVE UNITS
AND MX:MITTS IN THE AUTOMATIC SYSTFMS
The existing automatic systems are built essentially with the relay-contact
equipment. As an example it is sufficient to cite the automatic telephone stations
(ATS), railroad signalling systems, centralized control, automatic block control
(STsB),telegraph stations, telemechanics of the energy producing systems, of the gas
pipelines, of the oil pipelines, and production lines of machines and of automatic
factories in industry.
The number of operating pairs of contacts is very large (more than 10 billion).
The drawbacks of relay-contact equipment are known: impacts, jolts, and vibra-
tions can produce displacement and faulty closing of contacts; jamming of shafts,
dampers, and other movable parts can change the characteristics of elements and at
times make them inoperative. The sparking of contacts can be wie cause of fires and
explosions. The operation of this equipment requires frequent checkillg, adjusting,
and contact cleaning (for example, it is necessary in ATS to clean and adjust 1.5%
of the relays daily); it is necessary to provide access to the relays; thus, they are
placed on flat structures with an aisle between them (on racks).
Great hopes were placed in the use of electronic tubes and electron-bean tubes
instead of relay contact equipment. Rapid operation of the electron-tube equipment
made possible the construction of high-speed digital computers, electron-tube inte-
grators, models and trainers, and controlling machines. However, the alert service
life of tubes, their law efficiency, and the presence of delicate mechanical parts
(cathode, grid) do not permit the construction and sufficiently reliable use of com-
plicated automatic equipment which contains tube components.
The drawbacks of relay and tube equipment for a long time hindered the develop-
ment of automatic systems and the expansion of its field of application. It became
necessary to discover new principles for the construction of contactless and tubeless
elements and systems of automation that would be free of the above-mentioned defects.
84
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
The way to the solution of this problem lies in the use of semiconductor ele-
ments (crystal diodes, triodes), nonlinear resistances, magnetic elements (trans-
former and imped-nce relays and amplifiers, storage devices, commutators, logical
switches), and capacitance elements.
The automatic systems built with these elements will exhibit some characteristic
properties: static operation (i.e., they will consist of permanently connected elec-
trical networks made of mechanically nonmoving elements and units), vibration stabil-
ity, and moisture proofness; they will not be affected by the pressure of the ambient
atmosphere and they will operate with equal reliability in different positions in
space and when subjected to rapid accelerations.
Because of their great reliability, automatic systems can be constructed from these
elements, which can perform more complicated functions than can presently be performed
with many thousands and millions of separate elements in one network.
As a result of the design work on magnetic contactless and tubeless elements*
and assemblies for automatic systems, it now appears that it is possible to place more
than 100,000 relay cells in 1 cu m. The construction of units, in contrast to the
surface arrangement of contact aid tube systems, can be of a volumetric nature. At
the end of the assembly and testing, the units are filled with a viscous isolating?
material, are hermetically sealed, and possess a great mechanical and electrical stab-
ility. All elements are very durable. Their high-speed operation is close to the
speed of the electron tube equipment.
The mass use of these new industrial automatic facilities will bring about great
changes in the national econnmy of our country. The great reliability of the con-
tactless magnetic, capacitive, and semiconductor elements, as compared to the presently
existing elements in the relay contacting equipment, makes it possible to assume that
in 10-15 years the quantity of similar contactless equipment will increase by 100 to
1,000 times. This will result, not only in the increase of automatic telephone
stations, mathematicalmachines, railroad automatic signalling and other complicated
* This report does not separately examine crystal-triodes (transistors), which also
are of great importance in the creation of reliable miniature equipment for the auto-
matic systems. The possibility of the construction of such systems, using mainly
magnetic elements, is stressed on purpose. It is feasible to combine them with STAT
crystal amplifiers and relays.
85
in Part Sanitized COON/ Approved for Release @ 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13 CIA-RDP81-01043R004500020003-3
systems known to us new, but also to the appearance of qualitatively new automatic
systems. In this connection it is necessary to plan for the corresponding scientific
investigations and technical developments.
In examining the development of new systems of automation in the light of the
caning five-year plans, we must await the appearance of the new undertakings and in-
stitutions with new operating systems.
It is possible to see, for example, within this persepective a dense single net-
work of automatic information and control stations in our country, which will include
equipment of automatic telephone and telegraph stations of the new type. They will
include high-speed digital computing and controlling machines and statistical and
information machines.
Part of the stations--branch stations for information and control?will be tied
into a single system network of enterprises belonging to one combine and one ministry;
some of them will be located directly in the individual large enterprises and es-
tablishments.
Automatic Contactless Telephone Stations (ATS) of the New Type
The replacement of the contact telephone relay and step switches with the new,
contactless relays and commutators is within the realm of possibility in the very ?
near future. The realization of the new type of ATS represent one of the most im-
portant problems because this will result in the mass production of all the basic
elements of the contactless systems which are also necessary for other innumerable
uses in the national economy. In using the crystal and magnetic amplifiers for the
amplification of the intensity of the sound vibrations and with the use of high-speed
ferrite impulse relays, registers, and commutators, it is wise to plan for the appli-
cation of the digital interference-proof method of modulation. In this manner, the
entire commutation scheme and the ATS memory are reduced to the already tested arrange-
ment of units and elements of the contactless systems. Even now the ferrite magnetic
relay elements operate at a frequency larger than 300,000 operations per second and
86
STAT
'?' ' D,r+ rt.t. (nrv Aooroved for Release @ 50-Yr 2014/05/13 CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Co .y Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
by virture of their speed of operation they are suitable for the construction of the
new ATS.
The possibility of using new long-lasting capacitive and ferrite operative
memory installations will result in new interesting variations in the work of the ATS.
Thus, for example, in order to call any phone number busy at the time, it will only
be necessary to dial the number once. This number it: then transferred into the
operative memory of the ATS, with the result that the automatic arrangements of the
ATS watch that the connection with the dialing telephone will be made as soon as the
busy phone becomes available again. It will also be possible to install, for added
convenience, a signal on each telephone to indicate, during the conversation with
another telephone, that a third phone is awaiting the connection (signal "someone is
waiting for you").
The great speed of commutation and the presence of memory and controlling com-
puting installations will permit in the new A1S more efficient schemes of telephone
connections in each separate case.
Telegraph communications using the ATS system of the new type will be greatly
facilitated by the introduction of the pulse methods of transmission of telegraphic
signals and by the installation of amplifiers in. the output networks of the ATS. The
present-day telegraphic equipment, by virture of its principle of operation and con-
struction, represents complicated electromechanical instruments consisting of a
great number of relays and contacts.
In order to bring the number of telegraphic sets, for example, on a par with
the number of telephones, it is necessary to develop and massproduce small-size con-
tactless systems of telegraphic equipment.
It is necessary to develop complex telephone-telegraph portable apparatus.
The introduction into the cities and districts of a single telephone-telegraph
network through the AIS will give us new ways for the construction of control systems
used in the national economy. The entire correspondence among different establish-
ments, combines, ministries, and other organizations can be organized and conducted
87
rlarlaccifiPri in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
STAT
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
with the aid of telegraphic equipment and memory machines which receive the inform-
ation. This will result in a sharp reduction in time required for the transmission
and the processing of business .papers.
Automatic Statistical and Information Machines
The pulse systems for the transmission of various types of information will per-
mit the introduction of new forms of machines for statistical data processing con-
cerning the working of industry, the national economy, and transportation. Data per-
taining to the operation of establishments--receipt and use of materials, money re-
sources, distribution and utilization of vrork force, and tens of thousands of other
items of information--can be fed in parallel into the memory of high-speed, digital
statistical machines. This will make it possible to process with extreme speed the
material according to preset programs and to work out the necessary data required
for correct planning and control. This will result in an increase in the productiv-
it; of the employees of the state institutions and in an increase of the part played
by each worker who is using machine technology for the registration, classification,
and p-ocessing of all incoming and outgoing materials of the establishments.
It is necessary to add to this system the automatic reference machines* which
must contain reference material registered in long-lasting machine memory and pertain-
ing to science, industry, and the national economy. The reference machines will make
possible the process of utilization, in the national economy, of all the accomplish-
ments of science and industry and will result in the better use of all presently
known results of research and technical development in the future. They will also
improve the work of planning and controlline the national economy.
The need for the analysis and synthesis of the content of the world is scientific
* 1. Bystrodeystvuyushchiye nauchno-spravochnyye i statisticheskiye machiny
speed Scientific Reference and Statistical Machines7, Publishing House of the Academy
of Sciences USSR, 1954.
2. Puti postroyeniya informatsionnyich i statisticheskikh mashin novogo tipa,
L. I. Gutenmakher, /Ways to Construct Information and Statistical Machines of the New
Type7, Publishing House of the Academy of Sciences USSR, 1955.
88
im,,,ineeifiori in Part - Sanitized CODV Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
STAT
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
? and technical literature is now being felt by millions of workers in all branches of
the national economy. The uninterrupted growth of the number of books (up to 200,000
titles per year) does not permit even the most highly qualified workers to fell= the
publications in detail and to remember the needed information. It is well known that,
due to insufficient information, a substantial amount of effbrt and money is being
spent in vain by many establishments and enterprises. Figure 1 shows a block diagram
of such automatic reference machine.
Figure 1.. Block diagram of the scientific reference
information machine.
1 - machine "memory" (storage of the information units); 2 - reading
mechanism; 3 - converter of the ordinary information text and in-
terrogation into the machine text (pulse code), and conversely;
4 - input mechanism (introduction of information and interrogation
into the machine); 5 - output mechanism (printing of answers or trans-
mission of worked-out results (answers) through the telegraph
apparatus.
To solve this problem we need methods for text recording which will allow the
reproduction of the material with great speed as many times, as necessary, for 50 to
100 years and without damage to the material (until the content becomes obsolete).
To record, for instance, only the existing reference magazines accumulated up to the
present time, a memory machine is necessary with a capacity of about a billion
letters (25,000 printed sheets),- filled with telegraphic signs. In addition, it is
required that the size of the machine memory be relatively small. The memory of the
machine of a billion letters must occupy a space similar to the book storage of an
ordinary library. Present-day technology allows the reading of 250,000 pages of
text in one hour. A person can only study several pages of scientific text in one
89
STAT
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
hour. A person often reads to sort out the information he needs and partially
remembers interesting data he encounters on his way. These functions of sorting out
of information can be performed by the machines. Let us examine their possibilities
and the outlook for their development.
The machine can find and select in the memory the material which has given
qualitative characteristics. The entire material recorded in the machine memory is
thus reviewed for this purpose and is compared at all times with a certain number of
characteristics established by the interrogating person. As soon as the character-
istics of the material and of the request coincide, the selected article, reference,
or report is printed by the machine and is telegraphically communicated to the sub-
scriber. In the simplest case, the answer can contain the enumeration of the selected
items as a list of their numbers in the bibliographic yearbooks.
It must be noted in this case that the machine can select material according to
the combination of such concepts which cannot be foreseen with the aid of any system
of classification. Considering the great number of separate elementary character-
istics, the number of their possible combinations in response to an interrogation
is unusually large and practically cannot be exhausted. The theory of combinations
permits an easy determination of such "astronomical" numbers, which can be obtained
in counting the possible number of combinations to a question. Even 'withmodest
original data, this number will be larger than 101,000.
The comparison can be performed not only according to the qualitative character-
istics but also according to the quantitative data. For example, on can ask for the
data concerning the magnetic material which possesses certain given properties within
the current' frequency of 1 million to 10 million c, and which lies within the given
limits of magnetic permeability, induction, coercive force, mechanical strength, and
so forth. One can ask for information about the materials which possess maximum or
minimum values of given qualities.
The machine reading can be conducted also selectively according to the given
addresses. 'With the aid of an electric network, it is possible to send out currert
STAT
90
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part- Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
in accordance with a given program, in order to select any element of information
(reference, word sentence, l'ormula, diagram, etc.). In the same way as such elements
of information ar.. identified in the books by their "address"--book numbers, pages,
lines--so each element of informa,don within the system circuit of the machine has
its address. Just as subscribers of the telephone ATS can be reached by dialing
their numbers, so the information elements in the machine memory can be connected
by the commutator ("called out") to the mechanism for the analysis of their content.
The existence of the addresses makes it possible to conduct a search for the new
material using the literature references which are usually present in the scientific
articles, reports, etc. Such a process of searching for the material within the
internal circuits reminds one of the search through the association of ideas when a
person recollects more and more details of some question which interests him.
It is also possible to imagine the organization of a search for the required
information, using various tables previously introduced into the machine, which re-
late the particular concepts with more and more general ones. Thus, the machine
will permit the obtaining of newer and newer properties of the concepts within the
inquiry which interests us, replacing the particular concepts with more general ones.
Consequently, the machine "reading" as assigned does not represent, in this case,
a simple sorting out of information according to the characteristics of the inquiry.
The internal networks and the search program permit a rapid .and intelligent selection
of the needed information, a comparison of different data, and their analysis. In
many cases the inquirer may be interested, not in the bibliography, but in the anal-
ysis of the content of the articles. If one just imagines the tremendous amount of
information which can be examined by machine means, the need for and the prospective
uses of such machine reading becomes apparent.
The creation of such "reading" machines possessing a vast and stable memory can
radically change the methods for the use of the vast and inexhaustible potential
wealth which has been accumulated by human genius*. The development of such machines
* See the article by Academician A. N. Nesmeyanov, "Looking into the Future of
Science," Pravda iruth7, 31 December 1955.
91
STAT
im,,,ineeifiori in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
is one of the most important problems.
Translating Machines
The information machine cannot be constructed for the literature of only one
language. Its memory must accumulate the material of foreign literature. It is
apparent that, in order to utilize such literature, translating machines are needed
which muld automatically translate the foreign text into the Russian language.
Preliminary calculations showed that, using static durable memory and contactless
magnetic relays and commutators, it is possible to construct such a translating
.machine which will serve simultaneously about 1,000 subscribers connected to the
machine, through the systems of the city automatic telephone stations. Let us add,
for example, another 1,000 numbers of the translation machine to the numbers of the
future ATS of the new type; by dialing, for instance, on the ordinary telephone the
number 901, one will be connected through the subscriber's instrument to the trans-
lating machine. Let each subscriber have a special automatic typewriter. The key-
board of this typewriter transmits, by means of telegraphic code, the letters of the
foreign text which are manually assembled by the subscriber, while the machine types
the Russian 1)tters of the translated text, received from the translating station in
response to the assembled text.
With what speed can the subscriber work who is assembling the letters of the
foreign text on the keyboard? Let us assume that his speed consists on the average
of 100 letters per minute (about two letters per second). Thus, the assembly by the
subscriber of a sentence containing 20 words of 10 letters each (on the average) will
recuire 120 sec, while the future mchine will be able to make a translation of more
than 10 such sentences in one sec.
There will be within the instrument of each suoscriber a memory mechanism (made
of magnetic registers) for the receipt and transmission of coded signals of one sen-
tence.
The address system for the search of the address of a vocabulary unit in response
STAT
92
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
STAT
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
to an assigned word can be constructed from a system of two-coordinate ferrite grids.
Statistical Machines
The statistical machines, located in the autDmatic stations of information and
control, will receive, through the telegraph lines from establishments and organ-
izations such data, such information as is needed for durable safekeeping and for
processing as the need arises for planning and control.
The simplest of them can be used for planning and control in commerce. Thus,
for example, information about the sales progress and the results of receipts due
to the sale of various merchandise items can be fed directly into statistical machines
through the improved cash registers of department stores and shops. By automatically
processing this information according to given program, the machine itself will be
able to send out the necessary orders to the commercial bases for the delivery of
various other merchandise items from the trading points.
For example, the statistical machines of a main administration of the metal pro-
cessing industry willbe able to receive the complete information concerning the
progress of production. It will be possible to obtain information, for example,
about the following indicators of the operation of the enterprise: staff, number of
workers and their breakdown according to categories, production per worker, execution
of plan, production cost, electric energy used, number of orders and their character-
istics, raw materials, and parts in stock, etc. The study of these materials can be
conducted according to "different profiles."
With the existing techniques for processing these data in the accounting offices,
even a relatively narrow processing of records requires a large amount of time and
resources. If the data on each record are automatically recorded through the new
type of the ATS as a unit of information on the leaves of the machine memory, the
machine review of all the records of the enterprises of the combine will require sev-
eral seconds with a speed of about 30,000 units of information (leaves) per second.
The processing and the computation of the data according to any profile in
accordance with an indicator will only require several seconds.
93
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
STAT
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
?
?
The construction and use of such statistical machines will make it possible
to analyze the operation of enterprises from all angles through the most varied pro-
files.
Application of the Controlling Machines
The feasibility of reproduction, in accordance with a given address, of any kind
of curve makes it possible to propose a new system of program control of various
machines, machine tools, and other objects.
In the U.S. and in the USSR, work is being done on the automatic control of metal-
processing machine tools with the aid of perforated cards aid magnetic tape, on which
is recorded the program
speed and more reliable
and the magnetic tapes.
for the process of the machining of the product. Higher
static memory can be utilized instead of the perforated cards
The high-speed reproduction of the program for the
process permits the control of a large number of machines and machine tools
than 100) with the aid of a single memory installation.
The application of reliable contactless magnetic commutators, accumulators,
interpolators in conjunction with the new programming facilities will without a
doubt result in a control system which will be very productive and profitable.
operating
(more
Let us briefly recall that
operate automatically according
The program indicates the exact
and
in the case of program control the machine tools
to the schedule indicated by the tables of numbers.
trajectory of the instrument (cutter, knife), the
sequence of transitions from one kind of operation to another, as well as the control
signals for switching the operation off and on, the change in the number of revolutions
of the spindle, the supply of the cooling liquid, etc. The accuracy of numerical
programming is not limited: it depends practically only upon the accuracy of the mech-
anisms and the wear of the instrument.
This case of the use of contactless commutators and memory for program control
can be extended to the other types of program control of complicated equipment, with
a large number of elements linked together into one common system. Such equipment
can be found in chemical, metallurgical, oil, textile, and other industries.
94
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Application in Calculating Practice
The new memory devices can be successfully used in the analog computors and in
the universal digital computers.
Analog stands for the testing of the devices for automatic control and regulation
are becoming more end more important. It sufficies to cite examples of the use of
these models for testing the dynamic operation of autopilots and other types of
aircraft equipment. The most co4lidated flight and combat conditions can with suf-
ficient approximation be imitated in these analog stands.
Increasing importance is also being assumed by trainers, which are electrical
analogs used to train people for the control of moving objects.
These analog installations can assume great significance as a result of the new
type of memory equipment.
It can reproduce disturbances of various types, variable and constant coefficients,
tables of functions, and other data.
It is possible to introduce into the universal calculating machines one program
of operation after another, registered on static memory units. The gradual accumu-
lation of programs of different types in the machine, always ready for action, will
increase the scope of the machines.
It is possible to compile a large system of programs and subprograms, which in
various combinations will be able to perform the majority of problems encountered in
practice.
Figuratively speaking, these machines possess a "short" memory (good for one
problem) at the present time.
By using large-capacity, high-speed, and always-ready memory program units, the
efficiency of the machine will gradually grow.
The original data for the solution of different problems can also be registered
outside the machine on memory leaves. The units made with these leaves can then be
instantly put into operation at the same rate as the operative and arithmetical units
of the machine (at the sane speed).
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
95
STAT
Darf qnniti7Pd Coov Aooroved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13 CIA-RDP81-01043R004500020003-3
The above-mentioned examples of utilization can be realized already in the near
future. To confirm this statement let us examine the state of the corresponding
development in the Laboratory of Electrical Analog Computers of the Academy of Sciences
? VSSR. We will at the. same time point out also the data expected in the near future.
Durable Cappoitive-Type Memory Device
Figure 2 shows in a schematic form the electrical arrangement of a larse-capacity
unit whose construction is shown in Figure 3.
To simplify Figure 2, all elements located on one leaf of paper are shown on one
vertical line (T1-L1, T2-L2,..., T-I).
Figure 2. Schematic diagram of a durable capacity-type remory device.
C capacity; T - transformer windings of the address system.
Legend: 1) L1 4) screen
2) L2 5) Zoutput
3) Ln
One unit contains up to 1,000 leaves.
Each leaf has the electrodes of small condensers C with 3izia capacity. :he
recording of information is accomplished by leaving in or =itching off the circuit
of condensers C. To accomplish this, it is possible to omit the printing of separate
condenser electrodes on the designated places. It is also possible at first to pri StAT
?
96
in Dar!. qnniti7Pri (*CM ADDroved for Release @ 50-Yr 2014/05/13 CIA-RDP81-01043R004500020003-3
4
f.J
sheets with all connections and then, as the need arises, to punch out the connections
of the electrodes with the buses.
The method of punching out completely printed arrangements is used in the labor-
atory. When a large number of identical sheets is needed, it is reasonable to print
the sheets only with the necessary connections (without performations).
On the sheets L shown in Figure 2 are recorded the following codes of binary
numbers: L1-1l101; L2-10011; L3- 11011; 1401011;..., Ln-10111.
The presence of condenser C is shown here with the unit ("1"); the absence of
condenser C is shown with zero ("O"); the recording is done from top to bottom along
vertical buses (1.
The reading of the recorded information is done with the aid of voltage pulses,
induced in the secondary winding of transformers T1, T2,..., Ta(see Figure 2).
Figure 3. General view of a unit of the capacitive registering device.
When one of the transformers is excited, for example T3, the voltage will act
on the common electrode of the sheet L3. The capacitive current will pass through
condensers to the horizontal buses Bi, B2, 34,..., Bn. As we can see in the diagrams
there is no condenser C between bus L3 and bus B3; therefore, the capacitive current
will not flow directly from bus L3 to bus B3.
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
97
STAT
Dni-F rnpv Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Ap roved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Consequently, when T3 is pulsed, the capacitive current will flaw only to the
buses B1, B2, B4,..., Bn and will produce the corresponding voltage pulses across the
output resistances. If the distribution of these output pulses on buses B is re-
corded in the form of the binary number, the code 11011 is obtained.
The presence of a pulse is denoted here by the unit ("1"), while the absence of
the pulse is denoted by zero ("0").
In this manner, the action of one pulse, applied to a bus of one sheet, produces,
in parallel on the output buses, the pulses of the code previously registered by the
condensers C. The parallel reading of the code of the binary numbers from buses B
can be done in any sequence and as many times as necessary without disturbing the
record.
The secondary windings of T are placed on the ferromagnetic cores of the two
coordinate address grids.
The two nurribers of the coded address x and y produce the conditions of the ex-
citation of one of the transformers of the address grid. The voltage of the secondary
winding of the transformer acts on the buses of the sheet and the code of the binary
numbers registered on the sheet is reproduced on the output buses.
The sheets, when assembled into a unit, are subjected to a pressure of 50 to 100
kg/cm2. Under this pressure, electrodes A and B of the through I-oles form current-
-conducting zinc columns (the sheets consist of metallized zinc). The resistance of
such a column is very small.
