THE CARE AND FEEDING OF MAGNETIC TAPE

• Document Type: CREST
• Collection: General CIA Records
• Document Number (FOIA) /ESDN (CREST): CIA-RDP74-00390R000300440011-5
• Release Decision: RIFPUB
• Original Classification: K
• Document Page Count: 25
• Document Creation Date: December 9, 2016
• Document Release Date: August 2, 2001
• Sequence Number: 11
• Publication Date: April 22, 1966
• Content Type: SPEECH

Approved For Release 2001/08/09 : CIA-RDP74-0039OR000300440011-5
April 22, 1966 10:30a.m.
National Archives Theater
Washington, D.C.
THE CARE AND FEEDING OF MAGNETIC TAPE
Discussion of the physical properties of tape as well as the
advantages and limitations of tape as a record keeping vehicle.
SPEAKERS:
WILLIAM I. HUTTON
Manager, Engineering
and Field Service
IBM Corporation
COLONEL E. C. LAEDTKE
Chief, Data Processing Division
Headquarters USAF' Data Services Center
PRESIDING:
ALLEN N. HUMPHREY
General Accounting Office
ROBERT G. DEVITT
Field Manager
Approved For Release 2001/08/09 : CIA-RDP74-0039OR000300440011-5
 Approved For Release 2001/08/09 : CIA-RDP74-00390R000300440011-5
Allen N. Humphrey, General Accounting Office
Presiding
The program this morning, as you noticed from the flyer, will be
"The Care and Feeding of Magnetic Tape." The discussion will be on the
physical properties of the tape as well as the advantages and limitations
of the tape as a record-keeping vehicle. Our first speaker will be
Mr. Robert G. Devitt. Mr. Devitt is a native of St. Paul, Minnesota. Ile
is a graduate of the University of Minnesota and after serving two years
with the Air Force, he joined Minnesota Mining and Manufacturing Company.
He has been with 3M in various capacities for the past ten years and is
presently a Field Manager of the Magnetic Products Division. Mr. Devitt
will discuss the manufacturing of magnetic tape.
Robert G. Devitt
3M Company
Thank you, Mr. Humphrey. First I would like to express the appre-
ciation of 3M Company for the opportunity to participate in your
conference here on "The Care and Feeding of Magnetic Tape." What we have
this morning is a film strip which we put together for you to show how
magnetic tape is made. Looking at the manufacturing processes, we will be
seeing four types of magnetic tape made. The first is sound recording tape
which is used to record the voice and to record music, and this is the
first application for which magnetic tape was developed. The second type
of tape is video tape. Video tape is used to record pictures. I believe
you have all seen it on your television screens. We think in the next
few years that video tape will be used extensively in the home as well as
in industry. The third type of tape that will be seen is what we call
instrumentation tape. This tape is used to record analogue or continuous
data from remote areas such as satellites. The fourth is the tape that
is used as an input and output device and as a storage device in
computers. We call it computer tape.
This morning we would like to give you an idea of the critical
nature of the process and of the quality control used to make computer
tape that is required to be able to write and read more than one hundred
and sixty million discreet bits in one reel of tape, discreet pieces of
information on one reel. We know it is a requirement that the tape be
free from errors, so that the computer can operate at maximum efficiency.
In manufacturing a magnetic tape we strive to make a tape that is perfectly
free from flaws. So let's take a look at a slide film here on how
magnetic tape is made.
This report is distributed for the Interagency Records Administration
Conference by the National Archives and Records Service, General Services
Administration. For further information, write Conference Secretary,
Office of Records Management, National Archives and Records Service,
Washington, D.C. 20408, or call 963-6426 (Code 13-36426).
Approved For Release 2001/08/09 : CIA-RDP74-00390R000300440011-5
 Approved For Release 2001/08/09 : CIA-RDP74-0039OR000300440011-5
1. MAGNETIC TAPE MANUFACTURING
Magnetic tape is a plastic ribbon coated on one side with material
capable of being magnetized. Its three important components are the
plastic backing, the magnetic particles, and the binder which holds them
together. The raw material for magnetic coating is Ferric Oxide Hydrate,
FE203H20, also used in making paint pigments. However, for magnetic
coating, size and chemical purity are closely controlled.
The first manufacturing step is to remove the one molecule of water,
resulting in alpha ferric oxide. While this is in a very stable state,
the material is not yet capable of being magnetized. The alpha ferric
oxide is then subjected to a reducing gas, either hydrogen, carbon monoxide,
or water gas to produce this black oxide, commonly called magnetite. It
is now magnetic, but has two major drawbacks. First, it has continuing
oxidizing reaction. It still wants to rust. Second, tapes made from it
have a recurring magnetic memory. Erased recordings tend to come back
making later recordings noisy. These faults are eliminated by reducing
the material back to a ferric oxide state through a special heating process.
In this form, gamma ferric oxide, the material is now stable without the
rusting characteristic of black oxide. It also has retained the ability
to be magnetized.
The particle size of gamma ferric oxide for most magnetic coating
is about one micron in length and about a length of that in diameter.
Some newer oxides, however, have a particle size of just one third that
size. Both the minute size and uniformity of size are important to smooth
coating application and to control of the coating magnetic efficiency.
The binder used in the dispersion is likewise important. It is
specially prepared to be compatible with the backing material to be used
on the tape. It also has the durability demanded by the recording
application the tape will serve. Binder and oxide are thoroughly mixed
in large rotating drums which contain small metalic balls to help assure
blending of the two ingredients. This operation, as well as the earlier
oxide reduction process, is isolated from the tape coating and finishing
steps to prevent coating from being contaminated by the oxide dust found
in the air around these two steps. In fact, after a dispersion of oxide
and binder has been thoroughly mixed, it is transferred to these large
storage tanks and never again exposed to the uncontrolled, nonconditioned
atmosphere. In each tank a large power mixer constantly agitates the blend
to prevent premature hardening and provide proper oxide to binder balance.
To guarantee getting the cleanliness needed for magnetic tape production
and to house the especially developed types of coating equipment, the tape
is produced in manufacturing plants designed exclusively for applying
magnetic coatings. One is in Hutchinson, Minnesota. Another is located
in-Freehold, New Jersey and a third in Camarillo, California. This
assures prompt delivery of tapes to tape users in all parts of the
country.
Approved For Release 2001/08/09 : CIA-RDP74-0039OR000300440011-5
 Approved For Release 2001/08/09 : CIA-RDP74-0039OR000300440011-5
In these plants separate air-conditioning units, one for each tape
coating area, prevent dust from contaminating production. They operate
on positive air pressure that moves the air away from the coaters,
never toward them. Automatic vacuum door mats at each entrance remove
the dust and dirt from shoes as further protection. Special air-locked
double doors keep unsettled dust from entering critical areas. All per-
sonnel who work near the tape during manufacture wear special lint-free
clothing. While the need for these precautions is most important during
the time that the coating is still wet, they are found as well in all
parts of the process to guarantee completely uncontaminated tapes.
