[ARC5] Collins 18S-4, RT-380/AR (AN/ARC-59)

[Mike Morrow]

Michael wrote:

> The Tx portion of the 18S-4 seems to draw its heritage from the T-412 right
> down to the paired channels although the 18S-4's pairs have to be within 2%
> of each other.

I don't think the "T-412" JAN designation would have been issued before the
mid-1950s.  The 18S-1 predecessor to the 18S-4 appeared in 1946.  It used the
modulation transformer, 811 modulators, and 813 PA of the AN/ART-13, but not
much else.

The 18S-4 and the RT-380/AR require the two channel pairs to be within 1 percent
(20 kHz at 2000 kHz, 185 kHz at 18500 kHz).  One wonders how useful this feature
was in real application.  OTOH, the AN/ART-13B specification of 3 percent seems
a bit liberal for two channels to share the same tuning...that's 543 kHz allowable
separation at 18100 kHz.  The internal antenna tuning of the AN/ART-13B into
a short non-resonant antenna would likely be non-optimal for at least one of the
paired channels.

BTW, the RT-380/AR is actually an 18S-4A, which has a mechanical IF filter that the
plain -4 model doesn't have.  Thirty years ago Fair Radio had RT-380s for reasonable
cost.  I wish I had one.  My 18S-4A is a commercial model, with all the in-service
crystals still present.

The odd thing about the 1953 18S-4 is that it was in service with many years of
overlap with the 144-channel 618S-1.  The mid-1950s 618S-1 really wasn't a
replacement for the 18S-4.  Unlike the 618S-1, the 18S-4 had its dynamotor supply
built in, which would simplify installation.  The antenna tuner (180K-3, CU-527/AR)
was external just like it was for later Collins sets.  And each of the ten 
channels were factory configured for the frequency of that particular channel.
It's a lot like the Bendix RTA-1B (AN/ARC-9) that had to have each of its ten channels
factory-configured for the frequency to be used.  I think that the 1942 RTA-1B was as
much an inspiration for the 1946 Collins 18S-1 as any earlier set, including having
the dynamotors built in, all in the same-sized 1-1/2 ATR case.  One could ask why
Collins didn't make a crystal-controlled commercial version of their AN/ARC-2.  But
that was really just a MF/HF command set in one box.  It wouldn't have made sense
to market something with the low power of the 8-channel AN/ARC-2 when the 100-watt
output 20-channel 18S-1 could be ready in the same size case a year after WWII ended.

> If PanAm designed the CDA-T,  I have to wonder if they - or someone - didn't also
> go to Collins to see about something that would do the same thing.

I don't think the time lines support that.  I don't know when the CDA-T appeared
(Robert??).  It doesn't seem to have shown up on military AN/ART-13 sets until
sometime before 1952.  The Collins 18S-1 showed up in 1946.  I seriously doubt
that the CDA-T had any influence on what Collins was designing.  So, maybe the
lines of inspiration run:

 1942 Bendix RTA-1B (AN/ARC-9) --> 1945 Collins AN/ARC-2 --> 1946 Collins 18S-1

Or maybe, just:

 1942 Bendix RTA-1B (AN/ARC-9) --> 1946 Collins 18S-1

Or maybe, there are no valid lines of inspiration between any of these items!

I think that the Bendix RTA-1B is unjustly overlooked when people track aircraft
radio development lines.  It was state-of-the-art when introduced in 1942.

Mike / KK5F
PS:  There are some interesting 18S-4 and RT-380/AR photos at:
     http://www.aviation-radio.com/collins18S4.html