Investigations were conducted to determine haw reliable the contact connections
were made in the zinc collars of the openings.
The resistance R falls with the incrense of pressure up to 2:0 kg/m2; when the
pressure P decreases from 200 to 120 kg/cm2, the resistance remei.-. -crstent. Further,
a decrease in pressure produces a grad,1 ircreFse in resis-Ancn .:e curve
(P) has the shape of a hysteresis curve. Tie curve H = f2(P1, wr.ere H is the thick-
ness of the unit, hes the same shape.
STAT
If, after rewoving the rre-sure, one tries to separate sue; a %r:t into :-cera
98
? r. ' r4
r-nv Acoroved for Release @ 50-Yr 2014/05/13 CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
sheets, one finds that the zinc contacts forming the column are so solidly bound to-
gether that one can only disassemola the unit by tearing nwny the paper. After
compression, the pressure is maintained with compression bolts.
At the end of the asnambly, the unit can be completely hermetically sealed.
The following experiment was conducted to test the reliability of the bonding
of the contacts of the individunl collars on t}-e sheets within a un-t: the compressed
unit was immersed in transformer oil for 2 montls. The resistance of the contact
columns was checked after 14, 30, and 60 days. All these measurements did not sham
any change in the magnitude of the resistance. This proves the reliability of the
binding of the contacts.
The cross-sections of the unit made after the compression showed that the com-
pressed metallized sheets formed a structure similar to the structure of the laminated
compressed isolating material.
A system of automatic contact soldering was developed by the laboratory for the
reliable connection of the electrical circuits of separate units.
It is possible to replace the condensers on the sheets by ferromagnetic cores
or by resistors in the circuit similar to that of Figure 2.
Circuits with printed resistors and inductors can also be used for the recording
and the reading of information.
To commutate the output signals of the condenser units, magnetic commutators
were developed and used, consisting of cores saturated with currents flowing in the
coils of the control network. The principle of operation of these magnetic commuta-
tors is very simple. The ability to vary the saturation of the reactive resistance
of reactors or of the coefficient of transformer ratio by means of the ampere turns
is used for the commutation of currents.
Thus, for example, a model of a "scanning" device was developed which connects,
in succession, the registers of the condenser memory equipment to the output circuits
of the bus B. The commutation is also performed according to the signals of the
binary code of some address system. STAT
99
neclassified in Part - Sanitized Copy Approved for Release 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
The technology of the preparation of the sheets determines the dimensions of
the electrodes and size of the units. In the products of the years 1953-1955, the
thickness of the paper was 125/a.
? The capacity of the condenser is inversely proportional to the thickness of the
sheet. If the thickness of the sheet is reduced 10 times, which is possible in the
near future, it will be possible to reduce 10 times the dimensions of the electrodes
C for the same value of capacitance.
If the size of the sheets remains the same, the dimensions of the unit reduces
10 times, and the capacity of the sheet in the binary notations is greatly increased
due to the reduction of the electrode dimensions.
The smaller the capacity of the elementary condenser, the smaller is the power
required from the elements of tie address grid.
The technology of electrode printing practically determines the magnitude of
the assumed capacity of the elementary condenser (3 pf). From the point of view of
actual practice, it is desirable to make the electrodes on the sheets and the capac-
ities as small as possible.
For rough calculations it is possible to use the following formula to determine
the power:
P
in Xc
whereis the reactive power of the input bus (one the electrode L of the sheet);
Pin
m is the number of the binary digits of the information element, which corresponds to
the number of condensers, connected to the electrode L of the sheet: XC is the re-
active resistance of the condenser; is the coefficient which takes into account
the screen capacity. In practice eC has a clue from 2 to 3.
With Uin = 100 v, Xc = 5.105 ohms (C = 3 pf), m = 500, and aC= 3, a power of
30 va is required. With a tenfold decrease in the capacity C, puller of only 3 va is
needed.
With the decreese in capacity, the magnitude of interference, arising from the
parasitic couplings between buses through these capacities, also decreases.
ved for
100
STAT
@ 50 Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Co .y Approved for Release ? 50-Yr 2014/05/13 CIA-RDP81-01043R004500020003-3
The action of this interference is roughly approximated by the formula
R U(4h) 2,1Cc
r TrRT mzou't
where U ("1") is the voltage of the signal "1"; U ("0") is the voltage of the signal
"0" (i.e., of the ; interference) 7
, -out is the impedance of the output circuit of the
saturated transformers of the address grid; le is the ratio of U (1) to the inter-
ference U (0).
As we can see, the larger Xe gets when we have constant values of m and Zout,
the smaller will be the influence of the interference, i.e., the larger will be the
signal in relation to the voltage of the interference.
Let us cite an example. Let m = 500; Zout = 50 ohms; and Xe = 5e105 ohms (with
C = 3 pf). Then
2.540'
P = 5.102-540 ? 40'
The signal "1" is 40 times larger than the signal
"O "
.
If the capacity is reduced 10 times, the reactance Xc will increase 10 times,
and this will result in the increase of /51 from 40 to 400.
Thus, it is possible to deduce that the main problem in the further development
consists in the devising of a better technology for the printing of the electrodes
on thin sheets of paper or on sheets on some other insulating material (film).
One of the facets of this problem is similar to the problem of the production of
radio-type condensers Which is being solved by the radio industry. The other facet of
this problem is the special case of the general problem of the development of the
technology for the manufacturing of printed electrical circuits.
During the last three years, the laboratory has acquired a certain experience
in the manufacturing of sheets, in the construction of the units, and in the commutat-
ing arrangements (address grids, decoders, etc.).
On the basis of this experience it is possible to quote approximate figures for
the general dimensions of the units for the case of arc metallization and vacuum
metallization. The approximate cost calculations show that, with vacuum metallization
of one coding sheet, the screen and lining Will cost about 10 kopeks (for 512 digits
103.
) .S TAT
?c? ? in r+ mr-1T ri r.rnv Aooroved for Release @ 50-Yr 2014/05/13 CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
The sheet for one million binary digits will consequently cost 200 rubles; corres-
pondingly, for 100 million digits the cost will be 20,000 rubles, and so on. With
massproduction of sheets and mountings, the cost naturally will be greatly reduced,
i.e., by about 5 to 10 times. However, the abovementioned figures show that in
many cases they can even now be economically acceptable for the construction of a
series of machines mentioned previously.
In 1955 the commission tested a model of a durable memory unit (unit "A") with
parallel reading of cards.
The purpose of testing the model was to determine its possible use in computing
equipment.
The model of unit "A" consisted of the following component parts.
(1) A unit, assembled from 256 cards. The capacity of each card Was 128 binary
digits, this constitutes one information.
(2) Magnetic scanning for 256 cards with a reading speed of 30,000 cards per
second.
(3) Output amplifiers (equal to the number of digits read simultaneously).
The testing of the model took a long time, more than one month. The stability
of unit operation(correctness of reading of code) was tested.
The code under examination (obtained through the selective method, for example)
was:
On the tenth output (during the scanning of the 256 cards):
010111 11111 111 1010101 10001 100011 110011 1100111 1001 10111001 101 11101101 11
11010101 10001111 111111 11 1111111 lionoll 11^^n11 10011001 101101 10001 1010101
1110101 111 10001 11011.
The correctness of the code being red -vas directly checked at the card output
and also after the output amplifiers.
The average voltage of the output signal of the card was 60 my, with the aver-
age voltage of the scanning signal being 20 v.
After the output amplifiers, the signals from the cards were strong enough for
the triggering of the standard magnetic storage cell of the register (i.e., not less
102
STAT
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
than 15-30 ma, 6-8 v).
The ratio of the maximum interference to the minimum signal while reading 256
cards WAS less than 15%.
The results of the tests of the unit of the durable memory arrangement with
parallel reading show that the unit developed in the laboratory makes it possible to
obtain more than 3 million pulses per second (for a card capacity of 128 binary digits).
The number of delivered pulses per second is determined by the operating speed of
scanning and the number of the parallel outputs (by the number of digits on the card).
It is important to note that the accepted speed of reading can even now be in-
creased 10 times, i.e., brought up to 300,000 informations (cards) per second.
In practice, the speed can be taken as 30,000, or 100,000 cards per second,
for the initial operation.
Magnetic Elements and Units
The laboratory developed magnetic storage units for signals, logical switches,
registers, triggers-, counters, commutators, as well as arithmetical devices (multi-
plying and adding devices, functional.converters, etc.). All these elements and cir-
cuits were tested in the system where circuits were interconnected according to a
predetermined scheme.
1. Magnetic Storage Unit for Signals
The storage device developed in 1952 consists of an arrangement which stores in
its cells electric signals -which are fed in succession (one by one) into the first
cell, The capacity of the storage is determined by the number of cells.
The speed of storing is determined by the speed of the cyclic pulses from the
internal source of current, which serves for the transfer of the accumulated informa-
tion from one row of elements to the other.
In contrast to the known systems of storage units, in which cores with rectangular
shape of the hysteresis loop are used, the operating principle of the developed stor-
age device permits the use of cores with a nonrectangular hysteresis loop, in par-
ticular, the use of the ferrocart cores. STAT
103
Declassified in Part - Sanitized Copy Approved for Release 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13 CIA-RDP81-01043R004500020003-3
.?
k
MO.
The transformers used in the storage device of 1952 have an outside diameter
of 7.2 mm, an internal diameter of 3.9 mm, and a torus height of 2.2 mm; the weight
of one core is 0.32 g. The windings of the transformer are made with a conductor 0.1
mm in diameter. The weight of one winding is 0.51 g. The selenium discs have a
diameter of 7.2 mm. The weight of the disc is 0.1 g.
The speed of storing during the tests was equal to 30,000 storage pulses per
second. The active power used by the storage devices in the circuit of cyclic pulses
with 2C0 cells was about 20 w. The total power was of the order of 26 w. The tests
showed the operating stability of the storage device when the voltage was increased
by +20%(over the minimum allowable value), which supplied the output tubes of the
generator of the cyclic pulse current. The tests for the checking of the stability
under vibration showed that the storage device is stable when subjected to vibrations
with a frequency from 20 to 40 c and accelerations from 1.5 to 10 g.
The storage device unit (containing 50 cells) continued to operate in a stable
manner while the ambient medium temperature varied within the limits of 4. 56 and -390 C.
The signal amplitude in the output cells changed under these variations by :2CP7. with
relation to the amplitude at a temperature of +20? C. The amplitude decreased as the
temperature rose and increased as the temperature fell.
At the present time, under laboratory conditions, the manufacturing of one cell
costs about 11 rubles. With further improvement in manufacturing technology, the
cost of one cell will be several rubles.
At the end of 1952 and in the beginning of 1953 the laboratory prepared and
issued 10 standard storage units (1C x 200 = 2,000 storage cells). Besides, from
1953 to 1955 the laboratory prepared more tl-an 10,000 storage cells made out of
ferrocart and ferrite with the rectangular hysteresis loop.
2. Logical Circuits (Switches)
On the basis of the storage relay cell made with hysteresis relays, logical
switches were developed which, in combination with the storage, permitted cells the
realization of complicated arithmetical and logical equipment.
104
STAT
Darf qnniti7Pd Coov Aooroved for Release @ 50-Yr 2014/05/13 CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
7
Generating Triad. The storage cell, which has a circuit feeding back from the
?
output element to the input of the first element and which possesses an additional
coding winding on the first element, is called the generating triad "T." When the
coding signal is applied to the first coding winding, a corresponding signal appears
on the output of the cell. If a signal "1" is supplied, this signal is again applied
to the input of the cell because of the presence of the feedback circuit. In this
manner, the unit signal will appear on the output of the cell with each cyclic pulse
of the current source. There results a dynamic memory (circulation) of the signal,
appearing at the output of the cell. The second coding winding serves for the "eras-
ing" of the registered signal. When the signal "1" is applied to this winding, a
compensation of the circulating signal takes place and output of the signal to the
external circuit stops. If the pulses at the output of the generating triad are
smoothed out and shaped, it is possible to obtain a.pulse of constant current.
Coincidence Circuit "AND-AND." The storage cell with two output windings, cum-
ulatively wound, is called coincidence circuit "AMD-AND" or switch "AND."
Only when two signals "1" appear together on the inputs of the coding windings
will the signal "1" appear on the output of this cell.
Comparison or Noncoincidence Circuit "OR-OR." The storage cell with two input
coding windings is called logical circuit of comparison or noncoincidence (switch "OR").
When the coded signal "1" is applied to one of the coding windings, a signal
"1" appears on the output of the cell.
When the coded signal "1" is applied simultaneously to both coding windings,
a signal "0" appears on the output of the cell.
Prohibiting Circuit "NO." The prohibiting circuit (switch "NO") consists of a
storage cell with an additional coding winding. When a coded signal is applied to
the first main coding winding, a corresponding code appears on the output of this cell.
The prohibiting circuit brings about a selective "erasing" of the signals "1"
in the code which passes through this cell.
3. Registers, Triggers, Counters, and Commutators
105
STAT
r'rrw Anoroved for Release@ 50-Yr 2014/05/13 ? CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
The register consists of the signal storage device having a feedback circuit
from ti-3 output of the last cell to the input of the first cell. The feedback cir-
cuit is realized through the switches of the circulation "AND" and "NO."
The information entering the register is "memorized" and is circulated within it
as long as it is required by the program. If it is necessary to issue the informa-
tion, the control pulse is applied to switch "AND." If it is necessary to "erase"
the recorded information or part of it, the control pulse is applied to the key "NO."
Trigger. The combination of the logical switches "AND" and "NO" with the stor-
age cells produces a triggering circuit from hysteresis relays.
Counters and Division Links. To count the pulses in the binary system, a cir-
cuit for a binary counter was developed.
The circuit uses switches "T" and "NO" and simple storage cells. The counter
consists of division cells with each link performing the function of the division by
two of the binary code applied to its input. All division cells are of the same type
and are interchangeable.
A series-connection of these cells permits the division of the coded signals
applied to the input of the first cell by 2n, where n is the number of the dividing
links.
The counter can serve for the control of any circuit according to the given
number of the pulse.
The manufactured binary counter permits division by2 15
For the counting of pulses in the decimal system, a circuit was developed for the
decimal counter with the transfer of the tens.
The decimal counter made with triads c'n be used as a commutator and as a dis-
tributor.
Magnetic Synchronous Commutator. A magnetic field rotating in space can be used
for changing the magnetic permeability of the ferromagnetic material, located in the
path of the magnetic field. Under these conditions a sequential variation of the
magnetic state of the elements takes place with time. STAT
106
1:?????,..,,,
?
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
STAT
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
?
.flerram????14,9?1, Ii5,412tWF-1 v?-0,-.???
*. 1
Investigations conducted in the laboratory have shown that such a magnetic
commutator can be used as a synchronous distributor, as well as for the phase mod-
ulation of current pulses, for the Production of pulses displaced in time, clad for
the conversion of codes, etc.
Multicoordinate Contactless Relay Grids. Systems with two coordinate grids
which contain magnetic nonlinear elemen:s permit the construction of high-speed
"tables of multiplication," "tables of addition," and other tables of various functions.
A larger number of coordinates can be used for other purposes.
4. Summation and Multiplication Devices
Arithmetical devices were developed which perform the functions of adding and
multiplying the binary numbers.
With the selected speed of admission of the cyclic pulses at the rate of 30,000
pulses per second, the summation device can perform the addition of two numbers of
the 32nd rank with a speed of 1,000 additions per second. The speed of addition is
determined by n cycles, where n is the rank of the numbers to be added.
The summation device consists of the logical "AND" "OR" and simple storage cells.
The typical unit of the summation device for the addition of the number of the 64th
rank contains 17 switches "AND" and "OR" and about 130 simple storage cells. It is
placed in one typical box (see 3 in Figure 6).
The laboratory has manufactured six summation devices, out of which three were
of the storage type (integrating).
The developed multiplying devices perform the multiplication of the binary numbers
with a speed of 2n cycles, where n is the number of ranks of the multipliers (the
multiplication of two numbers of the 32nd rank is performed with a speed of about 500
multiplications per second). For the multiplication of numbers, the principle of
consecutive parallel subranking addition is applied. For example, the multiplying
device for the multiplication of numbers of the 8th rank contains about 55 switches
of the type "AND" "OR" and "T," and 50 simple storage cells.
5. Device for the Comparison of Numbers, Converters, Decoders
107
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part- Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
The device for the setting of the coordinates in the three-dimensional region
and for the comparison of numbers ("UZK") adds consecutively the numbers
x = x1 + c1A x
Y = Yi + K2 AY
z = z1 + K3 Az
x, y, and z are the coordinates of a certain given region: Ax, LNy, and Az are their
increments, and k1,2,3 are whole numbers (1, 2,...64).
Simultaneously with the summation there takes place the comparison of the current
values of the coordinates x, y, and z with their boundary values.
The comparison principle means that the number which is compared is subtracted
from the one which is given. When the number which is compared becomes larger than
the given one, the summation process stops and the addition of the increment .6. to
other coordinates takes place.
This device makes ,it possible to change the coordinates of the region x, y, and
z by 1 within the limits of 0 to 127. It contains about 45 switches of different
types and 243 simple storage cells.
The nonlinear functional converter (table of sines and cosines; devices for the
selection from the tables) represents a device which permits the obtaining of the
values of sinoc and cos oC simultaneously for the given value of o.C.
The device Consists of a grid, assembled as a table of sines and cosines, end
of a decoder with logical switches.
The device for the selection from the tables consists of about 60 logical switches
and of about 200 simple storage cells.
The decoder, which converts the code from the binary system into the binary-
decimal system, consists of summation links, connected among themselves with registers
and switches. This device Converts the binary code of the 32nd rank into binary-
decimal, which is necessary for printing the output data in decimal numbers.
The decoder consists of 60 logical switches and of 400 storage cells.
6.- Eagnetic Operative Memory Device "MOZU"
108
STAT
im,,,ineeifiori in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
As we mentioned p7eviously, circulating registers made from simple storage cells
are used to memorize the information.
In these registers, the number of digits of the memorized information is de-
termined by the number of memory cells connected in series.
It is evident that to memorize an information containing a large number of digits,
a large number of storage cells is required. The laboratory has developed a new
magnetic operative memory device "MOW" which uses bipolar current pulses, produced
by the pulse current transformers, for the recording and the reading of information.
Figure 4 shows the schematic diagram of connections of windings (or buses)
which thread through the toroids of one plate of the memory device.
Figure 4. Schematic diagram of the
magnetic operative memory device
(MOZU).
Figure 5. The hysteresis loop and
the diagram of the ampere-turns of
the element of (MOZU).
The number of toroids mounted along a single bus Z is determined by the rank of
the information or of the number (Figure 4 shows 6 ranks).
The number of stacks of toroids in the X coordinate of the plate is determined
by the required capacity of the memory device. Figure 4 shows four stacks; conse- STAT
109 .
? Z ?
? r)
if i1 Aroroved for Release @ 50-Yr 2014/05/13 CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
quently, on this plate we can record end read four numbers, each of which can con-
tain six ranks of the binary system.
In practice it is possible to install along the stack Z, 20, 40, 80, and more
? numbers of toroids, and the number of stacks along the X coordinate can be taken
as equal, for example, to 16, 32, 64, and more.
To increase the capacity of the device, it is possible to introduce one more
coordinate "Y," i.e., to connect several plates into one system.
Each torus of the plate has three windings: X-(k) for the reading of signals,
X-(k) for the registering of signals, and Z for the setting up of the address of
reading and of recording, i.e., for the selection of one number out of many.
Figure 5 shows the hysteresis loop of the toroid (Figure 5a), as well as the
distribution of the ampere-turns of the signal currents for the.recording of unity
(Figure 5b), of zero (Figure 5c), and of the ampere-turns of the address system
(Figure 5d).
The pulses of current are passed into one of the stacks along the bus Z with the
aid of the pulse transformers of the address system.
At the end of the first pulse, all the toroids of a given stack will have nega-
tive values of induction ("0").
To register "1," it is necessary to supply a current pulse in accordance with
Figure 5b; to register zero it is necessary to supply a current pulse in accordance
with Figure 5c.
In this manner there are two pairs of current pulses (from the address system
and from the recording bus) acting in each cycle on the toroids of one stack. De-
pending upon whether the ampere-turns Of signal currents add or subtract, a record
of unity or zero is produced in the given toroid.
A two-coordinate address system is applied in "Y.OZU" similar to the address
system used in the units of the capacitive memory devices.
In June 1955 a model of the magnetic operative memory device was tested by
STAT
the commission and approved for production.
110
Onr+ Ca .t. ri ('cnv A ID o ed for Release @ 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
Model of a System Made of Magnetic Elements and Links
In order to test the feasibility of common operation within a system of arith-
metical and logical units, and of the units of the capacitive memory device, a model
of the system was developed in 1954 which permitted the combination for Common oper-
ation of a capacitive memory device with different arithmetical and logical units.
Each arithmetical unit, included in this system has its input and output leads,
connected through plug-in arrangements.
All incoming and outgoing leads of the units are connected to a common switch-
board, where they can be externally connected according to a given plan.
Figure 6 shows the general view of the model of the system.
The right stand contains the units of the capacitive memory equipment (1) and
the magnetic decoders of the address system (2).
The left stand contains the arithmetical and logical units, and the register
units (3).
Figure 6. General view of the model of the system consisting of the
capacitive memory devices, arithmetical units, registers,
and other links.
The center part contains an arrangement for the setting of the coordinates of
the region and for the comparison of numbers "Mr (4) and the control board with the
indicating unit (5). STAT
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
111
norincQifipri in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13 CIA-RDP81-01043R004500020003-3
?,
Each unit has an internal program of control and can operate independently
outside of the system.
The magnetic elements, links, and units were tested. The observations, conducted
for 10 8-hour days of uninterrupted work, showed that they operated in a stable
manner with the variations of the anode potential in the source of the cycle pulses
within the limits of 280 to 440 v (the upper value is limited by the source). The
supply originated at the source with a frequency of 30 kc.
The arithmetical and logical units (multiplying, adding, device for the setting
of the coordinates "UZK," device for the selection from the tables "VT," binary
counter of 15 cells with the division of the 30-kc frequency by 215 and the converter
of the binary numbers into the binary-decimal ones) were tested during joint oper-
ation in the system.
The results of observations during the three days of the uninterrupted (nonstop)
operation of the units showed their working stability.
As the drawbacks of the magnetic elements, one can mention their upper temper-
ature limit (50-70?), the necessity for the use of rectifiers, and the frequency
limitations.
The elements, links, and units of the hysteresis relays have been approved for
production.
The work on the ferrite elements is conducted in the direction of the frequency
increase (high-speed operation). The electrocomputing laboratory in 1955 conducted
research and investigation on all types of relay elements and storage and ?logical
switches for the frequency of 330,000 operations per second. The magnetic operative
memory device Was designed for the selection and recording of numbers in 8,4tsec.