In the slitting operation each jumbo roll of tape is slit to one
of the tape widths needed in the particular recording uses for which the
tape was made. Before each roll of tape is packed for shipment, it is
vacuum cleaned as it is placed in the box, then exposed to a powerful
electro magnet located just below package conveyors which erases-degausses-
-magnetically cleans each roll of tape.
The many sizes and types of magnetic tape found in this vast ware-
house have undergone a manufacturing process based on the one you have
just seen. However, each different tape, since it is built for a particu-
lar recording application, has it's own set of manufacturing specifications.
Specifications based on the frequency to be recorded, the speed of the
recording system, environmental conditions and other important recording
characteristics in the many different tape uses.
2. BACKING SPECIFICS
Early tape recorders, while they were revolutionary and highly
advanced in their day, would be considered obsolete when compared to the
present day recording equipment. For such early recording equipment,
paper backing materials were used and were good enough, no less dependable
than the equipment. The tough, thin paper was flexible and readily avail-
able. at low cost. But as equipment improved and recording techniques
became more accurate, the physical and magnetic demands on tape became
likewise more critical, bringing out several disadvantages of paper backed
tapes. Paper's low strength factor limited the speeds and tensions that
could be used in the system. Also, humidity caused paper to expand and
thus be further reduced in strength.
While several plastic materials were researched, the first one to
be found satisfactory as a backing material was cellulous acetate.
Compared to paper, it was many times stronger and offered a very smooth
coating surface... Finding the right binder to bind the oxide to the
backing was one of the first problems in making plastic back tapes.
Because of the radically different physical properties between plastic
and paper, a binder had to be formulated which was compatible to the
particular plastic backing being used, possessing the same aging properties,
the same reaction to temperature and humidity, the same wear resistance.
Approved For Release 2001/08/09 : CIA-RDP74-0039OR000300440011-5
 Approved For Release 2001/08/09 : CIA-RDP74-00390R000300440011-5
If the backing and binder expanded at different temperature for example,
tape tended to cup. This held the tape away from the head and reduced the
output, especially at higher frequencies. This major problem was soon
solved and cellulose acetate backing proved dependable and economical in
many applications--applications which are less critical from the standpoint
of temperature and humidity or in which extremely fast tape transport speeds
are not necessary.
But for more critical tape uses like telemetry and other aero-space
applications, where difficult and varying conditions exists, or where high
speed transports generate frictional heat, base materials are made of a
very durable, precisely manufactured polyester formula. Polyester backings
have excellent dimensional stability, are virtually unaffected by atmospheric
changes, and therefore can be used in recording applications where high
humidity is prevalent. Polyester also has very favorable tensile strength
characteristics, important to tape wear ability to high speed recording
applications. Still another unique feature of polyester is its elastic
memory. When the tape is stretched less than it's yield point, in other
words, less than 5% of its length, the tape will return to its original
shape, something many other backing materials won't do.
Determining which backing material should be used for a particular
recording application, depends on the requirements of each application.
In one, the strength of the backing material might be important. In another,
tape stretching or breaking points, or heat and humidity characteristics
might be the prime considerations. This chart compares backing materials
as to stretch, yield point, and tear strength, three of several considerations
important in selecting the proper tape for an application.
As with all tape components, projects or improving present backings
and finding newer ones are the assignments of 3M research people. All kinds
of materials are tested for stress, friction, andwear properties, for their
ability to withstand temperature and humidity changes, and for the overall
compatability in being manufactured into a tape for specific recording
applications. The different combinations of cost advantage, resistance to
tearing or breaking, temperature and humidity, provide the tape user with
a variety of benefits to be considered when ordering tape for a specific
application.
3. COATING SPECIFICS
Different recording applications require different values of tape
coating properties. Some require minimum wear, others the best possible
output, still others high frequency response. 3M magnetic coating research
has developed magnetic dispersions which provide tapes capable of increased
values of all favorable characteristics.
Approved For Release 2001/08/09 : CIA-RDP74-00390R000300440011-5
 Approved For Release 2001/08/09 : CIA-RDP74-0039OR000300440011-5
One advancement that has increased the magnetic values of all
coatings is a manufacturing step which takes full advantage of each
magnetic particle. Early manufacturing applied the oxide as smoothly
as possible, but with no regard to the relationship of one oxide particle
to another. Now, by exposing the freshly coated dispersion to the field
of a powerful electro-magnet, the particles are aligned or oriented in
one direction. Aligned along the tape for most applications, but across
the tape for the vertical scanning of some video recording.
Orientation increases output and results in a better signal or
noise ratio. The increased output partly results from the fact that
more oriented particles can be packed in a given area. Like a pile of
toothpicks which are not evenly aligned, few unoriented particles will
fit into a specific space in the coating. Oriented particles, like neatly
aligned toothpicks, are more greatly concentrated in the same space.
Another coating requirement is uniformity of coating thickness more
critical in some applications than others. Uniform thickness is important
to uniform output, particularly when dealing with long wave lengths.
Since the best output from a tape occurs when enough record current is
used to completely saturate the coating, thickness variations will cause
variations in output. Simply leveling the coating doesn't do the job,.
because of possible irregularities in the thickness of the backing across
the width of the roll. A blade leveling technique would allow some areas
to have a thicker coating than other areas. While this picture is a gross
exaggeration of backing variation, it can be seen that a controlled
coating thickness is important to uniform output.
To. prevent variations in coating caliber, an exclusive coating
process was devised by 3M engineers which applies precisely in the same
thickness of dispersion, regardless of backing material variations.
With this highly precise application technique, coating as thin as 100
microinches would be held to thickness variation of just 10 microinches.
Signal losses or agglomerate of oxide which stick to the coating
surface protrude above the tape surface to cause an error in output,
not just in that spot, but since it holds the tape away from the head
in a much larger area around the spot. This umbrella effect is
especially detrimental in recordings of closely spaced pulse or short
wave lengths and to multiple track recording, since the affected area
often covers several tracks. It is most detrimental in broadcast video
recording where the vertical recorded tracks are only 5000th of an inch a
part. This is why with video tape and with every magnetic tape 3M makes,
every precaution is taken to assure absolute cleanliness during
manufacture.
Coating errors and foreign particles on tape become more serious as
slower tape speeds are used. Slow speeds mean reducing the tape wave
length of a signal so that an error affects a greater portion of an
Approved For Release 2001/08/09 : CIA-RDP74-0039OR000300440011-5
 Approved For Release 2001/08/09 : CIA-RDP74-0039OR000300440011-5
entire wave length. Because of the trend toward slower tape speeds in
many recording applications, the highly protective cleanliness and
coating application controls are constantly being improved to assure
dependable recording results regardless of tape speeds.
