FREQUENCY CORRECTION FOR A AND C RACKS

• Document Type: CREST
• Collection: General CIA Records
• Document Number (FOIA) /ESDN (CREST): CIA-RDP81B00878R001400030095-5
• Release Decision: RIPPUB
• Original Classification: S
• Document Page Count: 2
• Document Creation Date: December 9, 2016
• Document Release Date: January 27, 2000
• Sequence Number: 95
• Publication Date: January 17, 1957
• Content Type: MF

Approved For l ease 2000/ - . IIA-RDP81 Bfl 878RO01400030095-5
SAPC-12436
Copy/ of $
MEMORANDUM FOR: Contracting Officer
SUBJECT : Frequency Correction for Afpnd C Racks
1. It has come to my attention that the reproducing system
used in racks A and C drops off rapidly below 200 cycles and very
rapidly below 120 cycles, and varies considerably throughout the
remainder of their frequency range. In our conversations of the
fidelity requirements for these systems, it was requested that
you provide 50-60,000 cycle band pass systems. An earlier
request for 20-60,000 was reduced to 50-60,000 at your suggestion.
At this time, you indicated that 75,000 was easily reached as an
upper limit and I spoke of the usefulness of such a range. It
is required that a 50-60,000 essentially flat response be retained
so that the results of System 1, 4 and 5 can be fully interpreted.
The present low frequency roll-off will result in a serious loss
of intercept data.
2. In addition, it will be a considerable chore to interpret
the data unless this band-pass is essentially flat. Since a
duplicate will have been run through two systems before interpretation
and unless one can say the total amplitude degradation is neglectible
one will need make correction for each operation in order to retain
useable power level data. The Ampex equipment that we have used
in the past has met this requirement. A typical Ampex recorder and
playback system response is f 2 db from 20 cycles to an upper
frequency set by the tape speed. Since we usually analyze from
50 cycles to an upper limit of 4000 cycles, the Ampex equipment is
much better than t 2 db. Typically we get better than x- 1 db in
this range and after two operations ( a dupe and playback) still
have amplitude data retained within 1 or at most ? 2 db.
3. It is perhaps worth mentioning as background information
that we have no use for the pulse shape as recorded by any tape
recorder since all collection systems degrades this data. Rather
we work with the magnitude of the fundamental component of the
pulse as a measure of the magnitude of the original power. For
this reason, it is most significant to retain in so far as possible
the amplitude of the fundamental.
Approved For Release 2006ad CIA-RDP81 B00878R001400030095-5
 Approved For Release 20 9 1 : CIA-RDP81 Bft878RO01400030095-5
2 - SAPC-12436
Copy % of
4. It is requested that the A and C racks be frequency
corrected so that they are as good as Ampex equipment in the
range of 50 to 4000 cycles. It is further requested that the
50 to 60,000 cycle requirement be met so that high speed duping
and analysis work up to 60,000 cycles, which is occasionally
required, can be carried out satisfactorily. The 75,000 cycle
upper limit that you suggested would be desirable.
25X1A
Approved For Release 2000/O 4! FlA-RDP81 B00878RO01400030095-5