The increase in the speed may within the nearest future reach 1 million oper-
ations per second. However, this is related to the growth of the minimum power value
of the relaying magnetic element. To decrease this threshold of power, we need new
technological developments to decrease the size of the cores and to introduce the
machining or printing of their windings. As current sources for the supply of con STAT
112
,Annroved tor
@ 50 Yr 2014/05/13? CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
tactless magnetic systems we can use electrical machinexy(motor-generators and
static converters of alternating current), tube rectifiers, and crystal (semiconductor)
current generators.
The principle and the most important property of the magnetic contactless systems
is their operating reliability and their long life.
Together with the large anount of -work being done on the crystal triodes here
in the USSR and in the U.S., work is being done on an increasing scale on the magnetic
relays and amplifiers. For example, the firm of Remington-Rand, Division of Sperry
Rand Corp., reports about a new, cheap, high-speed calculating machine without elec-
tronic tubes; built with miniture magnetic elements called "microferractors." Mass-
production of these machines is expected to start at the beginning of 1957. In the
U.S. an increasing number of publications* are dealing with magnetic elements. There
is a general opinion that where very great operating speed is not required (more than
300,000 operations per second), a magnetic contactless system must be used.
To insure the greater reliability of these systems, work is being done in the
USSR and in the U.S. on the construction of standard units which could be manufactured
by means of a printing process. Investigations are being carried out on the multi-
circuit magnetic network on one common core having printed windings.' The aim is to
create mechanically sturdy units with a very large number of elements. Mile, for
example, in the electron tube of calculating machines a separate plug-in unit is
represented by a tube trigger, we find even new in the experimental magnetic systems
within one plug-in unit more than 100 magnetic triggers; and within one unit of the
magnetic memory there are more than 3,000 elements. In the future it is possible to
expect the production of units with the number of elements larger by 1 or 2 orders.
Deductions
The reliability of operation, long service life, small size, processing, compar-
ative cheapness, and other qualities of contactless equipment lead us to expect that
STAT
* Detailed bibliography can be found in the article, Karnaugh Proc., JRE, May 1955,
page 570.
113
npriassified in Part - Sanitized COPY Approved for Release
50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Co .y Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
?
?
this type of equipment will in the future receive wide application:
(1) for the creation of the new type of ATS by replacing contact relays and
selectors, introduction of the operative memory and computing-solving equipment for
the selection of the optimum circuit connections in each individual case: it will
be possible to include in the ATS the telegraphic network for the transmission of
numerical material and text;
(2) for the construction of new high-speed automatic systems and information,
statistical, and translating machines which cannot be realized in practice with the
old means of automation; only because of the introduction of durable high-speed
memory with a very large capacity can we plan the design of these machines and their
inclusion in the general system of the future common service of information and
control;
(3) for the programming of automatic control of industrial objectives, where
we need a memory for the registering of the program, very sturdy and reliable cal-
culating-solving equipment, and commutators for the communication between the central
point of control and the numerous objectives;
(4) for automatic signalling and control of railroad traffic, of gas pipes and
oil pipes.
(5) for the remote-controlled systems of automation of atomic electric stations
and atomic power installations and other types of production with difficult and
harmful operating conditions for the personnel; .
(6) for the construction of computing devices for popular use (arithmometers,
tabulators, special accounting machines), digital computers for testing the equipment
for automatic control and regulation, and digital trainers for the instruction of
flyers.
114
STAT
rlarlaccifiPri in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
B. I. Verkhovskiy, G. G. Iordan, and N. N. Shumilovskiy
OUTLOOK FOR THE UTILIZATION OF RADIOACTIVE ISOTOPES
IN THE AUTOMATION OF INDUSTRIAL PROCESSES
One of the most important problems facing us in the field of industrial auto-
mation is the task of ouilding accurate and dependable devices for controlling param-
eters that must either remain unchanged during the production process or follow a
set pattern of change. The constantly increasing speed of technological processes
and high expectations of accuracy of measurements compel the designers of automatic
devices constantly to improve the old methods of control utile searching for new
methods. Some of these new methods employ the radiations of radioactive isotopes.
The experience gained in the past several years in designing and operating such de-
vices underscores the growing Importance of radioactive isotopes among the tools of
measurement technology. Apparatus equipped with radioactive devices is being used
very effectively both in laboratory
cesses. In many instances, both in
being used or
are slated to be used
work and in controlling various industrial pro-
our country and abroad, devices of this type are
as transducers in automatic systems for the
regulation of industrial processes.
The present report does not purport to present a detailed survey of the work
being done to apply radioactive isotopes in the field of automation. The paragraphs
that follow describe briefly the basic paths along which this applied work is being
developed, and the possibilities of their expansion. At the conclusion we have for-
mulated a number of scientific-methodological and organizational problems whose
solution is vital to our success in this field.
I. In evaluating the prospects of utilization of radioactive isotopes in the
field of automation, attention has been centered in the main on the application of
these elements in devices for automatic control and guidance of industrial processes.
Let us therefore consider, first, some advantages common to the metering methods
based on the application of radioactive isotopes.
(1) A characteristic feature of such metering techniques is the noncontact
115
STAT
Darf qnniti7Pd Coov Aooroved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
nature of control and, as a corollary of this, the absence of a reverse reaction
of the device on the measured parameter. It thus becomes possible to exercise con-
trol without interference in the industrial process and without changing the proper-
ties of the object of measurement. On the other hand, such procedures protect the
?
device from any possible damage, and in many cases make possible the exercise of
control when the application of other methods proves impossible.
(2) As is well known, a change in external conditions, such as temperature,
pressure, humidity, etc., affects neither the action of the preparation nor the energy
of the radiation it emits. This fact serves to differentiate-radically such radio-
active sources from all other components used in building the mechanisms. The high
stability of the radioactive flux permits the carrying out of precise measuremehts
under the most difficult industrial conditions. Moreover, this characteristic of
the radioactive isotopes permits their use as gauging or calibrating devices and a
material lessening, through the setting up of various differential and balancing
systems, of errors introduced into the final metered data by changes in the parameters
of the other components of the device.
(3) Devices and mechanisms that make use of radioactive isotopes also manifest
a low degree of inertia. This fact is very important in the control and automatic
regulation of technological processes proceeding at high speeds.
(4) The long period of half-life of some radioactive isotopes (sane of these
periods last years or decades) opens up prospects of building devices with very long
terms of service.
(5) In the case of numerous parameters, such as, for example, thickness, density,
humidity, flaw, level altitude, etc., devices of the same type may be used in the
most diverse branches of industry. It thus becomes possible to devise standard
measuring systems and devices, or at least the basic assemblies for such devices,
that are capable of serving industry as a whole; this fact, naturally, facilitates
the use of such mechanisms in automatic regulating systems.
(6) In devices based on the use of radioactive isotopes, changes of the cont-,2
AT
116
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
STAT
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
led magnitude are converted directly into an electric signal. This fact, likewise,
serves to simplify materially the problem of combining such devices with systems of
automatic regulation and with computing mechanisms.
?
II. The use of radioactive isotopes in metering technology is developing along
the basic paths of the utilization of the penetrating capacity of the ray end of
the measurement of the degree of its absorption by the substance, of the utilization
of the phenomenon of the dispersion of radiation, as well as the process of ionization
of the substance as a result of the radioactive emission. Considerable interest
also centers on the utilization of neutron sources. As is well known, neutrons are
never emitted directly by radioactive isotopes, but can be readily produced with
their aid.
(1) Measurement of the degree of absorption of a radioactive emission by a
substance supplies the basic principle in the operation of a large number of devices.
Such devices are used effectively in metallurgy for measuring the thickness of the
rolled metals, inthe paper industry for measuring the thickness of paper, and in the
building industry for measuring the mud density in mudpumping machines, etc. The
penetrating capacity of the radioactive ray is utilized in a number of instruments
used to measure the level altitudes of liquid and free-flowing dry substances in
closed reservoirs, in devices used for signalling the reaching of a predetermined
point by the level, and in other signalling devices.
Further refinement of the degree of accuracy in measuring the thickness in each
specific instance (depending on the chemical compositioh and the density of the
material in question) is achieved by the use of an isotope that emits a ray with a
given anount of energy. A substantial broadening of the possible avenues of appli-
cation of such devices may be attained by the utilization of a "soft" urns. ray (with
energy on the order of 100 key) as well as of a braking and characteristic x-ray
radiation resulting from the absorption of fast beta particles. Somewhat further
down on the agenda is the possible utilization also of the "hard" gamma rays with
energy of 6 to 8 Mev that appear in certain nuclear reactions.
117
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Cop Approved for Release ? 50-Yr 2014/05/13 CIA-RDP81-01043R004500020003-3
Measurement of the extent of radioactive emissions is used, as is well known,
in the detection of flaws in products and materials. The development of flaw-detect-
int methods not based on the photographic recording of radiation is a most urgent
problem. Such methods may be used as a basis for evolving systems of complete auto-
matic control of manufactured products under conditions of mass production. The so-
lution of this problem depends on the further refinement of methods of fast and
accurate measurement characterized by highly effective recording of the flaw of
radioactive emissions, and of gamma rays in particular.
The degree of absorption of a "soft" gamma ray bears a very close relation to
the atomic number of the absorber. This fact provides the basis for developing a
whole series of devices to be used in determining and controlling the chemical
composition of various materials used in technology. The degree of absorption of
a "hard" gamma ray (with an energy of from 0.5 Mev to 2 to 3 Mev) and of a beta
ray depends on the number of electrons per unit of weight of the absorber. For this
reason, the absorption of such radiation is sensitive to the hydrogen content of the
absorber; this fact can be used in building devices for measuring the moisture con-
tent or, for example, the ratio of C to H in hydrocarbons.
In the case of signalling devices, a radiation detector is generally used as a
part of a relaying system, i.e., under conditions when, at one position of the measured
object, the magnitude of the recorded flaw is virtually zero, while at another pos-
ition of the same object it has the maximal value. In such cases, an accurate
measurement of the radioactive flow is not required.
It should be noted at this point that the development of methods permitting an
accurate metering of the measured parameter without calling at the same time for
high accuracy in the measurement of the radioactive flaw (whether frequency, phase,
based on utilizing modulated radioactive emissions, etc.) holds great promise. Such
methods will enable us to simplify materially the apparatus and instruments presently
in use, and in a number of cases to attain a degree of accuracy beyond our reach
when using other methods.
118
ir Dmrl. - Caniti7Pd CODV Approved for Release @ 50-Yr 2014/05/13 CIA-RDP81-01043R004500020003-3
STA1
Declassified in Part - Sanitized Co .y Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
The Geiger-Mueller counter--a simple, yet highly sensitive device--is used most
frequently as a radiation receiver in signalling systems. However, two factors tend
to limit their use--the large amount of dead time and their brief lifetime. The
lessening of the dead time and an increase in the length of service and in the sta-
bility of the Geiger-Mueller counters will bring about a marked increase of their
application in controlling and measuring devices. This, in turn, will improve the
prospects of a wide adoption of such devices by the most diverse branches of industry.
(2) The intensity of dispersion, or, as it is often called, of the reflection
of a radioactive ray, depends on the thickness, chemical composition, and density
of the reflector. This fact is utilized in devices used for determining the thickness
of electro-deposited coatings and the composition of alloys and mixtures.
The measurement of the thickness of electro-deposited coatings by means of
radioactive isotopes is often the only possible method of measurement that does not
result in the destruction of the controlled coating. The field of possible appli-
cations of this method is extremely large and is not limited, inter alia, by the
presence or absence of certain electric or magnetic properties in the measured
material. The methodological potentialities of measuring the thickness of electro-
deposited coatings by means of radioactive isotopes are far from being exhausted.
A severely limiting factor in the application of this method in many branches of in-
dustry is the fact that, with the aid of devices presently in use, the determination
of the coating's thickness must be made over a wide area. Development of devices
permitting the measurement of the thickness of an electro-deposited coating over a
small area and possessing a low degree of sensitivity to changes in the spatial
relationships between the device and the object to be measured will permit a mani-
fold increase in the application of radioactive isotopes in this field.
In measuring thicknesses, whenever the measured object may be reached from only
one direction, we may utilize the well-known interdependence of the intensity of
dispersed radiation and of the thickness of the object. In devices of this type, con-
siderable difficulties are presented by the need for protecting the radiation re-
119
STAT
rlarlaccifiPri in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13 CIA-RDP81-01043R004500020003-3
ceiver from the effect of the radiation flow coming directly from the source; this
problem is particularly acute in the case of gamma rays.. The use of lead collimators
and of protective screens to eliminate this difficulty calls for an increase in the
intensity of the radioactive source, leads to an increase in the size and weight of
the device mad very often precludes the attainment of the required accuracy of
measurement, particularly in the case of small areas. It is possible, however, with-
out making use of protective screens, to record dispersed radiation only, by using
the methods of gamma-spectroscopy, since the energy of gamma rays, when dispersed
backward, differs markedly from the energy of gamma rays emitted directly by the
source. This method permits an increase in the accuracy of measurement as well as
a material decrease in the required intensity of the radiation sources and in the
weight and size of the instrument. Further refinement of such devices must await
imiorovements in gamma-spectroscopic methods, above all, in scintillation methods.
It is possible, by the use of a set of radiation sources with differing energies, to
determine simultaneously the content of the separate components of multicomponent
systens. The building of automatic devices of this type will require, in all likeli-
hood, the enlisting of the methods and resources of computing technology.
(3) Metering technology utilizes the ionizing capacity of radioactive emissions
very extensively. The functional dependence of the degree of ionization on the dens-
ity and composition of the gas provides the basis for the operation of devices for
the measuring of pressures as well as for determining the contents of certain ad-
mixtures in gases. The increase in the range of measured densities and the construc-
tion of gas analyzers capable of detecting the various admixtures will make possible
a substantial expansion of the field of apElication of such devices for the purposes
of industrial automation.
The phenomenon of the ionization of a gas by a radioactive ray is also utilized
in devices that measure the speeds of gas fluxes and the flaw of gas. Refinement
of the temporal, frequency, and phase methods of measuring these parameters (methods
that are based on the use of modulated radiation and do not depend on a precise SI-AT
120
in Dar+ qnniti7Pri rODV ADDroved for Release @ 50-Yr 2014/05/13 CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
measurement of the degree of ionization) will permit an increase in the range of
the measured velocities, particularly of low and supersonic velocities, as well as
improve the accuracy of measurement; the possible fields of application of these
devices will be correspondingly broadened.
Measurement of the spectral composition and of the intensity of luminescence
caused by a radioactive emission may be used in deteraining chemical compositions,
as, for example, in determining the degree of purity of certain organic combinations.
(4) We shall consider next the prospects of utilization of neutron flows in
metering technology and in automation. It should be noted at this point that the
scale and volume of the work presently done in this field is utterly inadequate.
Yet, the areas of possible application of devices incorporating neutron sources are
extremely broad. Just as in the case of the use of radioactive emissions for de-
termining the various properties of a substance, we can make use in this field, too,
of the functionalrelationships that mark the processes of absorption mad dispersion
of neutrons by various media. Furthermore, the so-called activational analysis repre-
sents another distinct field that utilizes the neutron flows and will, no doubt,
contribute significantly to the automation of industrial processes.
Effective retardation of fast neutrons by hydrogenous media provides the physical
basis for instruments used to measure the moisture content of various materials.
Many branches of industry are interested in such instruments; among these may be
mentioned ferrous and nonferrous metallurgy, coal mining, the building industry, and
other leading branches of the national economy. The construction and wide adoption
of such instruments and devices would be greatly aided by the further refinement of
industrial nethods of measuring the flow of fast and thermal neutrons.
The' measurementof the degree of absorption of neutrons by various media may be
used to determine the content of elements with a large cross-section for absorbing
neutrons, such as, e.g., boron, cadmium, and a number of others.
Neutron radiations may also be used in flaw-detection of products with large
thicknesses when the penetrating capacity of gamma rays proves inadequate, and in
121
STAT
v Anoroved tor
Rel
@ 50-Yr 2014/05/13 CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
certain other cases.
Neutron sources may be utilized in devices used for measuring the levels of
hydrogenous media. A great merit of such devices is the fact that they may be used
in cases where the object to be measured can only be reached from one direction.
As the methods and the apparatus of neutron spectroscopy are improved and re-
fined and made ready to meet the exacting standards set for industrial devices, neutron
sources will probably also find application in the field of temperature measurement.
Neutron sources are already being used in the petroleum industry as well as in
other geologic exploration. With further development of methods of neutron and
neutron-Eamma core-sampling, the application of these sources in geologic exploration
work will, no doubt, steadily increase.
The methods of activations? analysis, i.e., the measurement of the content
of various elements by an analysis of radioactivity resulting from their exposure
to neutron rays, are of considerable value and interest to people concerned with
automation of the processing of ores. It is well known that many chemical analyses--
indispensable in the work of controlling metallurgical processes--require consider-
able time for their completion; this is true even in cases where high-speed methods
are used. A reduction of this time could be achieved in many cases by the use of
methods of activational analysis and of radioactive indicators, and would benefit
the industry greatly. Should these methods be improved in the reasonably near future
to the point where it would become possible to exercise automatic control (directly
in the industrial process) over the content of at least some elements, the automation
of metallurgical processes will be raterially advanced.
To achieve success in this field, it 1..!11 be necessary to improve still further
the present methods of recordinT radioactive rays, of the measurement of law intens-
ities, and of the analysis of the composition of rays emitted by a mixture of sev-
eral radioactive isotopes. In the work of automation, and especially in the deter-
mination of the chemical composition and of the measurement data of the individual
STAT
elements, interest attaches not only to nuclear reactions that result in the form-
122
v Annroved tor
Rel
@ 50-Yr 2014/05/13 CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
ation of radioactive isotopes, but to certain other nuclear reactions as well. Among
such reactions may be included the reaction (nvy.) regardless of whether the result-
ing isotope is radioactive or stable. Determination of content of the individual
elements in this case may be made by analyzing the specter of the gamma ray emitted
at the moment of the entrapment of neutrons. This method will, in many instances,
require considerably lower densities of neutron flows, since the gamma ray is emitted
within an infinitely small interval of time after the entrapment of the neutron. To
get a gamma ray flow of equal intensity by the use of methods of activational analysis
(these methods may be used if the resulting isotope isradioactive), will require
either a considerable span of time--equal approximately to three half-lives of the
formed isotope--or considerably stronger neutron flows. The main difficulty encounter-
ed in using this method of measurement stems from the fact that the emitted gamma
rays generally show a complex spectral composition, and the energy of this radiation
by the various elements lies, as a rule, in one part of the specter, the only points
of difference being the varying intensities of the separate lines.
Both the methods of activational analysis and the methods based on nuclear
reactions are closely linked with the measurement and analysis of complex radioactive
emissions. For this reason, the building of devices for automatic control and es-
pecially of transmitter devices for automatic regulation systems will involve, also
in this case, the enlisting of the resources of computing technology.
The use of computing apparatus will also make possible the creation of a whole
series of devices for simultaneous automatic measurement of several different param-
eters and will increase the potentialities of nechanisms used to measure one param
eter at a time. The building of such devices will permit a substantial broadening
of the field of application of radioactive isotopes in the automation of industrial
processes. The use of radioactive isotopes in determining the composition of multi-
component mixtures has been referred to previously. Examples of other possible appli-
cations of this method of measuring several parameters at the same time are the
following. The simultaneous determination of the moisture content of metallurgical Si-AT
123
npnlassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
pulp by means of neutrons and the measurement of the degree of absorption of a soft
gamma ray will enable us to define both the content of the hard component in the
pulp and the content of elements with high atomic numbers in this hard component (ore).
The measurement of the degree of absorption of soft and hard gamma rays by coal per-
mits the determination of its ash content. At the same time, changes--inevitable in
the process of production--in the thickness of the radiation-absorbing layers of coal
will introduce no error into the measurement data. In a device using only one source
of gamma rays, such changes would introduce errors far in excess of allowed toler-
ances.
The method of using radioactive indicators for chemical analysis, may, in add-
ition to the aforementioned possible application, find other, though limited, fields
of application in the control and regulation of industrial processes. Limitations
of the method stem mainly from the undesirability of contaminating manufactured pro-
ducts and factory space by any substantial amounts of radioactive substances or
wastes. However, improvements in the methods of measuring lam intensities, and the
use of radioactive isotopes with a trief half-life will, in all likelihood, permit
in some cases the application of radioactive indicators for the purposes of automation;
at present this method is used widely in research work. This method may be used, for
instance, in measuring the velocity of flow and flux of various technological. mix-
tures and solutions. A radioactive isotope may be either introduced into the system
from the outside or be obtained as a result of the exposure of some segment of a
mixture or a solution to neutron rays. For measuring velocities, it is most con-
venient to build separate "packets," each containing a radioactive isotope. Such
packets may be built, for instance, by the ,'plication of a time-modulated flux of
neutrons. The measurement of velocity and flux may then be accomplished by counting
the number of "packets" that pass the radiation detector in a given period of time.
In our examination of the prospects of application of radioactive isotopes in
the field of automation, we have limited ourselves up to this point to a review of
their use in controlling and measuring operations. However, the work of automationSTAT
124
oved tor
@ 50-Yr 2014/05/13? CIA-RDP81-01043R004500020003-3
STA
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
er)
??-?
very often calls for modifications and improvements of technological processes. In
this field, too, radioactive isotopes present considerable interest and definite
promise. One of such possible areas of application of radioactive isotopes is their
use for the removal of electrostatic charges. Application, of radioactive isotopes
for this purpose often presents the simplest and at times the only possible solution
of the problem, and in many cases permits a considerable increase of the speed of
technological processes.
Another possible field of application of radioactive isotopes in the area
defined above is the use of their radiation for increasing the speed of certain
chemical processes and for changing the properties of certain materials, e.g., semi-
conductors. This application of radioactive rays constitutes a large, independent
field by itself, and it would not be feasible to go into it in great detail in the
present report. However, the fact that this field of application of radioactive iso-
topes is of interest and value to the cause of automation should not be lost sight of.
III. Let us now consider some questions of scientific and methodological char-
acter whose solution, in our opinion, is of general interest to all areas of appli-
cation of radioactive isotopes in controlling and measuring devices and in systems
of automatic regulation.
(1) The problem of increasing the accuracy of data on the measured parameters
obtained by the use of devices incorporating radioactive isotopes is linked, as a
rule, with the need for increasing the accuracy of measurement of the flux of radio-
active rays. This task encounters considerable difficultiec, particularly in cases
involving time-consuming measurements under industrial conditions. For this reason,
parallel to the development of methods that do not require accurate measurement of
the radiation flux, there must be a continual refinement of methods of protracted,
continuous, and precise measurements of the fluxes of radioactive and neutron rad-
iations. And in the process, special consideration daould be given to methods of
measurement characterized by high efficiency in recording, particularly to the
methods of scintillation.
125
Declassified in Part - Sanitized Copy Approved for Release 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
STAT
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
?
(2) Development of spectroscopic methods for radioactive and neutron radiations.
(3) Development of methods of complex simultaneous measurement of a group of
parameters. Application of computing devices in the instruments.