The binder in a coating is likewise important to coating perfection.
It must flow smoothly to permit even application. It must have a controlled
drying time compatible with the time it takes to apply the mixture to
prevent the forming of oxide clumps or agglomerates. It must also be able
to supply that durable and flexible bond between the oxide particles and
backing to give the tape a long usable life.
If a binder is durable, but also brittle, the oxide is in danger of
being rubbed off. This again, is most visible in broadcast video
applications, where the rotating heads penetrate the coating at 57,000
times a minute and at a pressure of about 10,000 pounds per square inch.
Finally, the binder must have the same temperature coefficient of expansion
as the backing. Differences in expansion between the two can cause the
tape to cup, preventing good head to tape contact.
These many magnetic tape considerations; coating thickness; coating
defects; durability; smoothness; temperature are basically important to
the production of all magnetic tape. But they differ in importance
according to the applications a tape will eventually serve. That is
why each manufacturing run is controlled by specifications developed
for a specific type of tape and specifically used.
4. MANUFACTURING QUALITY CONTROL
Magnetic tape manufacture is actually three processes in one:
Manufacture of an oxide according to the specific needs of a particular
tape application. Manufacture of a:system of binder ingredients with
those same specific needs in mind. And the combining and coating of
those two components onto a specific backing material, according to a
specific degree of smoothness dictated by that particular tape application.
For this reason, control of quality cannot be simply a sorting system
grading tapes into quality groups.
The tape you use must be designed specifically for your recording
application before production ever begins, because tape requirements
differ with the many different types of recording processes. The testing
for a tape construction actually begins before its production ever starts.
The raw material for the oxide is not accepted for manufacture until a
test sample has first been completely processed into finished oxide,
then hand coated onto a backing and tested to see if it has the properties
needed in a particular recording process. The sample taken from each
new batch of raw material undergoes the same reduction steps as a full
kiln of oxide and the results compared to established standards to assure
magnetic efficiency, correct particle size and chemical purity.
Approved For Release 2001/08/09 : CIA-RDP74-0039OR000300440011-5
 Approved For Release 2001/08/09 : CIA-RDP74-00390R000300440011-5
To guarantee that each kiln-load of oxide meets the rigid standards,
close control is maintained of the heating time, temperature and reducing
gas atmosphere in each reduction step.. Constant inspection and adjustment,
much of it electronically accomplished, periodic spot checks and addi-
tional hand coated test samples take the guesswork out of assuring
uniformity of oxide manufacture. During the blending of these ingredients
into a dispersion, both temperature and mixing time are closely watched
to assure thorough mixing, with still another hand coated test sample
made as .a final magnetic check.
The backing material for tape, like the oxide raw material,
undergoes a rigid pre-inspection to determine smoothness, maximum strength,
uniform caliper and general quality. Backing thickness tolerances are
held to 4% of the materials thickness. That's about 40 microinches on one.
mil tapes. These tolerances, checked on special electronic thickness
gauges, apply not only to overall thickness of the material but also to the
maximum variations in thickness allowable across the web of material.
Probably the most critical and most controlled part of the entire
manufacturing process is in applying the dispersion to the backing.
Using a strict program of quality checks, the sensitive, exclusively
designed coating equipment applies the thin oxide coating within tolerances
as close as 10 microinches.
If the construction is to be an audible range tape, it's properties
are checked on the equipment typical of the professional audible range
recorders used in the field. By making the sample into a loop and
running it on test equipment the tape can be checked for magnetic
efficiency and physical durability as well as surface smoothness and
coating uniformity.
Once the jumbo roll of tape passes the rigid magnetic and physical
uniformity checks, it is then cleared for the tape slitting operation''
where a new kind of quality check comes into play - the job of assuring
cleanly slit, perfectly aligned tape edges. As each set of slitting
knives is set in the slitting equipment, slitting samples are then
checked to assure accuracy before proceding. Samples from each tape run
are then sent to a laboratory for testing on standard recording equipment
typical of the recording gear on which the tape will eventually be used.
Here tapes for instrumentation or computer use are tested on
equipment typical of field operations, with any defects electronically
counted to make certain that tapes meet the strict efficiency tolerance
for these uses. By testing tapes on the same types of equipment used in
the field, the eventual use of the tape can be perfectly duplicated and
its expected performance accurately measured. Video tapes for example
are tested on video recording equipment commonly used in television stations,
the equipment for which the tape was developed.
Approved For Release 2001/08/09 : CIA-RDP74-00390R000300440011-5
 Approved For Release 2001/08/09 : CIA-RDP74-00390R000300440011-5
Finally, packaging precautions - depending on the type of tape and
its eventual use includes special center supports on big reels, seen here
just behind the reel hub to protect the reel flanges from damage. Other
wrappings or cartons have space between the reel and outer surfaces as
further protection. For protection against moisture some tapes are sealed
in a protective plastic bag before being put into the carton. The
thorough precise quality control program developed through laboratory
research and well over a decade of experience in field applications is
the undisputed watch-dog of tape manufacture. Its demands on tape
perfection increased in proportion to the ever growing demands for
accuracy in new recording applications.
Mr. Humphrey
Thank you very much Mr. Devitt and I hope that you will thank 3M when
you get back home for having participated so nicely in our program.
We have as our next speaker Mr. William I. Hutton of IBM. Mr. Hutton
has been involved in the magnetic tape recording field since 1947. His
college background includes both Nebraska and Colorado Universities.
He has served as a broadcasting consulting engineer and as Vice
President of a commercial recording corporation. His first assignment
with IBM was at the Poughkeepsie Center in the field of advanced machine
design and development. For the past ten years Mr. Hutton has been
manager of the Engineering and Field Services at IBM Minneapolis
Magnetic Tape Center. Mr. Hutton will discuss the testing of computer
tape for us.
William I. Hutton
IBM Corporation
Members of IRAC and Mr. Humphrey, it is my pleasure to be here
this morning representing IBM and to give you a,,little background on
your meeting subject, the care and feeding of magnetic tapes, specifically
from the standpoint of the computer and the computer user.
To many people, the magnetic tape is looked upon as an incidental
item to be handled much the same as we do other information media we
use, such as paper and pencils. To those of us who are intimately
involved in the economies of the computer operations, the magnetic tape
is considered as an integral part of a computer. Actually it's equally
as important as any component that we have in the system. For an example,
the failure of a magnetic tape to be written on or be read back will
cause us a loss of valuable system time. And this loss, of course, will
be just as surely a failure as it would have been if there were electronic
components or mechanical components in the system that would have failed.
