[Collins] 32S-1 grid problem

[Gerald]

On Fri, 2005-12-02 at 08:10 -0800, Glen Zook wrote:
> The random serial numbers did not really start on the
> S-Line until the S-3 series.  Art started the random
> serial numbers after the 75A4 receivers started
> becoming valued by the serial number.  However, this
> did not become apparent until after the S-1 / S-2 line
> had been in production for a while.  Therefore, the
> serial numbers on the S-1 and S-2 were sequential. 
> Remember that the S-1 / S-2 series was not
> manufactured that long when compared with the S-3
> series. The S-1 / S-2 series was introduced in late
> 1957 for the "1958 model year".

I'm not so sure about that. I have two 32S-1 manuals. One has SN 2968
penciled on the front cover. Its 7th edition. The schematic says
"current as of 8-24-59". The other has SN 1874 on the front cover. 4th
edition, schematic says, "current as of 2-19-59". I haven't detected any
differences between the schematics. Both are missing the dots in the
neutralization circuit. Digging deeper I see a number of changes in the
VOX circuitry, including adding a variable VOX time constant control. I
still don't see any wiring differences in the PA grid circuit. I do
notice that the PA grid voltage is listed as -7.3v DC in one and -58
volts DC in the other. Based on my recollection of 6146 characteristics,
neither seems right. The two schematics show different driver DC plate
voltages and PA screen voltage above the nominal 275 volts.

Comparing 32S-3 (7th edition) to 32S-1 (7th edition) in the PA grid
circuit. I see no wiring differences. There are different parts values
for C71 and C72 in the neutralization, R144 and C84. On the bias supply
side of the meter there are different resistor values, with the S-1
having a single 18K to bias supply and a 220K to ground from the supply
side of the meter shunt (1k). In the S-3, the 220K is absent and there
is an additional 4700 ohm resistor between the 18K and the bias supply.
The connection between the two resistors is bypassed with a 25 uf
capacitor. But since that capacitor is on the supply side it won't
affect the grid current reading directly to make that reading go
negative.

>  The S-3 started in,
> if I remember correctly, 1961.  The serial numbers on
> the KWM-2 series started being "randomized" just
> before the S-3 series came out if I remember
> correctly.  At least that is the "story" that I was
> told when I was with Collins Radio.

I heard that story when I was a Collins, and I confirmed the randomness
by looking up the component specification on the serial number tags and
it sure said to be issued in random number order. I didn't try to chase
the revision history of the specification. I don't know if I could have
easily. Of course by demanding new models, especially of the receivers
and then new company logos and color styles, Art negated a lot of the
randomness. He seems to have neglected to keep track of his prior edicts
when issuing new ones. But no one would argue that kind of detail with
him and have a job that afternoon. I thought his complaint about 75A4
serial number could have gone way back to fairly early A4 production as
dealers did have different NEW prices depending on SN already.
> 
> The problem is not in the tubes.  I have removed them
> from the sockets and the problem is still there. 
> Also, I replaced all of the paper capacitors (2 "black
> beauties" and several "green beauties") with orange
> drops since those are the most likely to cause
> problems.
> 
> I agree that the problem is in the grid circuit. 
> However, since the actual circuit is basically
> "unknown" since it differs considerably from that in
> the manual and it is difficult to trace.

With the information I have, I can't explain the odd grid circuit.
> 
> What I was hoping to find was someone that has the
> actual schematic since that would make it considerably
> easier to trouble shoot.
> 
> As far as the neutralization circuitry goes the
> earlier versions of the 32S-3 and 32S-3A use basically
> the same circuitry as the 32S-1 (ceramic trimmer). 
> Remember that the 6146B was not introduced by RCA
> until February of 1964 and the military did not change
> the 6146W from the 6146A equivalent to the 6146B
> equivalent until some time later in 1964 (different
> suppliers had their contracts changed at different
> times).  It was sometime in late 1964 or early 1965
> that the problems with using the 6146B arose and
> Collins had to change the neutralization circuitry in
> the S-Line and KWM-2 series to the air variable and
> make a field kit to modify those units already in the
> field.
> 
> If you are not fully familiar with the 6146 family of
> tubes I suggest that you go to either website that is
> listed at the end of this message where you will find
> the article on the 6146 family of tubes that was in
> Electric Radio a while back.  It also has been
> reprinted (with my permission) in several other
> publications concerning amateur radio over the past
> several years.

I've used 6146 for most of my ham life when using tubes, but not always.
I've tried to use 6146 from the Electronic Tube Company and they didn't
have characteristics close enough to RCA tubes to even set the idle
current in the rig I put them in. At that time (~1968) the only ones in
the federal supply that I was able to find were those from ETC so to fix
the radio I bought RCA outside the federal supply and they worked. I
suppose I should have kept the ETC tubes as collectible junk.
> 
> Glen, K9STH
> 
Something is going to ground on the tube side of the meter. It may take
circuit tracing to find the true schematic and the wiring error or bad
part that is leaking.
-- 
73, Jerry, K0CQ, Technical Advisor to the CRA
All content copyright Dr. Gerald N. Johnson, electrical engineer