(4) Improvement of radiation receivers, particularly the development and prep-
aration of scintillating plastic materials and of liquid scintillators, as well as
of crystal scintillators of large dimensions; improvement of photoelectronic mul-
tipliers and the development of new types of this device with a higher stability of
parameters; lengthening of the lifetime of Geiger-Mueller counters.
(5) Increase of the lifetime of devices to a period comparable with a half-
life period of isotopes with a long lifetime (about a year or longer) and an increase
of their reliability. This task will require the development and application, in
the devices, of measuring systems built of semiconductor elements and an investigation
of the possibilities of constructing radiation receivers of such elements.
(6) An enlargement of the scale of application of devices incorporating sources
of radioactive rays and the securing of safe conditions of work both for the personnel
servicing the instrument and for personnel tending the technical equipment on which
the instrument is mounted; these will require the adoption of special measures and
the drafting of appropriate regulations. Let us note in passing that the present
Pravila raboty s radioaktirnymi izotopami /instructions for Working with Radioactive
Isotopes7 make no reference whatsoever to possibilities of working with such in-
struments.
(7) As the application of instruments and devices
isotopes grows in scale, the problems of decreasing the
incorporating radioactive
intensity of the emitter
in each model of the instrument will come '--reasingly to the fore; the solution
of these problems is important because it will help to lager the costs; even more
important is the effect it will have in insuring safe conditions of work. The natural
lowest limit of the required intensity of the emitter is represented by the level at
which the accuracy of measurement is determined entirely by the statistical fluctua-
tions of the recorded flux of radiation. The use of such instruments in systems of
126
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
automatic regulation requires a theoretical and experimental examination of the
questions of construction and behavior of such systems in the presence of a fluct-
uating input signal. Problems bearing on the accuracy of measurement and on the
selection of optional parameters in connection with the development of measuring
methods and the construction of instruments based on the application of radioactive
isotopes likewise call for theoretical analysis.
IV. In conclusion, we deem it essential to consider in detail some organization-
al questions linked with the application of radioactive isotopes in the field of
automat ion.
(1) The nature of problems facing us in this field requires for their solution
the services of specialists of the most varied types--physicists, instrument design-
ers, and specialists in automation. At the present time, parallel to the work of a
number of other organizations in effectively applying radioactive isotopes in instru-
ment design and automation, such work is also being done by the Laboratory of Isotopes
and Radiation of the Academy of Sciences USSR; by the Laboratory of the Institute
of Automation and Telemechanics, Academy of Sciences USSR; and by the laboratory of
the NIITyeplopribor of the MPSA USSR. Incorporation of these laboratories into the
complex of the above institutes can and should insure the carrying out of research
and studies on a high scientific and methodological level, with the utilization of
the most modern physical methods and means of instrument-designing technology and
automation.
While reserving comment on the other shortcomings in the work of these labora-
tories, it is proper to point out that the scale and the tempo of research being done
there can under no circumstances be termed satisfactory. The cause lies mainly in
the poor and at times utterly inadequate supplying of these laboratories with
scientific manpower, industrial quarters, and proper equipment. Without trying to
anticipate the measures that should be taken to remedy the present situation, we
feel that the end result of such measures should be a considerable increase in the
capacity of the above laboratories and their conversion into basic and leading or-
127
STAT I
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
ganizations in this field. The Academy of Sciences USSR should also take steps to
establish and strengthen laboratories attached to the academies of sciences in the
various republics.
(2) Fully congnizant of the fact that a number of problems linked with the
application of radioactive isotopes is peculiar to each specific branch of industry
and that the successful solution of these problems requires the closest possible
liaison with the appropriate institutes and enterprises in the various branches of
industry, we recommend the establishment of new laboratories as well as the expansion
of the existing laboratories in this field; these laboratories are to be a part of
the institutes engaged in the work of instrument design and automation for the basic
branches of industry. It seems to us, in the meantime, that the Academy of Sciences
should recommend an end to the present practice of the atomizing of resources by
the creation of a large number of miniature laboratories; it should advise instead
the concentration of effort in a comparatively small number of adequately powered
laboratories capable of conducting research on a high level and at the proper tempo.
(3) Under existing conditions, when the work of applying radioactive isotopes
in the fields of instrument design and automation is being done by a large number of
organizations, it is vital to establish an appropriate scientific and coordinating
center. Such a center could be set up in one of the branches of the Administration
for the Utilization of Atomic Energy of the Council of Ministers USSR. It could
be tentatively called the "Commission for the Application of Radioactive Isotopes in
Instrument Design and Automation." Needless to say, every effort should be made to
induce the leading specialists engaged directly in research in this field to join the
commission's staff.
High on the agenda of such a commission would be the following tasks.
A study of the situation existing with regard to the application of radioactive
isotopes in the field of automation; it should be noted that the number of instru-
ments and installations now in use is considerably smaller than the number of instru-
ments prepared by various organizations; a wide adoption of fully-evolved methods
FIAT
128
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13 CIA-RDP81-01043R004500020003-3
?
devices could have a considerable effect on the national economy.
The drafting of recommendations for a longterm plan of scientific research in
the field of application of radioactive isotopes in automation. In addition to the
permanent members of the commission, specialists in the branches of industry most
directly concerned with the development and dissemination of these methods could be
drawn into this work.
(4) Considering the fact that a numoer of devices incorporating radioactive
isotopes is of value to the industry generally and can be used without material changes
in various branches of industry, the Academy of Sciences would be well advised to
recommend the establishment, within the apparatus of the MPSA USSR, of a special
design bureau and of a pilot plant for the design and manufacture of such instruments;
such a stepvmuld tend to hasten the adoption of the newly-developed methods and in-
struments by the national economy.
(5) The Academy of Sciences should recommend appropriate measures for widening
the range and improving the qualitative characteristics of radioactive isotopes
produced at present. The need for sources of soft gamma rays and the necessity of
increasing the specific intensity of the supplied preparations should be particularly
borne in mind. It would be advisable, likewise, to consider the question of lowering
prices for radioactive isotopes, especially for isotopes that may be used as sources
of braking and characteristic x-rays, e.g., Strontium-90 isotopes whose present price
precludes their use for these purposes.
Finally, a solution must be found for the problem of supplying interested organ-
izations with radioactive isotopes made to fit their devices. The present situation,
in which a multitude of organizations is engaged in preparation of such sources, is
reducing this important activity to the level of a primitive handicraft. Technical
shortcomings inevitable under such circumstances may entail dangerous contamination
with radioactive isotopes of both laboratory and industrial buildings where such de-
vices are being used.
(6) The Academy of Sciences must also raise the question of the need of manu-
in Dad. - Aniti7ed Coov Aoproved for Release
129
?
50-Yr 2014/05/13 CIA-RDP81-01043R004500020003-3
STA1
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
facturing special equipment for equipping laboratories engaged in reaearch in the
field of application of radioactive isotopes. Furthermore, the problem of designing
standardized laboratories must also be resolved.
(7) The wide adoption of instruments featuring radioactive isotopes requires
even at this comparatively earl:- stage the creation of an authoritative body with
powers to police the manufacture and use of such devices in the interets of safety
and security. The establishment of such a body will help to prevent accidents and
will also help to dispel unfounded fears that arise occasionally upon the installation
of such devices.
(8) And last but not least we have the question of the training of "cadres" or
scientific and professional manpower. The Academy of Sciences should, in our opinion,
take the initiative and recommend that the institutes, which at present graduate
instrument-designing engineers, also train specialists in the fields of development,
design, and operation of instruments with radioactive isotopes, or that at the very
least the appropriate training courses be included in the programs of these institutes.
It would likewise be advisable to include such courses in the programs of instru-
ment-design technicums.
To insure competent and skilled handling of installed instruments, the ministries
concerned should undertake the proper training of engineers and technicians at the
plants and enterprises where instruments and devices based on the application of radio-
active isotopes have been installed or are slated to be installed.
Statement
by Engineer Ye. A. Nekhayevskiy
The All-Union Scientific and Research 'nstitute of the State Bureau of Engraving
in the past two years, thanks mainly to the efforts of G. G. Iordan and B. I.
Verkhovskiy, has developed, built, and tested a device for determining the weight
of an area unit of a paper tape; this device will be used in automatic production
controls of papermaking machinery.
The device is called a v.eiEhtmeter and measures the weight of a moving tape STAT
130
narlaccifiPri in Part - Sanitized Com Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
weighing from 5 to 25 mg/cm2. The weightmeter consists of a measuring pulley with
a differential system of ionization chambers with two thallium sources of 10
milli-
curies, and a specially designed dynamic condenser for transforming the measured
ionized current into an alternating one; the latter is amplified in an electronic
unit with an outlet to a standard recording device of an EPD type. Altogether, there
are three stabilizing cascades or stages.
It has been established that the error in measurement does not exceed 2.5%; error
in time measurement does not exceed 2 sec; the device is also characterized by high
stability--no drift of the initial control is observed in a period of 24 hours.
This device has passed industrial end laboratory tests and is now being in-
stalled in an industrial enterprise for regular use.
The following are the distinguishing features of the above mechanism: linearity
of the scale secured by a dividing device equipped with a compensating absorber; the
scale divisions are uniform throughout because of the use of the dividing device;
the measured data are completely independent of the shifts of the absorber within the
specified tolerance; the effect of temperature fluctuations without the use of the
compensator does not exceed 0.5% per 10C; a capacity to accommodate simultaneously
Up to five indicators with an accuracy of 2:5 mg/cm2 of the specified value; the
recording takes place in absolute units over the entire measuring range (up to 25
mg/cm2).
A measuring pulley controls the moving tace with a width of up to 800 mm in its
movement over a "path" with a width of 250 mm and can be set automatically over the
tape within a given time after a tear or break of the tape has been repaired; the
pulley can likewise be withdrawn automatically at the moment of the tape's break or
tear. Remote control of the measuring pulley is now under consideration.
Plans are afoot to combine the cescribed mechanism in the capacity of a trans-
ducer rtransmitter7 with a regulating device. The use of an automatic weight regu-
lator will permit the lowering of substandard production to 5%. The resulting econ-
omy will amount to about 1 million rubles per year per each papermaking machine
131
STAT
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
?
4
in the Bureau of Engraving.
And now please permit me to say a few words in connection with the report by
B. I. Verkhovskiy and others. The present method can lay no claim to freedom from
effects of outside media and outside conditions on the results of measurements.
The greatest trouble is caused by the effects of temperature changes. The low degree
of inertia is a relative concept, since it is difficult to achieve a time constant
of 1 sec.
On the questions of organization and methodology I would like to suggest the
following: an immediate effort should be made to draft rules and regulations on
the use of the instruments?the sanitary inspectors of the district will not permit
us to use the completed instrument for a test run under industrial conditions because
of the absence of such regulations.
A valuable suggestion has teen made in the report with regard to the preparation
of sources. There is no need to build expensive laboratories for the making of
sources in each separate enterprise. Such sources are not prepared very often and
are quite expensive as a result.
We support the proposal concerning the establishment of a coordinating scientific
center for the application of radioactive isotopes; I would only add to the suggestion
a list of organizations engaged in the design of instruments similar to the one
I have just described; the Central Laboratory of Automation (measurement of the
thickness of metallic tapes), the Central Scientific Research Institute of the
Cotton Industry (measurement of the extra weight of substances coated or added to
fabrics). LETI in Leningrad is designing a similar instrument for work on paper for
the "Kommunar" paper factory. Nonferrous re7R1lurgy, the textile industry, and the
paper and building-materials industries are all active in this field.
3.32
STAT
narlaccifiPri in Part - Sanitized Com Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
STAT
???
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
RADIO-FREQUENCY SPECTROSCOPY AND THE POSSIBILITY OF
ITS USE IN THE FIELD OF ANALYSIS
by
A. M. Prokhorov
Radiospectroscopic methods are in wide use at the present time in physics and
chemistry for the study of properties of nuclei, atoms, molecules, fluids and
solids. But radiospectroscopy also has its practical uses. For example, the
measurement of magnetic fields (with accuracy up to 10-6 oersteds) and their sta-
bilization by the use of nuclear and electronic paramagnetic resonances is becoming
increasingly widespread. A highly significant application of radiospectroscopy is
represented by the utilization of spectral absorption lines for setting up standards
of frequency (time). A large number of studies are concentrating at present on
this problem.
Finally, the question arises as to the possibility of using radiospectroscopic
methods for the purpose of analysis. However, in using such methods, a number of
difficulties are encountered. For a clearer understanding of the use of radiospec-
troscopy for analytical purposes and the corollary use of radiospectroscopic meth-
ods for process control, it will be necessary to make a brief survey of radiospec-
troscopic methods and to go over the conditions that favor the appearance of ab-
sorption spectra for substances and materials we are interested in studying.
Radiospectroscopic methods fall into the following four categories:
1) Gas radiospectroscopy which is primarily concerned with rotational transi-
tions.
2) Paramagnetic electronic resonance. This method is primarily concerned
with transitions between energy levels resulting from the interaction of paramag-
netic ions with an external magnetic field.
3) Nuclear paramagnetic resonance. In this method, observation centers on
energy ,level transitions that result from the interaction of the magnetic moment of
the nucleus with an external magnetic field.
133
npnlassified in Part - Sanitized Copy Approved for Release
50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
STAT
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
4) Pure quadrupole transitions; the phenomenona observed here are energy
level transitions resulting from the interaction of the electric quadrupole moment
of the nucleus with the electric field formed by charges arranged around the
nucleus.
Let us examine each of these methods in closer detail.
1) Rotational molecular spectra are observed in centimeter and millimeter
wave bands. The study of these spectra is effected by means of radiospectroscopy.
Substances under study are generally in a gaseous (vapor) state, contained in an
absorption cell with a total volume of about 1 liter, under pressure corresponding
to 10-2 - 10-3 millimeters of mercury. Under such pressure, the collisions of the
gas molecules are rather infrequent and the width of the line is obtained at about
1 megacycle; i.e., considerably below the value of electronic rearrangement of the
klystron, which is of the order of 40 - 60 megacycles. Lessening of pressure leads
to the lessening of the maximal volume of absorption without a narrowing of the
line, since the width of the line under conditions of low pressure is determined
in the main by the Doppler effect and by the collision of the molecule with the
walls of the vessel.
Increase of pressure leads to the widening of the lines but the maximal volume
of absorption does not change in the process. In order for the gas to have an ob-
servable absorption capacity it is necessary that its molecules possess a dipole
moment. The greater the dipole moment, the larger the volume of absorption. Should
we deal with an average molecule (in terms of geometry and molecular weight) with
average dipole moment, we may be able to detect 0.1 - 1% of concentration of such
molecules in a gas mixture. Moreover, the rossibilities of isotopic analysis should
not be lost sight of. Suppose we are given a mixture of isotopes and it is neces-
sary to determine their percentage-wise content; the problem may be solved by the
preparation of a suitable mixture for an analysis based on the fact that the lines
of absorption of molecules with differing isotopes have differing frequencies.
2) Observation of an electronic paramagnetic resonance takes place mainly
134
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
in waves within the centimeter range. Free radicals and paramagnetic ions with an
odd number of electrons have their lines of absorption on wave 3.2 centi-
meters, with the volume of tension of the magnetic field H = 3,300 oersteds. It
should be emphasized that both solutions and solids may be used for the observation
of a paramagnetic resonance. In many cases, it is best to use monocrystals rather
than polycrystalline substances in the observation of a paramagnetic resonance.
Moreover, to take the typical case of the paramagnetic resonance of rare earth ions,
measurements must be made at the temperature of liquid hydrogen; i.e., at 200 of
the Kelvin scale. Worth noting is the fact that when gadolinium ions are present
in a trivalent state, while the europium ions are bivalent, the paramagnetic
resonance of these rare earth ions may be observed in polycrystalline substances at
room temperature. Some idea of the effectiveness of this method may be gained from
the fact that it may be used to detect an admixture of gadolinium when its concen-
tration is as law as 10-2 -
In order to observe an electronic paramagnetic resonance, the substance (with
a volume 1 CM3) is placed in a volumetric resonator at the point of the maximum
of a high-frequency magnetic field. Next, a constant magnetic field is impressed
over the observed specimen in such a manner that the high-frequency and constant
fields are perpendicular to each other.
To observe the absorption line, it is necessary to modulate the constant mag-
netic field rather than the frequency of the klystron; in other words, use is made
of the modulation of the magnetic field.
3) Nuclear paramagnetic resonance may be observed in the case of nuclei whose
spin I is not equal to zero. However, should the spin I of the nucleus be ))L the
nucleus will be found to possess a quadruple moment.- The presence of this moment
in a nucleus brings about a rather strong interaction with the surrounding media;
this, in turn, results in a pronounced widening of the lines, and may, in some in-
stances, preclude the observation of a nuclear paramagnetic resonance. For this
reason, conditions for observation are most favorable for the paramagnetic resonancri.ma
135
Declassified in Part - Sanitized Copy Approved for Release 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
STAT
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
of nuclei whose spin 1= IT, and of nuclei with a small quadrupole moment. We
should bear in mind that the volume of the signal will be the stronger the greater
the magnitude of the nucleus' magnetic moment.
? Observations are generally made at frequencies on the order of tens of mega-
cycles with the magnetic fields measured in thousands of oersteds.
Men the strength of the outside magnetic field is indicated, the resonance
frequency of a given nucleus changes very little in transitions from one combination
to another. This fact provides us with a simple and effective method of detecting
the various nuclei in a combination that is being analyzed. This method is used to
conduct isotopic analysis. The effectiveness of the method may be illustrated by
the following example. The volume of signals of deuterium nuclei in natural water
exceeds the volume of noise tenfold. As is well known, the dispersion of deuterium
is about 0.02%.
Methods of observing the nuclear resonance do not differ materially from the
methods of electronic paramagnetic resonance; the only difference is that at the
usual radio frequencies use is made not of volumetric resonators but of self-induc-
tion and capacitory coils.
Specimens for analysis may be used in either liquid or solid states. In some
cases, it may be necessary to add some paramagnetic salts to the analyzed substances
so as to shorten the relaxation time of the nuclei, and thus to assure an adequate
observation of the nuclear paramagnetic resonance.
4) Pure quadrupole transitions are observed in a wide range of frequencies -
from tens to hundreds of mecracycles. Observations are made, as a rule, without the
use of a magnetic field.
The frequency of the resonance lines cepends both on the magnitude of the
quadrupole moment of the nuclei and on the magnitude of the second derivative of
the electrostatic potential caused by the surrounding charges at the locus of the
nucleus. The magnitude of the electrostatic potential's second derivative depends
closely on the type of the
cnemical combination; for this reason the location of the
136
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13 CIA-RDP81-01043R004500020003-3
resonance frequency of a given nucleus will depend on the type of chemical combina-
tions involved. The value of the ratio signal/noise will be rather small here.
The method of observation is comparatively simple - one observes the absorption line
at the given frequency of the generator.
It may be seen from this brief survey that radiospectrosoopic methods may be
used to solve only certain concrete analytical problems; accordingly, each specific
case should be studied carefully before we decide to apply spectroscopic methods.
137
STAT
in Dar!. qnniti7Pri (*CM ADDroved for Release @ 50-Yr 2014/05/13 CIA-RDP81-01043R004500020003-3
STAT
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
?
L. D. Rozenberg
ACOUSTICAL nETHODS FOR MEASURING NONACOUSTICAL QUANTITIES
General Considerations
Modern measuring practice is characterized by the expansion of its technical
foundations. Alongside with the direct methods of measurement of quantities, indirect
methods are appearing; at the basis of the latter we find the measurement, not of
quantities directly in question, but of some factors associated with them. Often
these methods are more convenient than the direct ones because of the better avail-
ability of the measured factor, better sensitivity, and, what is particularly im-
portant, the possibility of an automatic measuring process.
Among others, methods based on variations of the parameters of a sound field,
especially produced at the object of measurements, are beginning to be used now, In
spite of the fact that there are only a few disconnected and unrelated acoustical
methods of measurement of nonacoustical quantities, the technique as a whole appar-
ently is quite promising and merits a further systematic investigation.
Because of their mechanical nature, elastic (sound) vibrations and waves appear
to be the most adequate method for the investigation of the material objects and of the
variations of their mechanical properties. The parameters of the sound waves, which
are subject to variations, are the frequency of oscillations, amplitude, phase, time of
propagation (for traveling wave), and the traveling wave ratio (for standing wave).
The knowledge of these parameters permits the determination of the speed of pro-
pagation of sound waves in the medium under investigation, their attenuation in the
process of propagation, the coefficient of reflection from the boundary of two media,.
the coefficient of dispersion in a nonhomogeneous medium, and some other quantities.
The speed of sound, in its turn, is directly determined by the density of the
medium and the corresponding modulus of elasticity or its inverse quantity, the
compressibility.
Speed of propagation of sound waves
138
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Co .y Ap roved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Gas: A y
Liquid: A=
Solid body
longitudinal waves:
inside a bar Am.E,
in.a continuous medium A -
Shearing waves: A = G,
P. - density of the medium
A - modulus of elasticity
- adiabatic constant
130 - constant pressure
A- compressibility
E - Young's Modulus;
E(1 -4f)
(1 - 2d) (1 + d )
Poisson's coefficient
G - shear modulus.
In this way, the knowledge of the speed of sound propagation makes possible the
determination of the elastic constants of the medium with a known density or, in-
versely, the density of the medium with the knowledge of the constants of elasticity.
The coefficient of absorption of the sound waves is determined by the viscous
and the relaxation losses with the sound propagating in homogeneous media, and also
of the coefficient of sound dispersion in the case of nonhamogeneous media. Sound
absorption is expressed by comparatively complicated formulas, which are not exactly
known for all cases. However, the variations of the coefficient of absorption makes
it possible in each concrete case to follow directly the variations of the properties
of the medium, for example, viscosity.
In some cases it may be helpful to conduct the measurements at different frequen-
cies. The coefficient of absorption is almost always dependent upon the frequan
:MT
but the character of this dependency is different for different substances. Thus,
139
Declassified in Part - Sanitized Co .y Ap?roved for Release ? 50-Yr 2014/05/13 : CIA-RDP81-01043R004500070nm21
STAT
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
?
by conducting measurements on several frequencies, it is possible to obtain the nec-
essary data for the solution of a system of equations encompassing, for example, the
losses in the multicomponent mixtures, etc.
The natural frequencies of bodies are determined by their geometrical dimensions
and by the speed of propagation of sound waxes. The simplest example is presented
by the Helmoltz resonator for which the following equation is correct.
Wo= C ,
where: C is speed of sound propagation; S is the cross-section of throat; t is the
length of throat; and V is the volume. Thus, when the speed of sound propagation is
known, it is possible to use the variation of its frequencies to determine the geo-
metrical dimensions of bodies. This, for instance, is being used in the resonance
thickness gauge and cavity meter. The other method for the measurement of the geo-
metric dimensions of a body consists in measuring the time required.for the propagation
of a sound pulse whose spatial length is considerably mailer than the dimension to
be determined. This method, known as the echo method, is widely used in the 'hydro-
acousties, in supersonic fault location, and for the measurement of liquid levels.