As many of you are aware, the loss of computer time is expensive and in
Approved For Release 2001/08/09 : CIA-RDP74-00390R000300440011-5
 Approved For Release 2001/08/09 : CIA-RDP74-00390R000300440011-5
many cases the time lost cannot be made up because of the tight schedules
you have for-your equipment. Actually this may result in added expense in
that you may have to go to an outside facility.
The cost of error failures can exceed many times the price of a reel
of computer tape. Any reliability that can be tested into the tape prior
to its use on the computer is a savings for the customer. In order to be
constantly aware of the present and future requirements of magnetic tape
and to establish methods of identifying and measuring these qualities, the
IBM Magnetic Tape Center is continuously working with Tape Drive Engineering
and Development as well as Systems Engineering Development people in our
various laboratories.
Basically, the requirements that we are concerned with are the
reliability and the compatibility of a magnetic tape-reliability when used
in a given system, and compatibility when it's used interchangeably among
systems.
To show you the importance that IBM places on the quality of magnetic
tape, I will outline for you and give some examples of the operations that
are performed at our Magnetic Tape Center and product testing facilities.
At the Tape Center, our primary testing operations are divided into
two classifications.
1. Specification Testing - to determine whether or not each
production lot of tape will meet-the IBM specifications
for magnetic tape - computer use, and
2. If the production lot passes the IBM specification test,
then, Hundred Percent Testing of each and every reel that
makes up this lot. Here we accept only those reels that
can pass four additional writing and reading tests.
In our specification testing operations we make a total of 21 tests.
Each of these tests was designed to measure one or more of the perimeters
that we at IBM feel are essential to the operation of the computer itself.
It is quite important that the measurements obtained in each of these 21
tests be within their prescribed limits, for the state of the art is such
that the manufacturing process cannot be controlled to such an extent that
it can be relied upon to always produce tape which will meet the specifications.
When they do not, the end results can be a tape whose characteristics will
differ considerably.
The chart we have here lists the specification tests we make. One of
the tests listed is a measurement of friction. This measurement will give
us an indication of the starting and stopping as well as wear characteristics
of the tape surface. All have a direct bearing on tape reliability. As an
example, the tape must go from a stop condition to a speed of 112.5" per
Approved For Release 2001/08/09 : CIA-RDP74-00390R000300440011-5
 Approved For Release 2001/08/09 : CIA-RDP74-0039OR000300440011-5
SffCIFICATIO4' TESTS
SELECTION BY SPECIFICATION TESTING ASSURES BASIC
REQUIREMENTS FOR LONG TERM RELIABILITY
21 PHYSICAL, ELECTRICAL S .MAGNET/C TESTS MUST 8E MADE
? WIDTH
? FRICTION COATING TO BACKING
? WEAR OF TRANSPORT
? THICKNESS
? ADHESION COATING TO BACKING
? START TIME
? RESISTANCE OXIDE
? CURVATURE
? 0-H PROPERTIES
? ROUGHNESS BACKING
? CUPPING
? TOXICITY
? ROUGHNESS COATING
? ELASTIC CHARACTERISTICS
? FLAMMABILITY
? FRICTION COATING TO BRASS
? YIELD FORCE
? FRICTION COATINGTO RUBBER
? CREEPOCITY
? QUALITY RETEST ? ADHESION LAYER TO LAYER
Approved For Release 2001/08/09 : CIA-RDP74-0039OR000300440011-5
 0 sill /09 02Adkn 5
;PEI
/00 zo
ALL REELS
FINAL 100% TESTING
OF EACH REEL PROVIDES
DEFECT FREE TAPE...
? SEPARATE TESTS
MADE THE FULL LENGTH
OF EACH REEL
? OYNAM/C SKEW
S/NAL LEVEL
? SIGNAL O? 4/0/SE
Approved For Release 2001/08/09 : CIA-RDP74-0039OR000300440011-5
 Approved For Release 2001/08/09 : CIA-RDP74-00390R000300440011-5
second in a few milliseconds. and maintain this speed without variation.
A high friction characteristic would have an adverse effect on the
acceleration. You might look at the movement of tape the same as the
movement of the platen on the typewriter. If it were to go at various
speeds, the typed information would be spaced irregularly. Of course, in
the case of a typewriter you might be able to read it. On the computer, it
would be impossible to read because computer logic allows only a limited
period of time for each character. The start stop time and the speed of
this tape must therefore be quite rigidly controlled.
To give you an idea of the amount of force that is placed on a tape as it
is accelerated with a capstan, we can use a familiar measurement in terms
of gravity or G. As an example, when an astronaut or flier. When he is
subjected to from 5 to 10 G's will will Clack out. With forces of 15 or
20 G's most planes used today would lose their wings. As a contrast, when
we start a tape from a dead stop we are applying 500 G's of force. The
tape must be able to withstand this tremendous shock and those qualities
that might affect acceleration such as friction, must be watched very closely.
The adhesion, layer-to-layer is another area that we watch very closely.
This adhesion can be adversely effected by changes in pressure and
temperature. Pressures can change due to varying winding tensions on
a tape drive and temperature can vary during use and storage. Layer-to-
layer adhesion, therefore, must be checked quite closely to see that it
does not cause failures during the use of tape as well as the long storage
use of tape.
Another characteristic checked is the long length durability. This is a
check of the quality of the tape as it is used in a system under multiple
pass operations. For this we use a test that we call a 200 pass test.
During this test we write and read on a tape for a. total of 200 passes,
monitoring at all times for any temporary errors and permanent errors.
And at the same time we will be printing out with electronic instruments
the amount of time it takes to start this tape, bring it up to full speed,
and stop it. At the end of the 200 passes we will inspect the tape contact
points on the tape drive for excessive wear caused by the tape. In
addition we will retest the tape that was used to determine if there were
any unusual wear product build up on the tape during the long term test.
This is a very important test because a single pass on the tester could
not reveal the manufacturing variations that would effect multiple pass
usage of a tape.
If all of the lot samples, have passed the 21 test, then we accept the tape
for a second step. This is called 100% testing, where we test each and
every reel individually throughout its usable length. Here there are
four tests that must be passed by each reel of tape. A failure to do so
will cause a reel to be rejected. One of these four tests is called a
signal level test. Here we monitor the average signal level or strength
of the magnetic field recorded on the tape. This is done to be sure that
Approved For Release 2001/08/09 : CIA-RDP74-00390R000300440011-5
 Approved For Release 2001/08/09 : CIA-RDP74-0039OR000300440011-5
the tape will produce a constant signal within prescribed limits. Any
variation in the tape signal outside of these limits causes the tape to
be rejected. Controlling this value allows the IBM tape to be used
interchangeably throughout remote installations, without requiring
machine adjustments. As an example, the various agencies send tapes all
over the country and, of course, businesses do this too. If we can closely
control the signal level, changes to the equipment will not be required
when the tape is used interchangeably between remote locations.