The acoustical methods may be divided into four main groups, depending upon the
object of measurements: measurement of geometrical dimensions and volumes of bodies;
measurements of parameters of a medium, observation of the state of a medium; location
of internal defects and of nonhomogeneity of structure. The first group includes
the thickness gauges, altimeters, level gauges, cavity meters, etc. The methods of
the second group embrace the measurement of density, of all moduli of elasticity of
volumetric and shearing viscosity (viscosir.etry). The methods of the third group
allow the control of the progress of the chemical reactions and of the physicochemical
processes, as for example, the variation in the state of the aggregate, the hardening
of concrete, such processes as crystallization, polymerization, etc. Here we can
also mention the measurements of variations in the speed of flaw of liquids and gases
(consumption meters). The last group includes all types of supersonic fault locations
and underwater location.
140
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
Depending upon the method of measurements, one can distinguish the pulse method
and the stationary method. In its turn, the pulse method can be divided into two
groups. In the methods of the first group, the transit time of the pulse or the
change of its amplitude is measured. If we know the speed of propagation, the transit
time permits the determination of the length of the path of the traveling pulse (echo
meters, liquid level gauges) and if we know the distance, we can determine the speed
of propagation (consumption meters, measurement of elastic constants). The amplitude
variation of the pulse determines the aosorption in the process of propagation.
The schematic arrangement of such measurements is shown in Figure 1.
Let the medium to be investigated
be located in some container (if it is a
liquid or a gas). With the aid of
1.1 emitter I, a supersonic pulse is intro-
duced into the medium. This pulse, at
the end of its travel, is received at
the opposite end by receiver P. The out-
put voltage is fed from the receiver
amplifier to electron beam indicator SI.
Simultaneously, the same indicator re-
ceives the original electrical pulse
produced by generator IG which supplies
the emitter. On its war, this control
pulse passes through an artificial line
IL which permits the regulation of its
amplitude and introduces s.--;me time delay.
In this manner, it is possible to ooserve simultaneously on the screen of one electron-
beam indicator the fixed control pulse together with the examined pulse which appears
after a short time. The variation in the speed of propagation or of attenuation
E'T3
???????10
2 Asa..
Figure 1. Diagram of the pulse method
for measuring substance properties.
Legend: 1 - increase in speed; 2 -
decrease in speed; 3 - original pos-
ition; 4 - increase in at,;enuation;
5 - decrease in attenuation; 6 - gen-
erator IG; 7 - emitter I; 8 - receiver
P; 9 - amplifier U; 10 - artificial
line IL; 11 - beam indipator 31.
SI-AT
during Propagation produces t're corresponding change in the observed picture. Figure
141
narlaccifiPri in Part - Sanitized Com Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13 CIA-RDP81-01043R004500020003-3
1 shows the simplest case. This principle is employed, for example, in the cavitation
gauge developed by the Polytechnic Institute of Kharkov and used in hydraulic tur?
bines; this equipment is being demonstrated at the All?Union Industrial Exhibit.
In the methods of the second group, the pulse is used as a means for the ex?
citation of natural oscillations of a system. By investigating the process of atten?
uation of these oscillations, it is possible to measure, for example, the losses in
this system. By analogy with the architectural acoustics, this method is also called
the reverberation method. On this principle are based some viscosimeters and also
devices for the determination of the internal friction in metals, plastics, rubber,
etc.
The methods which use stationary sound waves can also be divided into two
groups: those using traveling waves and those using standing waves. The first group
of methods allows the determination of properties of a medium in which an elastic wave
is propagating. The methods of the second group are more universal. Because the
standing wave is the result of reflection of a traveling wave from the boundary be?
tween two media or from the boundary of an obstruction, the investigation of the par?
ameters of this standing wave permits the establishment of the properties of the
medium Which lies on the other side of the boundary or of the properties of the ob?
struction, as well as the distance to the reflecting boundary or obstacle.
In the vast majority of cases, the Primary sound field is produced externally;
however, there are also cases where the field is produced by the examined process
itself. In this sense, all acoustical methods are, as a rule, active, i.e., they
require the presence of a source (emitter) of sound oscillations. The measurement
of the sound field is accomplished with the -id of the sound receiver. In some cases
the functions of the emitter and the receiver can be combined in one converter; in
this way, particularly, is constructed almost all echo method equipment. At times
? the receiver is entirely absent: the results are judged by the reaction of the medium
on the emitter. This reaction is measured at the electric side of the emitter. For
example, when in the medium of finite length, connected to the sensing element of aSTAT
142
'?' ' D,r+ Cart.t. (nrv Aooroved for Release _@ 50-Yr 2014/05/13 CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Ap roved for Release ? 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
resonance thickness gauge, there is packed a whole number of half-waves, the loading
of the emitter changes sharply and this reacts upon the magnitude of the current
which supplies it. The change in tie current strength is detected by means of a
visual indicator. Sometimes, in order to produce the picture of the standing waves,
a special reflector is used; such a method is used in the interferometers.
The selection of the working frequency can be determined by various considerations.
In the resonance methods (cavity meters, resonance thickness gauges), the frequency
is selected on the basis of the resonance frequency of the system to be measured. In
the pulse methods, the selection of the frequency is determined on the one side by
such considerations as these: with the increase of frequency, the resolving power
and the accuracy of the method increase, while, on the other hand, we have the fact
that, due to the increase in absorption, the received signal decreases in magnitude.
In those cases where the magnitude which interests us is the absorption of sound
during its propagation, the working frequency must be selected sufficiently large
so that the absorption constitutes an appreciable magnitude.
As converters, the electromechanical systems have the monopoly; we have here
electrodynamic, magnetostrictive, and piezoelectric converters. Hydrodynamic and
aerodynamic emitters have not found acceptance in spite of the fact that they have a
better efficiency and that tie supersonic energy produced by them is considerably
cheaper. This is explained by the fact that, on one hand, great power is usually
not required for measurements, and, on the other hand, the aerodynamic and hydro-
dynamic emitters are considerably less stable in their frequency, amplitude, and
phase and are difficult to handle. Besides the electromechanical converters are
reversible; as we mentioned earlier, this proPerty is used in many echo methods and
also in the consumption meters.
The selection of the converter type is determined by the frequency range. Mth
a frequency range from 0 to 10-15 kc, electrodynamic converters are the most con-
venient; for frequencies of 5-150 kc, it is better to use the magnetostriction type,
and finally, the range of 100-15,000 kc is handled by the piezoelectric converters.
np.HARRifipn in Part - Sanitized Copy Approved for Release
143
50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
STAT
STAT
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Higher frequencies have not as yet been used; however, even here the only
possible converters are likewise the piezoelectric ones.
In the piezoelectric emitters, quartz is used in the majority of cases; it'
possesses a high mechanical and electrical strength and has a small temperature co-
efficient. The ceramics of titanate of barium, which have recently become so popular
for technological supersonic applications, are relatively rarely used for controlling
and measuring purposes, in spite of the very great magnitude of the piezoelectric
modulus. The point is that, along with their virtues, the ceramics of titanate of
barium possess great shortcomings which restrict their application for measuring
purposes: their piezoelectric modulus end specific inductive capacitance change con-
siderably with temperature, and, when heatod to 1200 C, a complete depolarization
takes place and the ceramics irreversibly lose their piezoelectric properties. Be-
cause of this, they are used only in those cases where the operating temperature
does not exceed 70-800 C and when the absolute magnitude of the emitted and received
signal does not have a tangible value. Examples of such an application are the
probes of supersonic fault locators or the resonance thickness gauges.
Description of Certain Existing Equipment
The resonance thickness gauges are manufactured in the USSR by the Ministry of
Aircraft Production and abroad by various firms in the U.S.,West Germany, France,
Belgium, Italy, and others. On the sheet whose thickness is to be measured, we placed
a piezoelectric converter which is supplied with a voltage of continuously variable
frequency. At the moment when the thickness of the sheet is equal to the even number
of half-waves, the reaction on tlaa convertr.r chances sharply, and this acts on an
indicator. Figure 2 shows one of these instruments. Four ranges of frequencies per-
mit the measurement of thickness within the limits of 1 to 100 mm with an error of
2-35. The advanta:e of acoustical thickness gauges consists in that it is not nec-
essary to reach the measured part from two sides; this makes it possible to measure
from the outside the thickness of pipe walls, of tanks, of closed volumes, and of
144
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
STAT
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
parts with complicated configuration eg.
Figure 2. External view of a resonance frequency meter.
Viscosimeter. At the present time, we have several different viscosimeters
which operate in the region of sound, as well as of supersonic frequencies. Let us
examine in detail one of them, manufactured by the firm "Ultraviscoson" if. The
1.1fmnyetacet inpainehw
oftmgai*Nutm
2 llowynaem roanow
Illutomm
3 Kontfauf ild,fla
1 1 1
1 I 1
I 1 I
1 1
To* I queAmoti qeiz
? ---11/911 6
PM),
thwoflaa 7
Figure 3. Principle of operation of a viscosimeter,
1 pulses for the control of excitation; 2 - pulses for the control of reception;
3 - oscillation of the probe; 4 - current in the receiving circuit; 5 - time;
6 - probe; 7 - winding.
145
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Ap roved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
operating principle of this viscosimeter is based on the fact that the attenuation
'of the natural oscillations of an elastic plate immersed in a liquid depends upon
the viscosity of the latter. The magnetostriction plate (Figure 3) oscillating in
the direction parallel to its surface, is excited to its natural frequency (20-100 kc)
by the regularly repeated square pulses. At the end of the pulse action, the os-
cillations of the plate attenuate. This attenuation process induces within a winding
an attenuating electrical pulse. A series of consecutive attenuating pulses is inte-
grated with a large time constant. The variable quantity is the frequency of the
sequence of the exciting pulses; it changes automatically so that the average current,
produced by the attenuating oscillations of the plate, remains constant. The cal-
culations show that the necessary sequence frequency of pulses is proportional to
the product of the density of the liquid under investigation and its kinetic viscosity.
Figure 4 shows the external view of such a viscosimeter. The magnetostriction probe
is seen in front.
According to the company's information, the equipment is good for operation with-
in the temperature range of -1200 C to +3400 C. The limits of measurements are from
Figure 4. External view of the viscosimeter; the probe is in front.
0 to 5 x 104 centipause x g/amS. The minimum amount of liquid required for measure-
ment is 2-4 cm3; the errcr of neesurement is 2-5%, the repetition error is 1%. The
sensing element can be separated from the indicator by a distance of 1.5 to 2 km.
Since the amplitude of oscillation of the device does not exceed 0.5A, it practically
has no influence on the medium being measured. The time required for one measurement
npHassifieri in Part - Sanitized Copy Approved for Release
146
50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
STAT
STAT
??
is about 1 sec. Later, an installation was described with 16 channels zu and
its external view is shown in _Figure 5. It permits the observation of 16 points
simultaneously with a recording of all results on one paper chart.
Figure 6 shows the arrangement of a sensing element of a viscosimeter, operating
at a considerably lower frequency of about 800 c /17. The active element is here
the disc clearly seen in the figure, oscillating in the plane parallel to its surface.
The coil of the exciting electromagnet can also be seen; it is supplied with a current
of variable frequency. Figure 7 daaws the calibration curve of this viscosimeter.
The calibration was conducted in a liquid with a known standard viscosity. We can
see that the scale of the device is practically linear within a very large variation
of magnitudes.
The cavity meter, an instrument for the acoustical measurement of cavity volume,
is described by Poole J77. It is shown in Figure 8. The depression P to be measured
(the diagram shows an example of measurement of the volume of the depression on the
Figure 5. The indicator unit of the Figure 6. Sensing element of the law-
viscosimeter with 16 channels.
frequency viscosimeter.
top of the block of an internal combustion engine) is closed with a cover Kp having
an opening 0. The sound waves from the sound source G can, through this opening,
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
147
norincQifipri in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
reach the cavity being measured. Another opening connects the cavity with the micro-
phone M.
The output of the microphone is connected to the input of an amplifier U and
the output of the amplifier feeds a loudspeaker. The stages are so selected that the
system is excited to the frequency which is determined by the natural resonance of
the acoustical resonazor which is formed by the cavity, the cover, and the opening
0. The frequency of self-excitation is compared, by means of comparator K, with the
frequency obtained from an opening of standard dimensions. The amount of frequency
deviation represents the amount of deviation of the volume from the required one.
7.7
+
?
? ? ? ? ? ? ? 4 . ? ? .
1" ?1?--"
. ? 1 -4
I 1
?
I
'I .1 I
M 4-
1il=a111:
MEMMUMUMEM
111=11111111111.11111
11v II, 1Juil 1.7v SI ii',,, 1.7v64
to 700 WOO CDC
qI7
? Figure 7. Calibration curve of a viscosimeter.
Legend: 1) divisions on the scale of the instrument.
The author shows that, in order to measure the volumes encountered in the ordinary
construction of machines, a frequency range of 50-500 c is sufficient.
Legend:
ilf
Figure 8. Diagram of operation of a cavity meter.
1 - from the gauge; 2 - U; 3 - K; 4 - M; 5 - G; 6 - 0; 7 - Kr; e-P.
148
STAT
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
STAT
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
Consumption meter. The principle of a supersonic consumption meter has been
described many times in literature. It is based on the fact that the sound speed
within a medium is equal to c, while the speed of flow of this medium is v; the time
of travel of the supersonic pulse from the stationary source to the stationary re-
ceiver will be 4-- + v if the direction of propagation of the sound wave coin-
(I
c
cides with the direction of flow of the liquid or gas, and t2 in the case it
c ? v
is directly opposite; d is the distance between the source and the receiver. From
Figure 9. Sensing element of the NITTeplopribor consumption meter.
1 - plug; 2 - connecting pipes; 3 - piezoelectric plates of the vibrators;
4 - body of the vibrator.
the two obtained equations the simple values of the speed of motion of the medium,
v, and the speed of propagation of sound are determined ,57. Such consumption meters
were developed and are manufactured by many companies. Figure 9 shows the sensing
elements of the consumption meter developed at NITTeplopribor Z7177. Figure 10 shows
the cross-section of the emitter-receiver, manufactured by one of the foreign firms
for the measurement of large consumption of water. Figure 11 thaws the photograph of
its installation within the spiral of a hydraulic turbine with a power of 42,500 hp.
In this last case the cross-section of the pipe is 5 x 7.5 in, the largest consumption
of water is 56 m3/sec, the distance between the sensing elements is 9 in, and the
operating frequency is 25 kc Arjr. The average error of the consumption meter of
this type is about 2%.
Analyzer of impurities. Since the acoustical methods permit indirect measure-
ment of different impurities through the change of sound propagation or its absorption,
they are, strictly speaking, not well adapted for the analysis in the full meaning of
149
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
this word, since different impurities can produce similar effects. Because of this,
acoustical methods are used eith3r in those cases when the expected composition of
impurities is known beforehand, or when it is only necessary to establish the fact
of the presence of impurities regardless of their composition or quantity. As an
example, one can cite an acoustical analyzer designed to monitor the appearance of
harmful gases in t!,e -dr, such as chlorine, carbon monoxide, or methane. This device
is based on the change in the speed of sound Propagation in the presence of impurities.
The measurement Itself is made
al view in Figure 12. Between
with a small interferometer, shown in the cross-section-
tie cuartz emitting plate q and the reflector R, a
system of standing waves forms. By depressing the pushbutton K, the reflector is
brought in contact with the surface of the emitting plate. When the pressure is
released, an external spring F returns
the reflector to the original position.
If, between the surface of the quartz
and the surface of the reflector, n half-
raves can be accommodated, then, when the
reflector moves, the reaction on the
emitter will change n times, i.e., the
magnitude of current in the circuit of
the emitter will change n times. The
resultant electric pulses are fed into a
pulse counter. The presence of impur-
ities change the speed of sound propagation,
?hange the length of the wave and as a
result the number n of the half-waves and
correspondingly of the pulses which reach
the counter. This permits not only signal-
ing of the pollution of the gas, but also
indicates the degree of this pollution FAT
4
iiove:e
4?7:67a-24..,
Figure 10. Antenna-type sensing
element.
Legend: 1 - crystals; 2 - linings;
3 - rubber packing; 4 - probe.
150
in Da rf qnniti7PO CODV Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Co Approved for Release ? 50-Yr 2014/05/13 CIA-RDP81-01043R004500020003-3
? The devices described above are far from representing the only possible methods
of acoustical measurements. In addition, the history of the appearance and of the
development of devices described above compels one to state that the appearance of
precisely such devices in the first place does not result from some general con-
siderations, but has often been determined by accidental causes and local require-
ments. As yet the application of acoustical methods is very small, for the control
of chemical reactions and physicochemical processes, where these methods can intro-
duce large improvements.
Figure 11. Installation of a sensing
element in the spiral of a powerful
hydraulic turbine.
Figure 12. Construction of a high-
speed interferometer.
Only very recently have acoustical methods begun to be applied here and abroad,
for the control of quality and for the supervision of the hardening of cement.
It is necessary to initiate a large amount of technical development and con-
struction efforts for the rapid growth of prospective methods of acoustical measure-
ment, considering in particular the circumstance that these methods are characterized
by high sensitivity and short time necessary for the performance of measurements;
they permit a renote location of sensors and can be successfully used for automation
and control.
151
STAT
1-1
for RI e@ 50-Yr 2014/05/13 CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Co .y Approved for Release ? 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
BIBLIOGRAPHY
1. See, for example, Bergman, L., U1itre.zyukffupersonics7, translation from
German, Moscow, 1956.
2. :Rueter, T.; Bolt, R., Sonics, New York, 1956.
3. Rev. Sci. Inst., 24, No. 2, Feb. 1953, page 186.
4. Woodward, J., Journ. Acoust. Soc. Am. 25, No. 1, 1953, page 147.
5. Poole, F., Product Eng., Jan. 1949, page 93.
6. Hess, W. B.; Swengel, R. C.; V;aldorf, S. K., Electr. Eng., 69, N 11, Nov. 1950,
page 983.
7. Shafranovskaya, Z. M., Priborostroyeniye ffnstrument-bui1dini7, No. 4, 1956,
page 12.
8. Swengel, R. C., Inst. Had. Eng. Cony. Rec. National Cony., Part 9, 1955, page 33.
9. Bergmann, L., Akustishche Beibefte, No. 2, 1954, page 591.
152
STAT
rlarlaccifiPri in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13 CIA-RDP81-01043R004500020003-3
V. L. Taltroze
MASS-SPECTROMETER AS AN INSTRUMENT FOR THE CONTROL OF
INDUSTRIAL PROCESSES
? Mass-spectrometer and the Principle of Mass-spectrometer
Analysis of Mixtures
Mass-spectrometer analysis?used during the early stages of its development in
a vast majority of cases to determine the nasses of atoms and the isotope composition
of elements--became during the last 15-20 years an efficient means for the investiga-
tion and determination of the chemical compositions of mixtures.
The modern mass-spectrometer represents an electric vacuum apparatus consisting
of the following major parts: ion emitter, designed to convert the molecules of the
substance to be analyzed into ions; analyzer, in which these ions by some means are
grouped according to their masses (more exactly, according to the ratio of the mass
to the charge); receiver, in which the ions reach a collector. The ion current is
measured by electrical means.
In the case of the analysis of gases or of more or less easily vaporized sub-
stances, with -which this article will be mainly concerned, a bombardment of molecules
with electrons possessing an energy of several tens of ev is used in an ion source
in order to obtain a beam of ions. Usually in this case, energies are used which
result in maximum cross-sections of single charge ionization (40-100 ev). (Additional
analytical potentialities can sometimes be obtained in determining the ionizing po-
tentials or working near the ionizing potentials.)
The application of the mass-spectrometer for the analysis of mixtures of chem-
ically different substances is based on the fact that, with constant energy and elec-
tron current, the ratio of ion currents of different materials is proportional to the
ratio of concentration of these materials in the analyzed mixture. This coefficient
of proportionality can be found by means of calibration. Besides, under a molecular
type of supply of gas into the ion source, the ion currents of the components of the
mixture are proportional to their partial pressures in the mixture. Under the elec-
153
STAT
nna-Inecifiarl in Part - Sanitized CODV Approved for Release @ 50-Yr 2014/05/13 CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
tronic bombardment of molecules by electrons with energies usually employed in mass-
spectrometer analysis their dissociative ionization takes place, in addition to
the simple ionization of molecules, with the result that each kind of molecule gives
rise to ions with different masses; for example, the molecules C2 H6 produce ions
C2 H6+, C2 H64-,C2 H4+,...3 C2+, H+. This produces amass-spectrum with many
lines. It is possible to assume that the spectra of different materials superimpose
themselves upon each other additively. Thus, in the general cases, to find the com-
position of the mixture with components, one has to solve, depending upon the spec-
trum, a system of k linear equations*:
. .
UI 1.\1 al?. v2 -f- ? ? ? --I- al ,x1 -: ? ? -I-
ri -4- (r,,A, .. rz, . 12
? ?
Cl 1XI + a 2.1 z . . .? at, V, . .a,X1 -
ii
. . . . .......
(1/.1 VI + ai2x2-1-?-?-Fahi xi+ + a
where xi is the unknown partial component; aij is the calibration coefficient; and
Ji is the intensity of the mass spectrometer lines in the spectrum of the analyzed
mixture.
In order that the coefficient a may remain constant, it is necessary--in addi-
tion to a sufficiently law pressure within the source and a constant electron energy--
to have a constant temperature in the region of ionization. The importance of this
condition increases with the increase in the number of atoms in the analyzed mole-
cules. However, the existing experience in the use of the mass-spectrometer gas
analyzers shows that with the usually em7loyed stabilization of electron emission
during the analysis of little distinguishal'e mixtures, the temperature of the source
remains practically stable, without the special stabilizers used in more complicated
analytical mass-spectrometers.
* A more exact, but more difficult method is the use of the number of equations
larger than k. In this case one has to use a method of the least squares (see the
article in /T7).
STAT
154
? L?
? r) r4
if i1 A d for Release @ 50-Yr 2014/05/13 CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
Figure 1.MAGS-2.
Anotler factor, vihich influences the reproducibility of the mass-spectrum and
the sensitivity of the mass-spectrometer is the degree of cleanliness of the internal
parts of the mass-spectrometer chamber. The author's experience shows, for example,
that in making systematic analyses of hydrocarbon mixtures, sudden changes in the
spectra are due to the contamination of walls of the analyzer tube (made of red
copper), which appear after 2-3 months of operation. These variations are due to the
formation of surface charges on the deposits, thus influencing the ion beam. These
layers are easily removed by simple mechanical cleaning.