Another 100% test that is made is one of checking the amplitude of each
individual pulse. This is done to be sure that you can properly write on
each possible information position in the tape. In this test, if a signal
should fall below a prescribed amount, the tester will stop then on error.
If this imperfection cannot be removed then the tape, of course, will have
to be rejected. This test is similar to taking a punched card and punching
out all of the possible punch positions then verifying that you can read
them all. In other words, we look at every bit position on the tape,
verify that it can be read with a good safety factor. To give you an idea
of the magnitude of this test and the way that it is continually growing
more demanding as the writing density increases; we began testing back in
1955 at 200 bits or characters per inch. This amounted to checking over
40 million individual positions that could be written on a 2,400 foot
roll of tape. Each and every one of these spots on the tape has to be
checked to make sure that they could be written and read properly before
we could accept the tape.
Starting with 200 bits per inch, we have since moved up to 556 bits or
characters per inch where we tested over 112 million character positions,
then to 800 where we tested in excess of 160 million and now we have added
1600 bits per inch where we are actually looking at and verify throughout
the length of tape, over 415 million individual magnetic spots. You can
see by these numbers, a test such as this is quite a rigid one. It has to
be for we certainly don't want to have a computer failure that will leave
a man up in orbit. Businesswise we are also quite particular. A failure
could change the location of a decimal point. As you can see, each and
every one of these bit positions is important.
The third test is the noise test. This is a test that is made to detect
any magnetic variations that might occur on the tape due to manfacturing
faults. These must be detected and removed as they will cause false
information spots on the tape. In this test we also stop on error and
examine the area. If we can remove the defect we will do so. If not,
then the tape would have to be rejected. This test would be similar to
passing an IBM card without any punches through a verifier and verifying
that there are no punches on the card. If there were to be a spot or hole
on this card that you could read as information of course you would want
to dispose of it. We do the same thing with magnetic tape.
Approved For Release 2001/08/09 : CIA-RDP74-0039OR000300440011-5
 Approved For Release 2001/08/09 : CIA-RDP74-0039OR000300440011-5
The fourth test is one that we call a skew test. This is a test where we
monitor the timing of the pulses on the outermost, or tape edge tracks.
Should this timing vary and go outside the prescribed limit, it indicates
that the physical characteristics of the tape in this area will cause it
to guide improperly. This condition cannot be corrected. It is a guiding
problem and, of course, there is nothing you can do to correct the guiding
condition of magnetic tape, so it would mean that we would have to reject
it. Going back to IBM card analogy again, this condition could best be
described as one you would experience if you were to try to pass a card
through a reader diagonally. It is impossible to do. The columns would
be tilted and you would not be able to read the information properly.
You would be reading parts of several columns instead of one column.
We have the same problem with magnetic tapes as we go to these higher and
higher densities. Any change physically as it goes through the guiding
points would cause you to read the information improperly. We call it
skewing. As an example of the sensitivity of this test at 800 characters
per inch density, if the tape were to be moved .057 degrees, which is
extremely small, this is all the movement that is necessary to cause us to
go into error with the type of equipment that we are using.
We would like to add here that all IBM products have to pass a rigid test
in what we call our IBM products testing laboratories. This, of course,
applies also to our magnetic tapes. The type of tape that we select for
IBM use must qualify in these labs before it can be used on IBM equipment.
The product testing group is a separate entity within IBM and, of course,
is very impartial as it is responsible for the total overall operations
of-the systems. The tests that they give are very rigid and can take up
to 8 months or more to determine all the characteristics required on a
given magnetic tape construction. The tape must pass tests on all types
of tape drives that are manufactured by IBM and that a customer would have
in his installation. This is the only way that we can be assured that the
user will have the compatibility required of todays tapes that are shipped
and used interchangeably between remotely located installations.
After we have delivered what we feel is the finest computer tape possible,
there are some precautions that should be taken when using magnetic tape.
During use, the tape is subject to wear product and handling contamination
as well as handling damage and machine damage. By contamination we are
referring to the accumulation of wear products, room dust, and the general
handling contamination that can build up either on. the tape itself from
handling or on the tape drives mechanism then transfer to the tape.
Cleanliness of the system, cleanliness in handling the tape and periodic
cleaning of equipment is quite important in keeping this contamination
t o a minimum.
Handling damage consists mostly of damages to the edge of the tape from
squeezing the flanges of the tape reel. Damage to the tapes edges
will prevent it from guiding properly. Information that could be written
near this edge could not be read properly.
Approved For Release 2001/08/09 : CIA-RDP74-0039OR000300440011-5
 Approved For Release 2001/08/09 : CIA-RDP74-0039OR000300440011-5
Never place the tape in such a position that it is needlessly exposed
to room dust. As an example, never place it on top of a tape drive
where it is hot and dusty and do not leave it open outside of its case
on a table for an extended period of time. Any of. these actions that
would allow excessive room dust products to accumulate on the tape will
cause the tape to be reduced in its reliability. There are a number of
specific handling and use procedures that should be followed quite closely
in order to maintain the accuracy of the tested IBM tape once it is
delivered to the customer. Now I believe that the handling and use of
tape within the system will be covered by the Colonel so I'll stop at this
point and again thank you for your invitation and we've enjoyed speaking
with you today.
Mr. Humphrey.
Thank you very much Mr. Hutton. The third speaker on our program will
be Colonel E. C. Laedtke. Colonel Laedtke is presently assigned to head-.
quarters USAF as Chief, Data Processing Division, USAF Data Services
Center, operating in. the Pentagon.
After entering the Army Air Corps he became a command pilot. During
World War II he was a_B-17 pilot in Europe flying with the 8th Air Force.
During the Korean War he flew logistics support for Japan and Korea.
Since 1946 Colonel Laedtke has been associated with the Air Force
Controller, the major specialization in the automatic data processing area,
with duty assignments at all levels of command. Data processing experience
began with punch card accounting operations and today encompasses the
management of 9 electronic digital computers. Colonel Laedtke is a graduate
of the University of Maryland. He has had graduate work at George Washington
University and is a graduate of the Air Force War college in Alabama.
Colonel Laedtke will discuss the use of tape as records.
Colonel E. C. Laedtke
Headquarters USAF Data Service Center
I will cover three points briefly. First, magnetic tape as a
document. Second, advantages of using magnetic tape as a document. And
third, disadvantages of using magnetic tape as a document.
Let's now look at tape as another form of a document. How many of
you have home recorders - stero-tape minded? Quite a few. Well, the
tape we use in a digital computer operation, although different in size
and better in quality, performs the same function for us on a computer
that your tape does for you in your music system. Music on your tape at
home is made possible by magnetizing spots on the oxide covering which
was explained to us earlier. In a computer operation it is the same thing.
Magnetizing of different spots on this tape gives us our numbers and our
letters. which the computer recognizes. There is a difference though in
how fast your home recorder will pass tape as compared with a computer.