However, even with careful stabilization of electronic emission and temperature
and using a carefully cleaned instrument of existing construction, one fails to
achieve a stability of the mass-spectrometer sensitivity, corresponding to variations STAT
155
Declassified in Part - Sanitized Copy Approved for Release 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
smaller than a few percent in several days. Many times better is the stability of
the ratios of sensitivities toward various substances and the stability of the distri-
bution of intensities in the spectra of individual matters /1:7% As a result, the
experimenter, attempting to obtain more exact results, often uses the method of
"ratios," introducing at times a srecial internal standard.
Let us dwell briefly on the MAGS-2 apparatus to illustrate the operation of a
mass-spectrometer gas analyzer. Its experimental models were issued by 1II MRTP 27:7%
The 1:AGS-2 apparatus (Figure 1) represents a construction unit containing a
single-beam mass-spectrometer, in which a segmental magnetic field is used for the
separation of the ion beam. It possesses a system of molecular supply of the analyzed
mixture introduced into the source of ions and a system of automatic recording of
mass-spectra. The diagram of t e instrument is shown in Figure 2. A small volume
of the analyzed mixture ("-, 8 cm3)--after the measurement of its pressure by means
of a mercury manometer, read through a microscope--is let into a supply vessel having
a volume of about 8 lit. From this vessel the mixture gets into the ion source
through an opening 10-15 in diameter, made in a thin (96,-- 1/46) aluminum foil
which insures the molecular character of supply. The system of automatic recording
of mass-spectra includes a wide-band scanning of the spectrum by changing the magnetic
field. The recording of the spectra takes place on a strip of an autopotentiameter
EFP-09, connected to the output of the electrometric amplifier with a full feedback
circuit. The low level of background noise within the gas analyzer is guaranteed by
the extensive use of stainless steel as a construction material, and the use of a
vacuum valve which separates the instmnent chamber from the pumping system. The
resolving capability of the instrument is 1/80, the range of mass numbers is 1-80, the
recording time of spectrum within the subrange 12-20 m.e. is 12 min. The stability
of the relative sensitivity is characterized by the deviation of t 1-1.5% from the
average value during several days.
156
STAT
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
ore
12.4tomelsoypow
Noce-coeurrpc"..ros w nputkpa Ampos
dormotftd,
ama
2OMNta
4"?.
mOt.11
fraVflai
II 4
YCLefillalei.
nOrM10"010
7
MeAWOMn gmAlupoy.0.6.
0#mmegacm
g ? - m./.7ammi
? radifirwom,s
I tont/orenft
_r Kaman jterurmpattuu vastrarrx grolta
_1 ismftw
&Tmr*fte
stattnAmeeit
10 _
i ?
14,,Afjo, X 5:1*.e.-7,471cma-
mciol --OA=
~wive
- 7
of11/117/1 mimosa .zetkorptfrevevahr
1Z ?
crivame-wA
amuumw
svomArow
JA4-4,l
/3_12.uvemmwommovolgiamonaAmg__________
------ ---- - --- - - -
Figure 2. Schematic arrangement of apparatus MAGS-2
18-
Ill:MMMthWNLM
1474.2Veftbref
1/0#4711
ef-ezzoec...1.:
WIPOw.A
0A40.4WM
A410,01,64,0
enakows,rop
Jo.vam,
A.7mida
110/.0.70,c4.02
itirman ?
uproe.eta
dolireleuti
No.vlogramman.
T
1 I
1
2.4
211
741..4farporw
AtlftWMy
FrsysItcw9
047M:Mfelffp
it
1
kInft4wa?
arpa.wer
?0
33
/UR
-c
ze
Alipako iferow.noxi 4
Mff4M1MIZ NOM.
pupa opellounte.nmare
rv "varletfir
-4 1
3 i
fi,nyamplidakg.rucioenil
riCATick7 S 1 r1/atoCHO- oinkaoszrA cacinemtz - j
to
? T
S
3
Anflsow
ruumlwomisagramuravy
eacmema Iftgenuo amonampyemod
cmeta I uemowar uthvid
L_ -J
1-15 valves.
Legend: 1 - mass-spectrum part of the apparatus; 2 - electromagnet; 3 - ion
receiver; 4 - electric meter; 5 - chamber; 6 - ion source; 7 - direct current
amplifier with 100% negative feedback circuit; 8 - regulator for the power
supply of the amplifier; 9 - section for the recording of ion currents;
10 - magnet current rectifier; 11 - section for the supply of the electro-
magnet; 12 - ferroresonance potential regulator EPA-58; 13 - measurement
and power supply part of equipment; 14 - suppressor battery; 15 - autopotent-
iometer EPP-09; 16 - manual switch for recording scales; 17 - stabilization
and regulation block for the magnet current; 18 - section for the supply of
ion sources; 19 - battery source of accelerating potential; 20 - emission
stabilizer for cathodes of ion source; 21 - section for the measurement of
pressure; 22 - ionization gauge; 23 - thermocouple gauge; 24 - mercury gauge;
25 - high vacuum trap; 26 - mercury diffusion pump of the chamber; 27 - high
vacuum trap; 28 - oil vapor trap; 29 - mechanical pump for the preliminary
evacuation; 30 - mercury diffusion pump for the supply vessel; 31 - ballast
vessel; 32 - evacuating system; 33 - supply diaphragm; 34 - supply vessel;
35 - flasks with investigated mixtures; 36 - system for the introduction of
mixture to be analyzed into the ion source.
157
STAT
(rnv Annroved tor
Rel
@ 50-Yr 2014/05/13 CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
Table 1
Mixture of Gas Hydrocarbons
Mixture Composi-
tion, Volume %
Devia
tion
From
dosage
data
Substances
Found
by mass-
spectro-
meter
method
Abso-
lute
Rela-
tive
Acetylene
Ethylene
Ethane
Propane
Table 2
35,1
? 32,2
14,6
? 18,3
:axture of Liquid Hydrocarbons
35,3
32,5
15,2
18,7
+0,2
+0,3
+0,6
+0,4
-I- 0,6
+0,1
+3,9
+2,1
Mixture Composi-
tion, Mol %
Diverge
nce %
From
weight
data
Substances
Found
by mass-
spectro-
meter
method
Abso-
lute
Rela-
tive
H-hexane
Cyclohexane
2,2,3-trimethylpenthane
H-de cane
Table 3
21,0
30,9
22,1
26,0
21,1
31,5
22,1
25,3
Mixture of Oxygen-containing Organic Compounds
4-0,1
+0,6
0
-0,7
+0,5
+1,9
0
--2,7
Mixture Composi-
tion, Mol 5
Divergence %
From
dosage
data
Substances
Found
by mass-
spectro-
meter Abso- Rela-
method lute tive
Ji-butyl alcohol
Ethyl alcohol .
Methyl alcohol
Acetone . . .
Methyl acetate
158
16,1 15,9 -0,2 -0,8
26,6 26,4 -0,2 -0,7
19,1 19,0 -0.1 -0,5
20,6 21,0 +0,4 +1,9
17,6 17,6 0 0
STAT
Dni-F rnpv Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Table 4
Mixture of Chlorine Compounded Methane
.Mixture Composition
Mol %
Substances
From
dosage
data
Found
by mass-
spectro-
meter
method
Divergence,
UI
Absolute
CC1,
13,6
14,3
+0,7
CHC13
20,9
18,4
?2,5
CH,C12
19,8
20,2
+0,4
CH,C1
28,7
27,4
+ 0,7
CH,
19,0
19,7
+0,7
r-
Substances Analyzed with the Aid of the Mass-spectrometer
-As a result of investigations during the recent years, the range of substances
which can be analyzed with the mass-spectrometer increased greatly. This range now
includes apparently all or practically all substances entering into the raw materials
or products of the oil and chemical industries. Among these substances can be
found hydrocarbons and organic compounds, containing oxygen, sulfur, nitrogen, not
to mention such comparatively simple materials for analysis as, for example, air
gases, inert gases, etc.
It must be noted that in more complicated cases, for example, in the analysis
of crude oil, of benzene and heavier fractions, the complete mixture analysis can
only be made through the combination of the mass-spectrometer with other methods
(adsorption separation, distillation, optical methods).
Tables 1-4 show examples of analysis made with domestic apparatus 27,
Application of Mass-spectrometer as a Control Instrument
under Industrial Conditions
The application of the mass-spectrometer as a control device is due to its
extensive analytical potentialities, its comparatively high speed of analysis, the
159
STAT
norinccifia,r1 in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13 CIA-RDP81-01043R004500020003-3
high sensitivity in many applications, and the possibility of continuous control
when connected to the stream.
It must be noted that in the U.S. mass-spectrometers have been widely used
during the last three or four years for production control /47, 5, 67. Let us point
out some application areas of mass-spectrometer control.
1. Continuous control of raw material composition and products in the production
of the acetylene and of the hydrocarbon gases. The special effectiveness of such a
control is due to the particularly critical conditions for the optimum course of the
process: the output of acetylene decreases by 405 with the temperature deviation of
50 C from the preset one.
2. The determination of harmful admixtures of ethylmercaptan (0.15) and other
substances in the derivation of methylmercaptan used in the production of methionine.
3. The analyses of gases, trapped in dirt extracted by borers, to determine
the location of the oil layer (the mass-spectrometer is placed close to the well).
4. The analysis of waste gases for H2S and SO2 in the plants processing hydro-
gene sulfide into sulfur. It is claimed that the use of the mass-spectrometer in-
creased the sulfur output from 91.5 to 94.55 or decreased the sulfur content in the
waste gases by 355.
5. The determination of the end point of water liberation in the treatment of
metals with hydrogen (in vacuum welding).
In the Soviet Union, a series of investigations were also undertaken during the
recent years. Let us mention some of them. In ITP ry, the mass-spectrometer
analysis used experimentally for the control of industrial processes for the elimin-
ation of high vacuum equipment, gave rise to new opportunities for the determination
of the best method of pumpout.
At the Uhl' of the Academy of Sciences U'S, G. D. Tantsyrev and the author of
this article were able to carry out a mass-spectrometer control of the purity of the
inert gases for various technological problems.
SI-AT
M. V. Tikhomirov, A. I. Kasperovich, and C. Sh. Bykov /-87 performed experiments
160
in Dar+ qnniti7Pri rODV ADDroved for Release @ 50-Yr 2014/05/13 CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Co .y Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
. for the determination of 03 H6 admixtures in the industrial gas mixtures, which has
ethylene as their main component. The experiments were conducted under plant operat-
ing conditions and gave successful results.
NIVI NRIP has already developed and manufactured several years ago several
models of the NAG-3 apparatus 7-97, which has been used, in particular, for the pur-
ity control of ethylene mentioned above. The MAG-3 apparatus (Figure 3) is an example
of a mass-spectrometer gas analyzer intended for continuous control. This apparatus
Figure 3. MAG-3 apparatus.
uses a 1800 deflection of ions. The mass-range is 12-100, the resolution capability
is 00. The apparatus uses a permanent magnet. The adjustment for the mass is
accomplished by means of step variations of the ion accelerating potential according
to a given program. A six-point electronic recording potentiometer can record six
lines of spectrum in 30 sec.
Construction Problems of Mass-spectrometer Gas Analyzer for the
Control of Production Processes
In the development of a mass-spectrometer for the control of industrial processes,
161
narinccifipri in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
STAT
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
many problems occur, which, in the case of mass-spectrometers intended for exper-
imental work, either do not play a substantial role or are not very important,
Let us examine these problems in order.
1. Problems Connected with the Increased Speed of Analysis
The process of mass-spectrometer analysis is divided into four stages: (a) the
establishment of fixed conditions after the introduction of the mixture into the
apparatus; as the work experience at the IKhF of the Academy of Sciences USSR shows
there occurs, in the system with molecular supply during this period, the establish-
Mem
to
C1.1C
Ct
Cat
z
8//d
Zerzon-
Az-liattfoot
114-n,
" /xot445
Mlt
/7 to
'tom
Ca
411??
14 SI
POrPO
;fay
2
H
0 *V(rMW) -3 i
MiwW
vnidef-33
em a di Ilr
4' timitid 14- lapd Pawl -37
.2.2:14,?
fa
c..16 /4? 7) -----
e /
.z.y
..... --. ..
....?
tt
k k
ul to, k Sep/ 4 4
zq
0 zS goy
3r*---- 'limy( ythot#ue -111
Afrnatfommeutor irptimornwie ,,c,i
Oropotmat ....seo I
-4.0 .1
Figure 4. Schematic diagram by the IKIIF of the Academy of Sciences USSR
of magnetic-field exponential scanning of the mass-spectrum.
1 - 350 ohms; 2 - SG3S; 3 - SG3S; 4 - SG3S; 5 - SG3S- 6 - tube 1 6N9;
7 - 1 kilohm; 8 - 12 kilohms; 9 - tube 2, 6N9; 10 - 1 kilohm; 11 - 1 kilohm;
12- SG3S; 13- 7.8 kilohms; 14 - tube 3, 61-49; 15-1 kilohm; 16-1 kilohm;
17 - SG3S; 18 - SG3S; 19 - 3:0 kilohms; 20 - 300 kilohms; 21 - 300 kilohms;
22 - magnet; 23 - 0.2 mf; 24 - 20 r...1-o?-s; 25 - 20 megohms; 26 - ma; 27 - tube
5, 6P6: 28 - 16 ohms; 29 - 130 nf; 3 - 8 megohms; 31 - +450 regulated; 32 -
16 ohms; 33 - from 1 to 4; 34- from :2 to 200; 35-4 megohms (R'5);
36,- charge; 37 - discharge; 3E: - megohms (R5); 39 - switch; 40 -'automatic
switching of speeds; 41 - 4 me,:nns; 42 - discharge; 43 - 123,000; 44 - 8
megohms; 45 - 360,000; 46 - charge; 47 - manual acceleration; 48 - tube 4, 6PP.
ment of thermal ecuilibrium in the ion source and partly the establishment of the
adsorption equilibrium; (b) the recording of the spectrum; (c) the pumping out of the
analyzed mixture until the erasing of the absorption memory; (d) calculation of theSTAT
162
npriassified in Part - Sanitized COPY Approved for Release
50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
mass composition from the mass-spectrum.
The first stage usually takes several minutes. The duration of the second stage,
generally speaking, is different for different apparatuses. In the majority of our
domestic apparatuses, the recording system consists of a direct current amplifier
with full negative feedback and of an automatic potentiometer EPP-09; the inertia
of this system, naturally, limits the speed of scanning. To obtain the maximum
possible (with a given inertia) speed of spectrum recording--as has been indicated
by the author J.27--the a...)plication of the optimum law of spectrum scanning is
essential, which happens to be an exponential law.
The arrangement by means of vihich the exponential scanning of the magnetic field
in the mass-spectrometer is accomplished, was developed by the IiChF of the Academy
of Sciences USSR /117. This arrangement, shown in Figure 4, also permits automatic
variations in the speed of scanning by accelerating it in the areas between peaks.
The 115-1A /17 apparatus makes it possible to record a spectrum in the range of 12-200
in 20 min. The duration of the third stage essentially is determined by the adsorp-
tion memory of the apparatus. This memory is considerably longer for substances
with polar molecules (water, alcohol, ammonia, etc.). The literature indicates that
the adsorption memory is related in the main to the region between the diaphragn-
admitter and the ion source. By bringing the diaphragm right against the region of
ionization, it is possible to decrease considerably the adsorption memory. Our exper-
ience shows that in such a case it is possible to reduce it to several minutes even
in the cpe of the strongly adsorbing substances.
It is possible to point here to another possibility for a large decrease in the
duration of the first and the third stages. This possibility consists in supplying
the gas to the mass-spectrometer as a modulated molecule beam. The experiments con-
ducted recently at the Ii{hF of the Academy of Sciences USSR indicate the possibility
of this method. The arrangement of the supply to the mass-spectrometer, used in these
tests, is shown in Figure 5. The analyzed gas is contained in volume 1 and is supplied
to the ion source in the shape of a molecular beam through channel 3. Gate 2, whichSTAT
163
narlaccifiinri in Part - Sanitized Com/ Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
is opened and olosed by means of magnet 4, modulates the beam. Inasmuch as the beam
does not practically touch the walls of the source and the materials which are ad-
sorbed by the walls of the vacuum part of the supply system are not introduced into
the modulated component of the molecular beam, one can expect, by increasing the ion
current in the modulated part of the molecular bean, a reduction to a minimum of the
inertia of the mass-spectrometer due to the establishment of thermal equilibrium and
of the adsorption memory. Our tests showed, in particular, that in this case it is
possible to decrease many times the memory in comparison with apparatus without the
molecular beam but with a diaphragm, located near the source.
Figure 6 shows a time variation curve of the intensity of the line 18 m.e. in
the mass-spectrum of water after its supply to the ion source has stopped. The curves
7/40% 111
7 ve?
al
4,4
di)
64rAzoiee,
P
A
Figure 5. Arrangement of the supply part of the mass-spectrometer.
Legend: a) to diffusion punp.
were obtained at the IKhF of the Academy of Sciences USSR on three apparatuses: on
the industrial isotope ns-2 apparatus, on the LS-1A 1-27 mass-spectrometer and on
the experimental mass-spectrometer IKhF with a molecular beam.
On the abscissa axis is shown the ti 7c after the beginning of the pumpout of
the supply volume; on the ordinate axis is shown the intensity of the line 18 m.e. in
percent of the original.
As to the decrease in the duration of the recording of the mass-spectrum, the
only way here is the introduction of the oscillograph registration with rapid scanning
of the spectrum. Although the descriptions of individual apparatuses of this type STAT
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
164
in Part Sanitized COON/ Approved for Release @ 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
STAT
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
was published p27, the real investigations of their capabilities were not made. In
those cases where the graphical representation and the speed of registration is more
important than the accuracy, this method can be very promising.
The calculation stage in the case of the multicomponent mixtures is found to
be the most lengthy and takes at times 80 to 905 of the time required for the entire
analysis. For these cases, the reduction of this stage through the use of computing
vu
fo
1ig
'
JO
V:
10
.ill
1
11
'
101
I
a
il
p
-
II
?.
1
.4
.-
III
I/
`011MIIM..
-.'illi
fi
tJI
CC
4.1
42
'I
I/
1
I;
2.1411:7110
5 ilintso oniavArt ?wit
Figure 5. Adsorption "memory"
in the case of water by mass-
apectrometers with different
supply systems.
1 - supply system YS-1 (needle valve); 2 - supply system ES-1A; 3 - supply
system with modulated molecular beam (experimental mass-spectrometer of
TKhF); 4 - intensity of the line = = 18 of water; 5 - pumpout tine, minutes.
machine techniaue becomes very greet. For the solution of systems of first-order
165
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
STAT
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
algebraic equations, relatively simple analog arrangements can be used. These are,
however, not sufficiently accurate when the number of equations is great; the use
of an electronic calculating machine becomes then very desirable.* With the preset
? programming, which can encompass practically all the cases one is interested in, the
time which is required by one machine to calculate the results of one analysis appar-
ently does not exceed several tens of seconds.
It is believed that, under present-day conditions, the use of computing technique
in the mass-spectrometer applications is more of an organizational problem than a
scientific one.
2. Problems Arising when it is Reouired to Connect the Mass-spectrometer to a Stream
for Continuous Automatic Control
It is evident that with t!e Lse of mass-spectrometer for continuous controls,
the problems connected with a more rapid analysis in certain cases preserve their
urgency. However, these cases in chemical and oil production will not be very fre-
quent at the beginning. These problems involve the various uses of the mass-spectro-
meter for the simultaneous control of many pipes or reactors with automatic consec-
utive switching of the ,supply system of the mass-spectrometer from one control object
to another.
In the majority of cases, the control involves one or two objects with very slow
possible variations of the composition of the mixture.
However, when the mass-spectrometer is connected to a stream, new problems arise,
connected with the gas supply systems for the ion source, the recording of the spectrum,
the automatic programming of the operatio% of the apparatus, and with the insurance
of a prolonged and stable operation.
In principle two systems for supplying the gas from the plant piping into the
ion source are possible: with cne stage or with two stages of pressure reduction.
In the system with one stage, the supply to the ion source must be accomplished pri-
* The problems of the use of calculating machine technique in the mass-spectrometer
is described in detail in the monograph A.V.
-
166
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
STA
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
manly according to the law of viscosity. In this case the quantitative character-
istics of the component content of this mixture are not the magnitudes themselves of
the mass-spectrometer peaks, but only their ratios*. Here we find the main drawback
of this system: when we use the automatic calibration with a-standard gas, which
appears to be desirable, it is necessary to add a measured amount of this gas directly
into the stream directed into the mass-spectrometer, i.e., to create an internal stand-
ard. The two-stage system is free from this drawback, where the gas is continuously
eleiminated with a vacu= pump from the volume, with v resistance at the inlet, in-
suring a volume pressure of about 0.1 Trml of the mercury column. This volume is con-
nected to an ion source by means of a diaphragm, which insures a molecular supply of
gas into the source. In this case the supply of the standard gas can be alternated
with the supply of the mixture to be analyzed. Such an installation is used in the
U.S. for the control of the gas products in the cracking plants /-47. In the Soviet
Union a similar system has been used by IAT of the Academy of Sciences USSR, in de-
veloping a system of mass-spectrometer control for the products of underground gas-
ification of coal 11.57.
The other problem encountered in the continuous mass-spectrometer control is
the stability of the calibrating characteristics. As all mass-spectrometer operators
know and as wy mentioned above, the relative sensitivities of a mass-spectrometer for
different gases are more stable than the absolute sensitivities. Even in the case
of a more careful preparation of ion sources, the mistakes in the analyses, performed
with the "method of ratios" are two or three times smaller than with the use of
absolute sensitivities. Because of this, it is desirable, in continuous analysis,
to use a standard gas which is supplied automatically into the ion source alternative-
ly with the mixture to be analyzed. The IAT of the Academy of Sciences USSR is work-
ing on the further development of this method: the use of the ion current of the
standard gas for an automatic correction of the amplification factor of the mass-
spectrometer amplifier LW.
Two systems of spectrum recording are used in the mass-spectrometer for con-
* See, for example, the survey LW%
167
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part- Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
tinuous analysis: a continuous system with an automatic prograrming of the record-
ing ranges and a dot system for the maximum values of the ion current in the mass-
spectrometers (an example of the system of this last type is the recording system
in the MAG-3 7-97 apparatus). The continuous system naturally insures a completely
reliable recording of a maximum; the dot system requires for reliable recording, a
special automatic arrangement which is now being worked out in particular by the
authors of the MAG-3. With continuous control, the use of an electronic oscillo-
graph appears particularly desirable.
In case when the problem requires the watching of only two or three spectrum
lines, it is most convenient to use correspondingly two or three collector mass-
spectrometers with a simultaneous recording of the ion current in each section (or
the ratio of these currents) /n7.
In case of continuous automatic control, a substantial role is assumed by the
system of automatic programming. In the general case, this system must perform the
following tasks: select the mass-spectrum lines, which must be registered; establish
the range of sensitivities, within which each line must be recorded; control by means
of automatic valves the supply of the mixture for analysis, alternating the connection
with various pipes and with the vessel containing the standard gas.
3. Problems of Simplicity and Small Dimensions of the Apparatus
One of the main courses, followed by physicists and designers who are attempting
to simplify the apparatus, is the development of the new and the improvement of the
ordinary methods for the distribution of ions according to their masses.
At the present time tree systems baed on this principle are being used.
In one of the systems, proposed by Bennett Z17, 187, the ions emitted by a
source pass through several groups of grids, each containing three grids. The first
and the third grid of each group have a constant potential; the second grid has an
alternating potential, oscillating with a frequency of several megacycles; the ions
which do not get into the proper phase are slowed dorn and, as they pass through each
STAT
168
IThno-Inecifiari in Part - Sanitized Com/ Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
succeeding group, they are slowed more and more. The probability of an ion getting
into phase depends upon the ion speed, which, with similar kinetic energy of ions W,
depends upon their mass:
1/2W
v=v ?in ?
The inhibiting field in front of the collector does not pass any ion with a kinetic
energy smaller than a given magnitude.
The other system employed is similar to the linear accelerator (or retarder)
in which the ions pass through a series of accelerating (or retarding) intervals
farmed by grids which have potentials oscillating at high frequency. At the
output of the tube we have an electrostatic analyzer which passes to the collector
the ions which have accumulated (or lost) a definite energy.* According to the
infarmation found in literature /7.8, 6:7, apparatuses of this type are being manu-
factured. The scanning in the first apparatus is achieved by the variation of the
accelerating potential, and in the second one by the change in frequency.
The third system was developed in the Soviet Union at the FTI of the Academy
of Sciences USSR /30, 237. In this system the source emits pulses of ions with a
given frequency which move toward the collector. In front of the collector there is
located a grid to which is applied a shutoff potential pulse a certain time t after
the source had emitted the ion pulse. To the different times correspond the groups
of ions of different masses.
Another method used at the present time in the mass-spectrometry is the method
of cyclotron resonance A.17. The apparatuses of this type are the miniature cyclo-
trons in which the collector receives ions with masses satisfying the equation
m=q?FIT
2m '
where T is the oscillation period of the accelerating potential, H is the intensity
of the magnetic field in which is located the chamber, and q is the ion charge.
* It must be noted that the idea of using a linear accelerator as a mass-spectrometer
was brought up by Alkhazov end Yurin in 1941 597.
STAT
169
? Z ?
? r)
if i1 ('rirw Aooroved for Release @ 50-Yr 2014/05/13 CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13 CIA-RDP81-01043R004500020003-3
The scanning of the mass-spectrometer in this apparatus is achieved through the
variation of T.
It is now generally accepted that all these apparatuses are related to the class
of dynamic mass-spectrometers.
Anong the industrial static instruments, a mass analyzer 13787 is beginning to
be used, employirg the intersection of mgnetic and electric fields, and thus poss-
essing a double focusing (for the direction and the speed) 2:-.2V. The operating prin-
ciple of this apparatus is clear from Figure 7. The magnetic field is directed per-
pendicularly to the surface of the figure, the direction of the electric field is
shown by the arrow. S1 is the slot of the ion source, K is the collector. The sub-
stantial advantage of this analyzer with a cycloidal trajectory is also its capa-
bility of focusing ion beams with a large aperture.
Figure 7. Principle of operation of the analyzer with double focusing.
Legend: 1 - ions with lower energy; 2 - ions with larger energy.
In the Soviet Union, N. Ye. Alekseyevskiy and his collaborators Zpy very
effectively used a magnetic mass-sn,-ctrometer with e nonuniform field to increase the
resolution power of the mass-spectrometer. ';ne of the results of this work proved
to be the possibility of obtaining an instrument with a smaller radius without the
loss of the resolving power. An experimental model of the mass-spectrometer for the
analysis of mixtures, using the focusing of N. Ye. Alekseyevskiy was demonstrated
at the All-Union Industrial Exhibit in 1956 by the "Neftepribor" construction
department. STAT
170
?s? .4 ? in,r+ Car-NT ri ('rnv Aooroved for Release @50-Yr 2014/05/13 CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
It must be noted, however, that in spite of the large scope of efforts expended
on the new mass-spectrometers, each of it by itself does not result in the simplifi-
cation and cost reduction of mass-spectrometers. The reduction in the dimensions of
the analyzer part itself, obtained in certain cases, affects very little the overall
dimensions because these ,,re determined Primarily by the vacuum pump and the electron-
ic equipment. One cannot also say that this achieves a simplification of electronic
equipment, since the replacement of the magnet, in particular of a permanent one,
hardly constitutes a simplification of electronic circuits. It is more correct to
consider that tke main advantage of the new types of mass-spectrometers, resulting
from the use of broad particle streams or beams with a large aperture, consists in
the increase in the ion currents and consequently in the increase in sensitivity for
the similar resolving power. Even this conclusion, at least with respect to dynamic
instruments, must be reached with a certain caution. Thus, as has been reported, in
particular all American dynamic instruments, mentioned in this paper, which have a
resolving power of V100 to 1/120, produce ion currents of 10-1? to 10-11 a, i.e.,
practically the same as those of the ordinary static instruments.
The present state is such that, in the selection of the mass-snectrometer type
for industrial production, it is not yet possible in many cases to make a flat decision
in favor of one or the other type. The decisive reasons in such a selection can be
considerations linked to the reliability of operation, based on a sufficiently ex-
tensive experience. A considerably larger area for simplification and reduction in
the cost of mass-spectrometers is found, in the opinion of the author, in its vacuum
and electronic parts. It suffices to say that semiconductor technique has not yet
found its way into this field. It can also be stated that the use of atomic batteries
for the acceleration of ions in the mass-spectrometer is very promising due to the
comparatively low currents. Finally, of very great importance is the conversion in
the mass-spectrometers, particularly intended for continuous control, from the mer-
cury pumps and traps with liquid nitrogen to the oil pumps and efficient charcoal
traps, used in many mass-spectrometers in the U.S.
171
STAT
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
In concluding this section let 116 note, that the use of the mass-spectrometer
in industry can require its construction in an explosion-proof form.
Possibilities for the Use of the Mass-spectrometer Gas Analyzer
for Automatic Control of Industrial Processes
Although the application of the mass-spectrometer for control has a long history,
its application for automatic control, judging by the Published information, is still
somewhere off in the future. The most interesting subjects for regulation would be
the chemical processes in the industrial reactors and columns. It would appear that
the problem is clear: to supervise the concentration of several components and act
upon the corresponding parameter (temperature, pressure, speed flow, mixture compos-
ition) so that these concentrations or their ratios may be optimum. However, when
the problem is so stated, serious difficulties are immediately encountered. The
point is that, at the start of any chemical production, the technician usually selects
the best conditions Which are then maintained constant. Thus, let us say, when the
output of the product falls, it is usually due to some noncontrolled or poorly con-
trolled factor, for example, the state of the catalyst. At first glance it is not
clear how it is possible to apply here the principle of automatic control. However,
the author thinks that there is one possibility which has not yet been examined. Let
us illustrate it with an example in which the reason for the decrease in the output
is due to the change in the state of the catalyst which must be replaced. Naturally,
such a replacement is made when the output falls below some value of the allowable
output previously determined, which lies, let us say, 55 below the maximum. The
question now is whether during the entire period of the decrease in activity of the
catalyst all other conditions were kept at optimum values. Apparently, this is
far from always the case. It is possible, for example, that during this period the
largest possible product output with each given state of the catalyst is reached
under somewnat different reactor temperature than when the catalyst retains its
original activity. In such a case the application of the automatic temperature reg-
ulation of the reactor becomes justifiable with the aid of the mess-spectrometer
172
STAT
Declassified in Part - Sanitized Copy Approved for Release 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release . 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
40
which supervises the output of the product and maintains its concentration 'at its
maximum. In another case such a parameter can be not the temperature but, let us say,
the composition of the mixture, the speed of product flow, etc. In this case it is
possible to obtain optimum conditions during the entire period of the decrease in the
output--from the original to the "allowable." It seems that the possibility of such
a control must exist and that the chemists, after making the corresponding investi-
gations, and maybe on the basis of already existing experience, could point to the
appropriate concrete examples. In examining these concrete examples, the decisive
factor will apparently be the econemics. It may happen, for example, that we will
lose more if we maintain for a long time an output slightly higher than the allowable
one, instead of changing the catalyst more often. In any case, the role is assumed
by several percent of the product in a production with a large tonnage is clear,
and in this connection the value of the examined approach is clear for the use of
the mass-spectrometer for the automatic control of chemical processes.
Another, apparently more long-range use of the ma.ss-spectrometer consists in
the automatic control of a mixture composition, resulting from the mixing of the out-
flow from the two pipes. In the simplest case, the control can be accomplished by
acting automatically on the valves in the pipes.
The designers of the MAG-3 apparatus in particular are now working on the develop-
ment of a device for the solution of problems of this type.
It is possible to increase the number of similar examples; however, it is ques-
tionable whether it is necessary to do this in such a general manner. As to the
fonnulation of concrete problems, the best way to examine them is through an extensive
use of the mass-spectrometers, first of all, for automatic control, particularly in
'pilot installations.
In order to use the mass-spectrometer as an instrument for automatic control, it
is necessary in the general case to employ a more or less complicated computing in-
stallation, which, depending on one or another relationship of the ion currents,
will send a signal to the operating devices. The problems arisingin such a case aiSTAT
173
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Cop Ap roved for Release ? 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
examined in tEe report by professor N. N. Shumilovskiy in this session.
Conclusion
Above were examined the problems related to the construction and the application
of the mass spectrometer as an instrument for the control and automatic regulation
of industrial processes.
In discussing the construction of the mass-spectrometer we have concentrated
our attention only on the problems Which are characteristic in its application for
the control purposes, leaving aside, to a certain measure, the general problems
pertaining to mass-spectrometry as a thole, the problems arising with the increase
in the sensitivity of the devices, in the suppression of the background noise, etc.
We have not attempted to develop final recommendations for the possible types
of devices, and have limited ourselves only to the obvious selection of the class
of these devices for the continuous control. Different problems naturally require
instruments of different types: it is useful to note that it is always possible
to build the spectrometer in such a way that the different problem can be solved
through the addition of the corresponding blocks and links to the main instrument.
Such en approach, as is evident, makes the instruthent cheaper when massproduced.
In conclusion, it is possible to state that the existing experience in the use
of mass-spectrometers for control points to the great importance of the development
in this field. In the Soviet Union, vast scientific and technical experience has
already been accumulated in the area of mass-spectrometry. From this standpoint, there
is practically nothing to prevent an exLensive application of this new and progressive
method for automatic control except the ectil lack of industrial manufacturing of
mass-spectrometers for the analysis of mixtures. All the domestic work, which we have
mentioned, was done on the experimental models, constructed at times 3 to 4 years ago.
It appears to us that the present session must call the attention of the existing
organization to the abnormality of the current situation.
The preparation of this report uas facilitated by a highly valuable discussionswr
174
M
? ori ('rnv Annroved for Release @ 50-Yr 2014/05/13 ? CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13 CIA-RDP81-01043R004500020003-3
???
in which participated at different times R. Ye. Rybchinskiy, M. V. Tikhomirov, G. D.
Tantsyrev, V. R. Anders, P. A. Florovskiy, M. S. Olkhovskiy, and Ya. A. Yukhvidin,
to wham the author expressed his thanks.
BIBLIOGRAPHY
1. Taltroze, V. L., Tantsyrev, G. D., Yukhvidin, Ya. A., Zav. lab. /Plant Laboratorg,
21, 1955, page 1174.
2. Yukhvidin, Ya. A., Zav. lab., in print.
3. Kupriyanov, S. Ye., Dzhagatspanyan, R. V., Tikhomirov, M. V., Tunitskiy, N. N.,
Zav. lab., 21, 1955, page 1182.
4. Lanneau, K. P., Article in the collection Applied Mass Spectrometry, Inst. Petr.,
London, 1954.
5. Walker, J. K., Gifford, A. P., Nelson, R. H., Ind. Eng. Chem., 46, 1954, page 1400.
6. Priestly, W., Dudenbostel, B. F., Ind. Eng. Chem., 48, No. 2, 1956, page 81A.
7. Kalyabina, I. A., Otchet 5epor7 No. 54--1100, 1955.
8. Tikhomirov, M. V., Kasperovich, A. I., 3yk, S. Sh., Zav. lab., 20, 1954, page 830.
9. Menshikov, M. I., Rybchinskiy, R. Ye., Zav. lab. 20, 1954, page 826.
10. Talfroze, V. L., ZhTF, 25, 1955, page 2280.
11. Dekabrun, L. L., Lyubimova, A. K., ZhTF, 25, 1955, page 2280.
12. See, for example, Sin, VI., Rev. Sc!. Instr., 18i--1-947/,-page 540.
13. Barnard, G. P., Modern Mass-spectrometry, London 1953.
14. Taltroze, V. L., Zav. lab., 20, 1954, page 292.
15. Report, IAT of the Academy of Sciences USSR, NM-604, 1954.
16. Hunter, I. A., Stacy, R. W., Hitchcock, F. A., Rev. Sci. Instr., 20, 1949, page 333.
17. Bennett, W. H., J. Appl. Phys., 21, 1950, page 143.
18. Thomas, B. W., Ind. Eng. Chem., 46, No. 8, 1954, page 7IA.
19. Alkhazov, D., Murin, A., DAN, 30, 1941, page 204.
20. Ionov, H. I., Mamyrin, B. A., ZhTF, 23, 1953, page 2101.
21. Morgan, W. A., Jernakoff, G., Lanneau, K. P., Ind. Eng. Chem., 46, 1954, page 1404.
22.. Alekseyevskiy, N. Ye., Prudkovskiy, G. P., Kosourov, G. I., and Filimonov, S. I.,
DAN, 100, 1955, page 229.
SI-AT
23. Agishev, Ye. I., Ionov, N. I., ZhTF, 26, 1956, page 203.
24. Mariner, T., Bleakney, W. P s. Rev., 72, 1947, page 792.
175
in Dar+ qnniti7Pri rODV ADDroved for Release @ 50-Yr 2014/05/13 CIA-RDP81-01043R004500020003-3
N. N. Shumilovskiy and R. I. Stakhovskiy
AUTOMATIC GAS ANALYZERS OF THE LA.SS-SPECTROMETER TYPE
General Considerations
In the automation of continuous industrial processes, a very important Part
must be played by the proper control of Qualitative indicators of the principal manu-
factured product in the various stages of its processing through the use of auto-
matic analyzing devices operating on one or another principle. Iodern production
requirements bring out mc-re distinctly and more urgently the problem of analysis
of multicomponent mixtures for all or at least for many of their component parts.
In this connection, mass-spectrometers have been becoming more and more wide-
spread in recent years. They make possible the obtaining of a mass-spectrum of an
analyzed product and, after the corresponding computations, permit the determination
of the gas composition of the analyzed mixtures.
The great attention devoted to this field of measuring technique in recent years
is explained by the advantages of the equipment of this type over other equipment
for gas analysis. These advantages are:
possibility to analyze multicomponent mixtures for all components;
relative speed of analysis;
great sensitivity of instrument and accuracy of measurements;
possibility, in many cases, to perform the analyses of mixtures only with the
aid of this equipment.
Equipment of the mass-spectrometer type ur.fortunately has many drawbacks, among
which can be counted, first of all, bulkine-s, complexity, high cost, and so far
inadequate adaptation for automatic, continuous analysis of production under direct
manufacturing conditions.
The presently known types of mass-spectrometers can be divided according to the
method of analysis of ion beams into two groups: equipre-t of static and dynamic
types.
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
176
STAT
Dni-F niti Pri Cnnv Aporo ed for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
The first group (static type) comprises equipment in v&ich the separation of
ions is accomplished either in the electric or the magnetic, slowly varying field.
In the equipment of the second group, rapid variation of an electric field is
used, in which the ionized particles separate either according to their transit time
between the ionization source and the receiving collector, depending upon the mass
of the particle (ratio of mass to charge), or upon their own frequency, also depending
upon the mass.
At the present time, mass-spectrometers of the static type can be considered
as hating been mastered in the USSR. At GSKB ZS-tate Special Design Bureau7 for gas
analyses of the :!PSA, the industrial production of mass-spectrometers of types :.:S-2,
LLS-3, and MS-4 for isotope analysis has been set up. Simultaneously, other organ-
izations have developed me:ny designs for mass-spectrometers: MAGS-2 (NII-160),
(NIZI LEP), mass-spectromezer for PGU (IAT of the Academy of Sciences USSR and
ITNIIPodzemgaz) and for the analysis of crude oil products (SKB-6 i,21P).
Industrial models of the dynamic type of mass-spectrometers have not been pro-
duced as yet in the USSR, and only laboratory and developmental models are in exist-
ence. These are: the laboratory model of the transit time mass-spectrometer at the
Physicotechnical Institute of the Academy of Sciences USSR, the developmental model
of the radio-frequency mass-spectrometer with a parabolic field in the Department of
Industrial Electronics of LEI, the developmental industrial specimen of a mass-
spectrometer of the transit time type at GSKB for gas analysis of :ESA.
In the light of problems set forth by the resolution of the Twentieth Congress
of the Communist Party of the Soviet Union in the field of automation of the national
economy, the instrument of the mass-spectrometer type must receive wide application
in the installations for automatic control, as well as in the installations of auto-
matic regulation of industrial processes in accordance with qualitative character-
istics of the manufactured products. It is therefore very important right now to
determine correctly the directions which will be followed by the development of this
type of equipment for their best adaptation to the demands of industry. STAT
177
#
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Many characteristics (speed of analysis, smaller complexity, and smaller di-
mensions of the vacuum section) make equipment of the dynamic type more suitable
for the above-mentioned aims. In the U.S. a series of such instruments have been
produced for industrial purposes. Among them can be listed: the apparatus of the
firm Philips Petroleum Company (of the Bennett type) for the analysis of gas mixtures
(hydrogen, helium, methane, nitrogen, argon), the "omegatron" of the General Electric
Esso Company for the analysis of hydrocarbons, the apparatus of the Beckman Instrument
(of the Bennett type with stagnation feature) for the analysis of hydrocarbons.
In this connection, one of the actual problems is the development of new dynamic
mass-spectrometers, intended for automatic control and regulation.
At the same time, extensive work must be undertaken for further development of
equipment of the static mass-spectrometer type intended for automatic control as
well as for automatic regulation, as well as their manufacturing for use as automatic
continuous-action gas analyzers for direct determination of multicamponent composition
? of an analyzed mixture in industrial departments.
Industrial tests made with the mass-spectrometers MAG-3 (NIVI) and PGU (IAT of
the Academy of Sciences USSR and .VIIIIPodzemgaz), with step scanning of the accelerat-
ing potential for the definite number of t:e most characteristic masses, gave positive
results. The time required to perform the analysis of the six compotents with this
type of scanning was only several minutes.
For the regulation of high-speed industrial processes (where this is required)
in two or three components, it is possible to use a static mass-spectrometer with
several collectors for the simultaneous automatic control or regulation of the in-
vestigated mixture for each characteristic component.
The direct analyses of complex mixtures and t:e determination of the components
of such a mixture requires a long mathematical processing of the measurement results.
This can be achieved in the equipment through the use of special computing arrangements,
which will considerably decrease the expanded time for the interpreting of the records
obtained in the instrument.
178
STAT
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
Such a computing installation can consist of a separate computing desk with
manual input of data obtained from the ordinary mass-spectrometer, or it can consist
of an additional intermediary block within the instrument itself, which automatically
receives the output pulses of the masses and which automatically process them in the
form of pulses suitable for the typewriter which writes corresponding to the magni-
tude of components in the analyzed mixture.
In this last case, the mass-spectrometer directly assumes the functions of a
multicomponent gas analyzer. In the manufacturing of such a gas analyzer of the mass-
spectrometer type for automatic control and regulation of production processes, it
is necessary to satisfy the following requirements:
provision for continuous automatic control;
suitability for oppration under plant conditions in the presence of the plant's
corrosive gases;
provision for automatic processing of primary results of measurement of masses,
and for automatic recording on a receiver device of the composition of the analyzed
mixture (automatic gas analyzer);
reception of signals at the output of the gas analyzer intended for automatic
regulation;
insurance of stability, with time, of the sensitivity of the entire measuring
installation as a whole.
In the majority of cases, the mass-spectrometer can insure only a periodic con-
trol of masses, with a subsequent automatic amputation of the component parts of
the analyzed mixtures. The time between individual measurements will determine the
periodicity of control. The reduction of the time measurements for each individual
cycle can be achieved through the step-type variations of the scanned fields (magnetic
or electric). In addition, the instrument must have an automatic regulating arrange-
ment, which allows for measurement taking according to a definite developed program.
The necessity for installation of this instrument near the analyzed praiuct
in the plant requires its protection from harmful external agents. At the same time
179
s' n r+ Ca .t. ri ('cnv A ID o ed for Release @ 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
?
?
I.
the servicing of the measuring section of the installation requires its periodic
inspection. In this connection one of the measures for combating the action of cor-
rosive gases is the separation of the entire installation into two parts: the
vacuum part, installed directly in the neighborhood of the control point and located
in a hermetical enclosure and the measuring part of usual construction, located in
one of the central locations. This last arrangenent requires the installation of a
remote-control automatic regulation for a coordinate operation of these parts in
accordance with a preset program.
Instability of Ion Currents in Mass-spectrometer Gas .Analyzer
The use of mass-spectrometers as automatic instruments of the industrial type
creates as one of the basic problems the achievement of stability of its indication
together with unchangeable overall sensitivity.
At the same time it is known that calibrating characteristics of mass-spectro-
meters vary with time and that repeated calibrations are necessary. In the case of
automatic gas analysis, for well-understood reasons, the stability reouirements of
the sensitivity are considerably higher and additional corresponding measures must
be taken.
One of the main reasons for the instability of a mass-spectrometer is the in-
stability of ion currents caused by the physical trocesses which take place within
the ionization chamber and the analyzing tube.
It is known 7 that the magnitude of the ion current depends upon the temper-
ature of the ionizing chamber of the ion source. This is determined by the perform-
ance of the cathode of the electron gun, the external conditions, the composition of
the analyzed gas, etc. In order to avoid this influence, it is necessary to control
the chamber thermostatically.
The analysis of the operation of the chamber shows that its temperature instabil-
ity is not the main reason for the instrument instability; at the same time, the re-
quired temperature stability (within the limits ?-10 C) complicates the arrangement
of the equipment. In this connection, the domestic mass-spectrometers of the indus-
S TAT
180
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Co .y Approved for Release ? 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
trial type do not use temperature stabilization of the ionization chamber.
With the passage of time, under the action of electron and ion bombardment, n
semiconductor layer is formed on the metallic surfaces subjected to their actionn.
This phenomenon is connected with the recombination of ionized hydrocarbon molecules
and their subsequent deposition on the nearest metallic surfaces 1-27. The most
favorable conditions for the appearance of such layers occur in the ionization chamber
of the ion source mad the analyzing tube. The hydrocarbon molecules can reach the
vacuum system of the gas analyzer from the analyzed hydrocarbon, as well as from the
oil vapors occurring when oil diffusion pumps are used.