A present day computer tape unit will move tape much faster than could
ever be accomplished on a home recorder system.
Approved For Release 2001/08/09 : CIA-RDP74-0039OR000300440011-5
 Approved For Release 2001/08/09 : CIA-RDP74-0039OR000300440011-5
In thinking about magnetic tape as a document I would like to run
through briefly, the steps normally required before the manual records in
your offices are represented in another form on a one-half inch or three-
quaeter inch piece of tape. First, before you can take a document and say,
"I'm going to place it on magnetic tape, and I'm going to get a computer
to do my work," you must sit down and organize your data and plan on how
you will use it. Too many people even today, after approximately 15 years
of computer experience to reflect on, still fail at this first very
important point of organizing information so it will go on the computer
in an efficient and effective manner. Many offices will automate, things
will go wrong, they will blame the computer and they will blame the tape.
Really they are blaming a fault in their thinking in that very,early step
of organizing their records so that they can be mechanized. This then is
the first important step for you to consider.
Now let's assume we have accomplished our initial planning and know
what we want. We have our fields of information decided. We have some of
these fields coded, possibly. We know just what we are going to do. Now
we have to get those manual records on tape. To do this we go through
a conversion process and I will briefly identify ways we do it today. A
principle major method of getting that information from your office, from
your filing cabinets, and onto magnetic tape is through the old fashioned
card punch machine. This is still a big "number one" way of accomplishing
this operation but not the only way. Today we also use punched paper
tapes. Punched cards and punched paper tape go through another process we
call "card-to-tape" or "'paper-to-tape" processing in which this information
is transferred to a reel of magnetic tape. Another way, and a relatively
new way, is direct key punching onto magnetic tape. We do not use this latter
method in the installation to which I am presently assigned because its use
is limited by application. It has definite advantages if used properly
and another way to get information from your filing cabinet to magnetic
tape. Today there is also much experimentation going on in the optical
scanning field, wherein a machine "looks at" your manual record and creates
a computer record. I'm brushing over these very fast and want only to
identify this transition step.
The question usually asked at this point by those not too-familiar
with data processing is "How many of my records can I put on a certain
length of tape?" There is no single answer to this question. It depends
on how long your records are, how many characters of information you use,
and how tightly you press these characters together on tape. For example,
over in the Pentagon today, on our particular tape drives we can put
information on tape in three different densities... 200 characters to an
inch, 556 characters to an inch, or on our later drives, 800 characters
to an inch. Certain equipment manufacturers provide capability for even
greater densities. Density means more if we site an example: Consider
a density of 800 characters of information to an inch. An 80 column
punched card....the kind you are familiar with....can be reduced to 1/10
of an inch of magnetic tape. So you see you can handle a great deal of
information on a single 2400 foot reel of tape.
Approved For Release 2001/08/09 : CIA-RDP74-0039OR000300440011-5
 Approved For Release 2001/08/09 : CIA-RDP74-00390R000300440011-5
We now have this information on tape...what next? We now enter the
field of "tape handling", and one very important step must be accomplished.
This is the labeling of the tape reel at the time our information is placed
on it. You know at home how you can "goof-up" your library if you mis-name
or mis-label a particular reel. In computer processing work, the same
danger prevails. So we follow a very strict procedure called "tape labeling"
which can be initiated either in the data processing installation or back in
your office. A new procedure we instituted in the Pentagon recently is one
that requires the customer to prepare the tape label and send it to us
together with their processing request. We of course distribute detailed
preparation instructions for completing the label. Such advance preparation
can't be done in all cases, since there are times when the customer really
doesn't know in advance certain of the identification items required by the
label. We find this system very helpful and it has decreased the number
of labeling discrepancies. You might ask, "What identification and control
items are found on a tape label?" This varies by organization but the
common items include the name or title of the particular data recorded, the
number of the reel, the "as of date" of this particular information, the
customer's code or name, an all important data retention date, the security
classification assigned and the character density at which the reel was
created. As I said, the specific items used vary by installation and the
method of magnetic tape inventory control-employed.
Your data is now recorded on magnetic tape and is ready for filing
in what we commonly call a tape library. A tape library is very similiar
in many respects to a book library. You have your racks and filing sequence
which must be followed religously. We in the Pentagon, for example, at the
present time have approximately 25 thousand reels in our Air Force Library.
Twenty-five thousand reels is classed as a very substantial library, but
number of reels by itself is really not the all important factor. Your tape
activity is of upmost importance. "How often to you withdraw and refile a
tape." This is tape activity and we have great activity. For example, on
a typical day over in the Pentagon, servicing the Air Staff of the Air Force
and Mr. McNamara's office, we utilize on an average day anywhere from 600
to a 1000 reels....sometimes it even exceeds 1000.
Your records are now on tape, filed in a tape library, and ready
for future processing. Now what are some of the advantages of having
your records in this new form? Briefly let's go over a few of them:
First, one I mentioned earlier....the compact filing. You can eliminate
numerous filing cabinets and end up with several reels of tape which is
certainly a space saving advantage. Then, you have flexibility of use.
Previously you had to thumb through your records to find something. Now
you have it on.tape and through computer processing techniques we can
search for, discover, sort, compile, calculate and printout whatever you
want. The flexibility of use is really up to you....the customer. Once
you have the data created, once such recording is accurate, once it is
on tape there is no end to what you can do with it.
Approved For Release 2001/08/09 : CIA-RDP74-00390R000300440011-5
 Approved For Release 2001/08/09 : CIA-RDP74-0039OR000300440011-5
Tied directly to flexibility is of course frequency in use possi-
bilities. Your now "mechanized" basic information can be used as often
as you desire. You might ask "Can I wear out my tape?" Here again I
don't think you will find anybody who will give you a firm solid answer,
particularly if you are using some of the new tape, on the market today.
I will say that it is almost impossible to wear it out if you take care of
your equipment, keep your tape heads clean and exercise care in handling.
Then in the event something does happen you can always duplicate your reel
by "copying" the data on a new reel. Let's also put "realiability" on our
list of advantages. Once your information gets on the tape, it will never
change unless you change it. This, I'm sure, you must consider a definite
advantage.
Ease of shipment and/or transmission is another advantage. It is cer-
tainly easier to ship a reel of tape representing information in half-dozen
filing cabinets than it is to ship the basic data. Also a new innovation
of the past several years permits the transmission of your data to another
location over the Defense Communication Systems Automatic Digital Network
....AUTODIN. This is a world-wide system wherein data from a reel of
magnetic tape or punched card information can be loaded on an AUTODIN
terminal and transmitted practically anywhere in the world and picked up
on the receiving end as magnetic tape or punched cards. The information
might have been originated by Burroughs-type equipment and received else-
where in a manner permiting immediate use by -- IBA equipment. This is a
flexible and rapid way of getting information from one place to another.