The occurrence of semiconductor layers on the walls of the ionizing chamber can
bring about a change in the effective potentials of electrodes, as compared with
those which are set externally by the sources of supply. These variations can be as
high as 10 v or more. As a result of this, a disruption of the focusing of the ion
beam can take place.
As far as the analyzer tube is concerned, the semiconductor film formed on its
surface will result, under the action of a prolonged ion bombardment, in erroneous
potentials, but of somewhat lower magnitudes, because of significantly smaller den-
sities of ion beams. However, it must be taken into account that the produced false
potentials act on the ion beam along the entire length of the analyzer tube, and this
increases their influence on tl-e total instability of the output ion currents. As
was shown by the experiments conducted at the Institute of Chemical Physics of-the
Academy of Sciences USSR, the film formed in the analyzer can result in the instabil-
ity of the ion currents on the order of several percent.
The removal of the film can only be accomplished by mechanical or chemical means,
end only at the time of the dismantling of the instrument. During the operation of
the instrument, between the regular dismantling and cleaning, it is necessary to take
into consideration the occurrence of a certain instability of the ion current due to
the formation of this film.
STAT
The magnitude of the total ion current i* in the ionization chamber of the ion
181
narlaccifiPrl in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
source is determined by the equation:
= P ? Qi ? $
(1)
where ie is the value of ionizing electrical current; p is the pressure inside the
ionizing chamber; Qi is the cross-section of the ionization of the investigated
molecule, depending upon the energy of the electron; and t is the length of electron
trajectory within the ionizing chenber.
Usually it is considered that the ionizing electrons are of the single-energy
type and the magnitude of the ionizing current is measured by the anode current of
the electron gun, which is stabilized by one or another means. In addition, it is
considered that the negative space charge of the electron beam, although noticeably
influencing the magnitude of the ion current, does not influence the stability of the
ion current because it remains constant.
Actually the situation is complicated by the electron secondary emission, occur-
ring at the walls of the ionization chamber and at the anode of the electron gun DV.
The results of experiments conducted at the IAT of the Academy of Sciences USSR
established that the electron beam, inside the ionizing chamber is not of a single
energy type because of the presence of the secondary emission from various surfaces
of this chamber. The amount of slow secondary electrodes within the entire beam de-
pends upon the magnitude of the magnetic field in the region of the source; in its
absence this may be as high as 70%; when we have a magnetic field with an intensity
of 200 oersteds, this amount decreases to 25%.
The presence of slow secondary electrons in the beam affects the magnitude of
the cross-section of ionization and gretly increases the negative space potential,
created by the electron beam. This leads to the disturbance of ion-optical proper-
ties of the source and from here leads to the variation of the magnitude of the ion
current.
In this manner, under normal operating conditions of the ionizing chamber, there
exists within the electron beam of the ion source a large amount of secondary electrons,
which materially affect the magnitude of the ion current. With time, the properties STAT
182
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
of metallic surfaces of the ionizing chamber change as a result of the formation of
4
various films and deposits. In addition, each time the instrument is operated, a slow
degasification of the electrodes of the ion source takes place, with the result that
the coefficient of the secondary emission changes. All this leads to the uncontrol-
lable fluctuations of ion current which can reach a value of 20-30% and even more.
The magnitude of the ion current is proportional to the pressure within the ion-
izing champer, which depends upon ti.e speed of specimen supply into the ion source,
on the aerodynamic resistance of the corresponding parts of the system for the supply
of the specimen, and on the speed of elimination by the diffusion pump, which us-
ually does not remain constant.
In this manner, the maintenance of a prolonged operating stability of the mass-
spectrometer by preserving its sensitivity causes many material difficulties be-
cause of the impossibility of maintaining the stability of the ion current whose mag-
nitude is substantially influenced by the uncontrollable processes which arise within
the ion source and the analyzer tube.
The technical level in the area of mass-spectrometry is such at the present time
that there is no hope for eliminating all these influences entirely. It is necessary
to continue in the future the investigations of these phenomena in order to find
further means for their complete suppression. Concurrently with this, in the design
of an industrial type of an automatic gas analyzer, it is necessary to insure its
stability by corresponding automatic correcting arrangements, intended for the com-
pensation of all the detrimental phenomena, which interfere with the correct oper-
ation of the gas analyzer.
Methods for Periodic Automatic Calibration of a Mass-spectrometer
To adjust the indications of mass-spectrometers continuously and automatically,
IAT of the Academy of Sciences USSR is developing a special system of periodic cal-
ibration for automatic gas analyzers of the mass-spectrometer type.
The method of automatic calibration of the instrument (standardization) is used
183
STAT
im,,,ineeifiori in Part - Sanitized CODV Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Co .y Approved for Release ? 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
i
r 7 7f^-1
Figure 1. Block diagram of a gas analyzer with a periodic
automatic calibration.
1 - vessel with comparing gas; 2 - three-,way valve; 3 - supplier; 4 - pump;
5 - analyzer; 6 - ion current amplifier; 7 - automatic corrective arrangement;
8 - recording potentiometer.
very extensively in American practice; in the Soviet Union it has been used in iso-
lated cases, where it gave definite positive results.
The method of a periodic calibration proposed by IAT of the Academy of Sciences
USSR is shown in Figure 1. The gas analyzer is supplied with a vessel 1, containing
a comparing gas of constant and known composition. At predetermined time intervals
the gas analyzer is disconnected automatically, with the aid of an automatic valve 2,
from the line of the controlled product of production andis connected to the vessel
containing the comparing gas with a composition close to that of the controlled-
product.
After the recording of the partial spectrum of the comparing gas, the instrument
is reconnected for the control of the production product. The fixed initial recording
of the composition of the comparing gas and of the controlled product makes it possible-
-each time during the interpretation of the records--to use the scale coefficients
from the preceding recording of the comparing gas and thus eliminates the errors accum-
ulating during the time between periodic calibrations.
During the comparative tests of the mass spectrometer for tie five masses of
PGU, using the calibration system with a standard gas, as well as without it, note-
worthy positive results were obtained, as shown in Table 1. (The comparator gas was
the PGU gas of known composition.) S TAT
184
narlaccifiPrl in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
?-?000Z0009170n1?1701.0-1.8dC11-V10 ? I./SO/171.0Z -1A-09 ? eseeiej .104 panaiddv Ado ? Pez!l!ueS -1-led LI! PeWsseloaCI
K ? Coefficient of the decrease in error.
981
lijil-).:.f.,
Components -
.1::;.-7, 2: b:
_
o I
0ec..7...10
7.-3 le .3
tn
-
without
correction
Re
' 10 hours
cc, e Ca, di. 0
14 IC ---.4
cp
17ith
correction
?
x
*
I { ;
,
, ,
4 0 0 4 0 g g g
o o o o o o o o
?,-1 *4-1 or( .4-1 st-1 .8-1 .1-1 91
+3 4' 4-) 43.p -P +3 44 4) 4)4.) 44
0 0 0 * 0 0 C.) w* Z 0
g K* 0. . v*
0 CD 0 K o 0 0 0 o 0 0 n
.o ;.4 .4 S-? A s-i A E "
-p si -P S-. 44 $.4 PS.I -P Si -P si +3 SA 4.) ,-4
?ri 0 .4-1 0 .4.-1 0 .4-1 0 ...4-1 0 ?.4-1 0 ??ri 0 LH 0
:-.- 0 :-.? t) t-- 0 .7.- 0
4?-? .-- .. r. 0 t;I: 0 i LI 0 ?-" 0
-.?
?-?000Z000917001?1701.0-1.8dC11-V10 ? I./9O/171.0Z -1A-09 ? eseeiej .104 P8AOJCWV Ado ? PeZP-IeS -1-led LI! PeWsseloaCI
Declassified in Part - Sanitized Copy Ap roved for Release ? 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
As we can see from the table, the use of correction from the automatic calibra-
tion reduced the errors of each measurement by 7 times on the average.
The development of this method is due to the use of special schemes and arrange-
ments, which insure an automatic correction of the instrument and maintain the scale
coefficients (of sensitivity) constant at all times.
For this purpose, the device 7 is connected to the amplifier output of ion
currents, which is able to change at the time of calibration the sensitivity of the
entire measuring instrument sufficiently so that, independently of the instability
of the ion current, the signal at the output of the analyzer tube, supplied to the
input of the receiver of the measuring device, will remain constant and equal to
the value preset beforehand. At the end of such a calibration, the preset sensitivity
of the mass-spectrometer gas analyzer will remain unregulated for the entire time
interval between the consecutive calibrations. As a result, the maintenance of the
required scale constant in the given case is accomplished automatically.
? It is difficult to determine in advance the time necessary for this automatic
calibration, because the instability of the mass-spectrometer still possesses anindividual
character. This can be done only with the help of corresponding tests of the instru-
ment by means of numerous recordings of the spectrum of the analyzed gas. The in-
stallation =1st, however, allow for the establishment of different spans of time for
such a calibration.
Such an automatic correction consists in the automatic maintenance of a constant
magnitude of the mass current of the comparing gas, which is supplied to the instru-
ment at the time of the automatic calibration. To accomplish this, it is necessary
to ha-:e an amplifier which is capable of a...4:onatically changing its sensitivity
sufficiently so that its output signal will remain constant independently of the in-
stability of the ion source. The sensitivity of the entire instrument established
in this manner must be "remembered" until the next cycle of calibration.
The arrangement of such an installation is shown in Figure 2.
In order that the calibration of the instrument indications may be made compar-
186
nprlaRRifipci in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
STAT
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
.1
a.
atively quickly, it is necessary to insure a rapid and complete changeover from the
investigated gas to the comparing gas and conversely. This can be achieved by a
special admitting system of sample selection, one variation of which is shown in Figure
1. The given arrangement provides only for the viscous supply of the gas into the
ion source.
In the case under examination, the selection of the sample is carried out by
an uninterrupted suction of the gas through the automatic three-way supply valve;
a small amount of this sample is drawn into the ion source. The suction can be
applied to the gas pipe with the controlled gas as well as to the vessel with the
comparing gas. The required switching is done with the aid of special electromagnetic
Figure 2. Amplifying arrangement of the mass-spectrometer gas analyzer.
1 - amplifier of ion currents; 2 - setting potentiometers; 3 - servoqmplifier
EU-18; 4 - calibrating potentiometers; 5 - recording potentiometers;
6 - P1-4 plates of the searcher ShI-50.
switching valves of the solenoid type having an operating time of 0.1 sec.
The cm-
plete automatic switching of the instrument takes from 2 to 3 sec.
Figure 3 shams a variation of the input system with a small expenditure of the
comparing-gas and with molecular feeding of the sample into the ion source. In the
given arrangement the valves Kl, K2, K3, and K& are automatic small size of law-vacuum
type, developed at the IAT of the Academy of Sciences USSR.
The system provides for a periodic picking of a portion of the gas from the pipe.
When the valve K1 is opened, the volume between valves Ki, K2, and K3 (a few cubic
centimeters) is filled instantly with a pressure equal to that of the pipe, because
S TAT
187
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
4
the entire system, has been evacuated beforehand by the vacuum pump with Kland K2
closed and K3 and K4 opened. Vihen the vacuum is reached (2-3.10-2 mm of mercury
column) K3 and K4 are closed. When the intercepted volume is filled, K1 is closed
and K2 is opened; the volume thus intercepted now expands in the supply chamber from
where the gas begins to enter the ion source. lath this expansion the pressure of
the gas drops to a value of a few millimeters of the mercury column, thus resulting
in a molecular type of supply. To change the sample, the valves K3 and K4 are opened
and the gas is pumped out by the vacuum pump. In order to admit the comparing gas
into the instrument, valves Ki and K2 are interchanged. In constructing the admission
/low" gpactratamtedoletwit
Ere
5
Immwwymmorsoly
1?77
Figure 3. System for the admission of a picked sample.
1 - elastic ccntainer; 2 - hermetic container; 3 - vacuum pump; 1(1_4 - remote-
controlled valves; V - supply volume; 4 - pipe for pressure equalization;
5 - to ion source.
system it is necessary to take into account the fact that, to insure a correct cal-
ibration of the mass-spectrometer, it is necessary to maintain the pressures of the
analyzed and of the comparing gas equal. However, there is no necessity to have
them constant all the time; but care shou1:1 "_e taken that, during the calibration,
the pressure does not change appreciably. If the raintenance of equal pressures for
sone reason becomes difficult, it is possible to maintain a constant ratio of these
pressures.
As a comnrsing gas, it is desirable to select a gas which is close in composition
the analyzed as and w4lic
all the components of the latter. The comparing as
STAT
188
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13 CIA-RDP81-01043R004500020003-3
4
must be available in a sufficient quantity to insure periodic calibrations during a
prolonged time. From this standpoint, the determination of the possible use of air
as a comparing gas is of great interest. It is true that in this case limitations
-
and certain difficulties arise; however, the convenience of such an application be-
comes very evident.
Computing Installation for the Determination of Component Parts of the
Controlled Multicomponent Mixtures
The system of equations which must be solved in determining tne component parts
of the controlled mixttz-es by measuring their masses, is written in the following
manner:
A )= 2 .?xb
(2)
where Ai is the directly measured peak of the mass, aji is the scale of each mass
spectrum, determined at the time of calibration; xj is the value of concentration of
the j-th component of the controlled
valueof.results in the equality
xj
where
.Aj.
mixture.
Ai
xi= y,
an; an;
au; a24; ? ? ? ?
an1; an2
au; an; an
an; a22; a23
an1; an2; an3
The solution
Al am
A2 ? ? ? a2,i
An ? ? ? ann
aln
a2rt
anh
of this system for the
(3)
(4)
(5)
After carrying out the solution for-all components of the multicomponent mixture, we
obtain as a result the following system of equations
189
STAT
Rel
@ 50-Yr 2014/05/13 CIA-RDP81-01043R004500020003-3
STAT
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
?
?
I.
?
11
A
Xi = Ljt. firsI
X2 =
jCjaAj
/-1
A I
?
X n Ecnv-aii
where Cji is the function of coofficients aji.
The solution of each equation is obtained from the computer through the use of
a single summator (solver amplifier) with the use of variable resistors for the
setting of coefficients. The switching of resistors going from one equation to an-
other must be done automatically. If, in addition, a recording device is connected
(6)
to the output of the summator, it will register on the diagram the values of
x2,..., xn after a definite tine interval.
-4 22
1,
vizi rii
F-I
-I 4
Figure 4. Blocks of computing installation with manual
setting of input data.
1 - block for -anual setting of data input; 2 - blocks for setting coefficients;
3 - solution transfer switch; 4 - summator; 5 - recording device.
This type of computer can be made in two versions.
1. Computing Device with Y.ual Setting of Input data.
This device consists of an indivi&-al stand dn which are manually set the values
of input potential resistors in accordancc '_th the obtained records of the controlled
masses from the mass-spectrometer (vrlues of Ai, A2, etc.).
The structural arrf-nrm-ent of such a corputing device for the recording of six
masses (six equations) is shown in Figure A. Here blocks 21 ,W 9 ,O0O,V g are intended for
1
thesettingofresistancesR,inaccordancewithcoefficientsC,which are determined
JI
each time in accordance with t.e controlled gns mixture. The scale coefficients,
190
Declassified in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
in determining the resistances for the computer, are selected in such a way that any
summation of potentials, appearing at the output of the solver amplifier (summator),
4
will be within the allowable limits (in particulLr within the limits of 10-100 v).
A recording device is installed at the output of the solver amplifier to measure the
electric potential (specifically, EPP--9).
2. Automatic Corputer Block, Built into the Automatic Gas Analyzer of the ::ass?
spectrometer Type
The computing arrangement with an autam%tic input of data in accordance with the
value of the ion current, appearing at the output of the analyzer tube, differs from
the manual computer by the presence of a special input block. This block includes
a condenser memory device and a device for the synchronization of pulses.
Figure 5. Blocks for setting coefficients and their signs
in an automatic computer.
1 - input channel for peaks; 2 - channel for the synchronization of pulses;
3 - memory block; 41_6 - blocks for setting of coefficients; 5 - solution
transfer switch; 6 = summator; 7 - recording device.
Within the condenser memory device, the pulses which come from the intermediate
amplifier as a series of individual peaks are sequentially remembered on individual
condensers.
The synchronizing device distributes the received signals to the individual com-
densers according to a predetermined order. After the last, the sixth peak, has been
memorized (the memorizing process is equivalent to the data setting in the manual com-
puter), the solving process begins with the starting of the step-by-step switch. This
process of solution does not differ in principle in any way from that of the computer
with manual data setting. From this point on, the remaining part of the structural
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
191
STAT
norincQifipri in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
arrangement does not differ in any way from the corresponding part of the first
version, as we can see from Figure 5,
4
Figure 6 shows the schematic arrangement of an automatic gas analyzer of the
mass-spectrometer type, which incorporates all the above-mentioned considerations.
The instrument is intended for the control of any six masses within the limits of
1 to 100 in accordance with the controlled multicomponent mixture.
The setting of the masses to be controlled is accomplished manually each time
during the adjustment of the gas analyzer for the particular mixture to be controlled.
A computer block can be included in this arrangement, the a recording device registers
1 - hermetical container; 2 - elastic container with comparing gas;
3, 4, 5, 8, 11, 13, 14 - remote-controlled valves; 6 - supply volume;
7 - diaphragm; 10 - analyzing tube with an electromagnet; 12 - diffusion
pump with a trap; 9, 15 - vacuum punp; 16 - supply regulators; 17 - ion
current amplifier; 18 - calibrating potentiometers; 19 - scanning arrange-
ment; 20 - setting potentiometer; 21 - servoamplifier; 22 - recording
potentiometer; 23 - controlling device.
directly the components of the gas mixture. It is also possible to disconnect this
block from the instrument and to connect the recording device directly to the input
of the intermediary amplifier.
In this case we obtain a record in t're shape of a mass-spectrogram. The input
system of the instrument is altomated anc yvides for a periodic input into the
instrument of the compering :as under a pressure e'aial to the pressure in the control-
led pipe. This is insured by placing the comparing gas into a soft container of
the oxygen pillow type which is loccted in a hermetical compartment, connected with
the inlet of the sample. The automatic picking of a portion of the analyzed gas is
accomplished with the aid of small solenoid valves developed by the Institute of STAT
192
norinccifipri in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Cop Approved for Release ? 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
Automation and Telemechanics. The operating order of valves is shown above.
When a portion of the gas has been pick,:d through a sudden change of the current
in the electromagnet, the six selected masses are recorded, whereupon a smooth
scanning takes place within the range of the single peak to determine its height
accurately.
At the end of the recording of the last peak, the input system changes the
sample by means of the already, described means, and the control cycle repeats itself.
After a predetermined number of control cycles, the comparing gas is admitted into
the instrument by opening the valve K2. This requires a negligible amount of the
comparing gas, determined by the volume between the valves Kl, K2, and K3. now,
in order to adjust for the selected masses, it is necessary to set the sensitivity
for each mass by means of the six rheostats with induction drives. For the automatic
setting, the signal from the mass-spectrometer is memorized at its peak value. After
this the slider on the rheostat changes its position with the aid of a servoamplifier
until the potential on the rheostat becomes equal to the potential on the comparing
potentiometer, manually set. At the end of the automatic setting of sensitivity
for all masses, the instrument again admits a portion of the controlled sample in
accordance with already described method and so on.
Inasmuch as the vacuum section of the instrument must be controlled remotely,
IAT of the Academy of Sciences USSR undertook the development of automatic low-vacuum
and high-vacuum valves. It the present time, IAT of the Academy of Sciences USSR
has developed vacuum valves of the solenoid type with rubber packing; the through
opening has a diameter of 7 =which insures a reliable shutting-off of the system
from the atmosphere.
High-vacuum automatic valves have also been developed and produced with an open-
ing diameter of 30 mm in two versions: with metallic and with rubber packing.
BIBLIOGRAPHY
1. Stevenson, Journ. Chem. Phys., 17, 101, 1949.
193
Dmri? - Caniti7Ad CODV Approved for Release @ 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
STA1
????
Declassified in Part - Sanitized Cop Approved for Release ? 50-Yr 2014/05/13 : CIA-RDP81-01043R004500020003-3
A
2. Koenig, Zschr. f. Phys., 129, 491, 1951.
3. Barnard, Modern Mass-spectrometry, Inst. of Physics Press, 1953.
Address
by Comrade P. Ye. Rybchinskiy
Because of the high cost of mass-spectrometer analyzers (on the order of 50,000
rubles in massproduction), their use will be justifiable only in those industries
where they will result in substantial savings, in particular in the production of
artificial alcohol and in the manufacturing of ethylene oxide. In this last case,
it will be possible to reduce the employed personnel from 100 to 30 persons.
Since the extent of such industries is compartively limited, it will be desir-
able, at least for the present, to design mass-spectrometers, not of the universal
type, but for special applications, calculated for the control of definite types
of production.
The accuracy of analysis, amounting to 2 to 3%, is suitable for present-day
chemical production; thus the various complicated features introduced into the mass-
ppectrometer by IAT in order to increase the accuracy are, in my opinion, not justi-
fiable. In addition, I do not consider desirable the division of the instrument
into two psrts: the analytical block aad the electrical block.
Address
by Comrade R. I. Stakhovskiy
The present-day state of mass-snectrcmeter technique does not allow a long-time
(weeks, months) stability of ion currents, required for an industrial instrument
used in automatic control. It is possible to obtain 2 to 3% only for a fen days.
After this, a new calibration of tne instrument is required. Because of this, the
automatic calibration allows for the increase in reliability and accuracy of the
instrument. Inasmuch as a mass-spectrometer is quite a large and cumbersome instru-
ment, the addition of one block for the automatic correction does not change the
194
STAT
nariaccifipci in Part - Sanitized Copy Approved for Release @ 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
Declassified in Part - Sanitized Copy Approved for Release ? 50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3
structure of the instrument, and also does not change in a noticeable manner
complexity and its dimensions. In addition, this block can be made removable and
can be taken out if it is not needed.
For production cases, where a mass-spectrometer must be located immediately
at the place where the sample is taken, it is wise to have two blocks; in addition,
the electronic part of the instrument must be put in a binding where there are no
corrosive media.
The vacuum part of the instrument can be controlled ramt.tely.
* * *
In a concluding address, B. S. Sotskov, chairman of the session, informed the
gathering of the content of the draft resolution, in particular, about the proposed
organization of the Scientific Council on Automation attached to the Presidium of
the Academy of Sciences USSR, which will be charged with the coordination of scientific
investigations in the field of automation. B. S. Sotskov also announced that a
series of special consultations will be conducted next year, concerning semiconductors,
magnetic amplifiers, and contactless elements, and concerning the problems of regu-
lation, etc.
195
STAT
npnlassified in Part - Sanitized Copy Approved for Release
50-Yr 2014/05/13: CIA-RDP81-01043R004500020003-3