In Air Force Headquarters we are today tied together, through AUTODIN,
with all of our major Air Force commands world-wide and during an average
month will receive and transmit in the neighborhood of 2 million data
records.
Another advantage of having your records on magnetic tape is related
to the output products from those records. Today you might have your
record "mechanized" but are wondering what to do with all of the reports
and listings that are printed from these records.... in other words you
are "snowed-under" with management documents that you desire to retain.
Today there is a piece of gear available that will take this digital
information direct from tape and, through a cathode ray tube connection
and microphotographing techniques, photograph these reports for storage
in the form of microfilm ...an advantage.
Let's list "life expectancy" of your taped records as an advantage.
For years we thought a taped record would last forever and many of us today
still think that, under certain conditions, this is true. Since I had to
think of some disadvantages to-give you I'll turn this point around a
bit later on and call it a disadvantage..
As a last advantage let's list security classification and declassifica-
tion. In creating a magnetic tape reel we can classify it in whatever manner
you, as the customer desire. It can also be removed at your request.
Approved For Release 2001/08/09 : CIA-RDP74-0039OR000300440011-5
 Approved For Release 2001/08/09 : CIA-RDP74-00390R000300440011-5
Declassification has been a rather recent innovation. It was only late
in 1965 that the Department of Defense gave the military departments the
"green-light" to declassify previously classified tapes. To accomplish
this a reel is passed through a special degaussing or erasing operation,
using approved equipment, and it again becomes'unclassified. Before
receiving this release permission the only way to declassify a previously
classified tape was to completely destroy it by burning.
My third point this morning concerns disadvantages of placing your
records on magnetic tape. As I said earlier it wouldn't be appropriate
for me to only list advantages so I have come up with several items that
could go on the other side of the ledger. Some of these might appear
"doubtful disadvantages", and if you want to disagree with me you might
not find strong opposition.
First, let's consider "ease of destruction". You have your file on
tape, a file that you spent thousands and thousands of dollars accumulating
and you want to keep it as such. This taped file can be very easily erased
as those of you in the home recording business well know. Just take a
soldering iron and hold it close over the reel and your data is gone. Or
you can inadvertently, although you have to go out of your way to do it,
write somebody else's information on your particular reel of tape. As I
say, you have to go out of your way to do this because equipment manu-
facturers have gone to great lengths to keep operating personnel from
inadvertently destroying your file.
Let's list the requirement for duplicate tape records as another
disadvantage. We found that when you have magnetic tape files it's a
good idea to have some duplicates... not duplicates of everything... but
selected reels. I mentioned that our Headquarters USAF tape library
contains approximately 25 thousand reels and we have our several hundred
of these duplicated and filed in an alternate physical location. The
Air Force learned its lesson back in 1959 when. we tried to burn-down
the Pentagon. I'm sure some of you remember that. That fire destroyed
our entire tape library which, at that time, contained only 4 or 5
thousand reels. Many months were spent and thousands of manhours to
recoup basic data files. So today we have selected alternate files...
files our customers have selected as those requiring such protection.
This additional precaution could, I suppose, be cited as a disadvantage
of having your records on tape.
It might be well to list special tape handling as a disadvantage,
but here I'm sure we will have disagreement. The requirement to exercise
care in handling tape was touched on by Mr. Hutton and I won't repeat what
he said. When handling tape you must exercise a certain amount of care
in removing a reel from its container, in placing it on a tape drive, and
finally in removing it and returning it to its case. But we in our
installation have found out one thing about tape handling.: Some of the
handling precautions we thought years ago were ultra-critical to a computer
installation, are not quite so critical. Dirt surely, elimination of
dirt is important and you should do everything possible to have a clean
installation. You might remember that many of the early data processing
Approved For Release 2001/08/09 : CIA-RDP74-00390R000300440011-5
 Approved For Release 2001/08/09 : CIA-RDP74-00390R000300440011-5
installations prohibited cigarette smoking anywhere near a computer....
even today many do not allow smoking. Some installations required that
you don a white coat and clean your shoes before entering the computer
room....that is if you were even allowed to enter. These precautions,
although good, are extreme and no longer absolutely necessary.
In this area of tape handling I would like to cite an example. Last
November we went through a major building renovation project over in our
computer center, where six of our nine computers are installed. During
this renovation work the Air Staff and Mr. McNamara didn't say "You boys
close up shop". We were expected to continue producing in the normal
manner, so even during the tearing-down and rebuilding we continued our
operation. Part of the renovation involved removing the entire floor
under the computer area. To accomplish this a very detailed timing
schedule was worked out so a portion of our operation would be underway
at all times. As we removed the old floor to put down the new one, we
discovered dirt from the fire of '59 still underneath. In our particular
installation, because of the limited floor space and the close placement
of our computers we have a heat problem which is solved by having two
air-conditioning systems. One of these blows the air in from underneath
the floor and the other one from on top. So, when we took this old floor
up and disturbed the packed dirt you can imagine what happened. We had
more dirt in that computer room than I think any place in Washington.
But we kept operating, and at the same time made it a special point to
keep track of what we could identify as increased tape problems because
of the disturbed environment. It might be difficult to believe, but our
identifiable tape problems showed no increase.
In citing this example please don't take me wrong. I am not advocating
or encouraging computer operations in an environment I just described.
Certainly you should take whatever precautions are necessary, but I wanted
to bring out that when using certain equipment you can continue processing
under very adverse conditions.... possibly I should have placed this item
in the advantage column.
Another problem when your records go to tape, is of course the danger
of using defective tape. As with almost every product on the market today
there is good and bad....or good and not so good. To have effective and
efficient data processing you must use good reliable tape because there
is no end to the problems if you don't.
Another possible problem concerns storage of your taped records.
Although the new tape that we are using today....heavy duty tape....is
not as susceptible to humidity and temperature problems as some of the
early mag tape used a decade ago, we still attempt to maintain certain
temperature and humidity controls in the tape storage area. Such controls,
although not as critical as for the data processing equipment itself,
must still be within certain limits. In our installation we consider
between 40 and 90 degrees fahrenheit on temperature and 20 to 80 percent
Approved For Release 2001/08/09 : CIA-RDP74-00390R000300440011-5
 Approved For Release 2001/08/09 : CIA-RDP74-0039OR000300440011-5
humidity as the acceptable ranges. True, a rather wide range here but
still some control necessary. When a tape comes to us from an outside
environment we like, if at all possible, to have it in our processing
installation atmosphere for at least 24 hours prior to use.
Still another possible disadvantage might be the requirement for
periodic tape'rehabilitation. You can't just keep using tape over and over.
At some point you are required to clean it and check it for defects, both
of a temporary and permanent nature. There is special equipment on the
market used in accomplishing this task and I won't go into this.
Now the last possible disadvantage, and one that could really develop
into a true disadvantage, of using.a tape record. But my personal feeling
is that if it becomes a true problem, ways will certainly be developed to
overcome it. This possible problem was brought to light by one of the
civilian agencies of the Federal Government..... Social Security Administration
in Baltimore. They reported at a recent magnetic tape user conference a
difficulty in reading recorded information in the area of the reel hub if
such a reel was stored for more than three months. A preliminary diagnosis
seems to show a slippage of the adhesive used to hold the iron oxide
particles, and consequently created a situation where the tape is unreadable.
This is something new in our business..... something we have not identified
as a problem in our installation..... to date anyway. Possibly one reason
we don't is that we do not "pack" data onto a reel of tape in the manner
that Social Security does. Usually our files end before we reach the hub
area.....maybe if we were constantly using "fully-packed" reels we also
might have a problem. We understand Social Security has solved or at least
minimized this problem by running their reels through a controlled tension
winding operation which establishes a constant tension as the reel goes
through its rewind cycle. If it is a true problem rest assured it will he
solved.
So.......... There you have some possible disadvantages, and I put
possible in "quotes". You might disagree with me on their classification
and if you do I. won't fight too hard to hold my position since my main
objective was to inform. This morning, in a few minutes, I tried to tell
you a short story about the use of magnetic tape as a record, and advantages
and disadvantages of doing so. And when considering this subject there is
one.thing we should keep in mind.....mag tape records are here to stay.
Improvements in data processing equipment.will certainly continue as well
as improvements in the manufacturing and use of magnetic tape. History
demands the recording of information and today one of the best ways of
doing this is through the medium of the magnetic tape. If your office,
your agency, your activity is not "automated" today, rest assured that one
of these days it will be, and you will then have a firsthand experience
with the subject discussed here this morning.
Thank you ladies and gentlemen for the opportunity to be with you
today.
Approved For Release 2001/08/09 : CIA-RDP74-0039OR000300440011-5
 Approved For Release 2001/08/09 : CIA-RDP74-00390R000300440011-5
Mr. Humphrey
Thank you Colonel Laedtke, I am glad for one that you took the
extra time. It was very interesting.
Ladies and Gentlemen we will now have a brief few moments for
questions and answers. Will you please pass your questions forward
and address them to the speaker that you would like to answer. Mr.
Devitt would you like to answer a couple of questions here?
Mr. Devitt
There are two questions here. One is - Some plastic reels have
cracked or broken on the inside of the hub. Is this caused by storage
conditions or improper mounting on the tape drive?
I think this could be caused by a couple of things. Also I think
that which are coming up, instead of 50-60 million characters of
information can now store in some of their components, hundreds and
hundreds of millions of characters. Some people suggest that we store
everything on these drums and not worry about tape. My personal feeling
is that tape is going to be here for a long time and possibly in increased
use.
Sure we can take your basic file. We can store it on a drum. We
can process from that drum. We need not go to the library and pull out
all these reels. But my feeling is, that if we get to that point in the
Pentagon, we will back up those basic files with a reserve data file for
security purposes. For history is another thing. Sometime you are going
to clean these hundreds and hundreds and millions of bits of information
and put new information on the tapes. So you will have need for historical
files. Tape will he with us-. You'll see it. If your office, your agency,
your activity is not mechanized today, rest assured one of these days it -
will be. That is the way the push is going. Thank you ladies and gentle-
men for these minutes with you.
Mr. Hutton
My question asks: "Are you working with the group at GSA which is
developing Federal Specs for computer tapes?" This is true, we have
representatives sitting on almost every committee that is involved in
this matter so that there can be a cross referencing of ideas as to what
will make a computer work the best. This is our primary reason for being
in the tape testing industry today. We are most interested in what will
give you the greatest degree of reliability and accuracy on the computer.
Approved For Release 2001/08/09 : CIA-RDP74-00390R000300440011-5
 Approved For Release 2001/08/09 : CIA-RDP74-00390R000300440011-5
I think this is a good group, this Federal Specs Committee and we do,
of course, have a man with this group. I'm sure that there is a 3M
man working with GSA developing Federal specs for computer tapes.
Mr. Humphrey
We are answering the questions in the order that the speaker had
in the program. Colonel would you take - there are two there I think.
Colonel Laedtke
The first question asks "Would you trust keeping very important
records for 50 years solely on magnetic tape or would you also keep the
printout?" It's difficult to give a brief answer to this question since
I'm not sure of the volume of printout that the gentleman was referring
to. If the volume was small and you had the space I would say there
would be no harm in keeping the printout. If such output products were
large and space consuming, I would suggest looking into the microfilming
possibility mentioned in my presentation. Then of course, if the records
were very important they would almost certainly continue to be retained
in magnetic tape form to facilitate possible re-use or recall in the
future. Today we really don't know how long our new heavy duty type tape
can be retained before it deteriorates .... might-be 50 years .... or it might
be longer. This is still one of the questionable areas in our business.
I have a little difficulty reading the second question, but I believe
.that the party is requesting information on how often should a tape be
checked for "fade or possible deterioration"? I believe the party is
possibly asking for information on how often you should check a reel of
tape for possible defects. Some data processing installations accomplish
this on a scheduled basis which requires each reel in the inventory, after
the lapse of a certain period of time, to go through a tape rehabilitation
cycle. During this rehabilitation the tape might be washed to remove all
loose foreigh matter and then processed through a certification machine
which will identify location and number of all defects on the reel at
that time. Such a program might call for the rehabilitation of each reel
on an annual basis....or it could be less frequent. Other data processing
installations, and I include ours in this group, do not have a specifically
scheduled program but accomplish a rehabilitation operation on an as
required basis. In our installation it works like this: Whenever a
computer operator detects what he believes is a defective tape he immediately
prepares what we call a."suspect ticket" and attaches it to the questionable
reel. In some cases a suspect reel will permit us to complete the particular
run that is on the computer, and in such a case the reel with the attached
suspect ticket will go to our rehabilitation section prior to its return
to the magnetic tape library. Because of the number of reels in our
inventory, and our somewhat limited rehabilitation capability, we make
no attempt to periodically check reels which are not in a suspect category.
Approved For Release 2001/08/09 : CIA-RDP74-00390R000300440011-5
 Approved For Release 2001/08/09 : CIA-RDP74-0039OR000300440011-5
Again, I say, the program is up to the particular installation. But all
installations should possess some type of quality checking capability.
I hope I have answered this question to someones satisfaction.
Mr. Humphrey
Thank you very much Colonel. I want to thank all of you gentlemen
for putting on our program today. It was mighty, mighty interesting.
Ladies and Gentlemen we stand adjourned.
Approved For Release 2001/08/09 : CIA-RDP74-0039OR000300440011-5