[Collins] Collins military S line

[Dr. Gerald N. Johnson]

On 7/9/2014 7:02 PM, Carl wrote:
>
>>
>> On 7/7/2014 7:39 PM, Carl wrote:
>>> If I had a full run of the tube manual into the 70's Id use a 6EH7 RF,
>>> 6ES8 as a Pullen mixer with AGC, 6DC6's in the IF and 6ES8's in the 2nd
>>> mixer and PD. Separate AGC loops would handle the different tube curves.
>>> This model would have the highest
>>
>> Many a radio has "delayed" AGC to the RF stage, essentially a voltage
>> divider so it sees less AGC voltage. Probably from use of remote
>> cutoff tubes in the IF stages and a semi remote or sharp cutoff for
>> the RF stage. Adding some threshold effect would enhance the AGC
>> qualities but wreck the linearity of the S-meter reading, just like
>> putting a sharp cutoff RF stage in most receivers.
>
> ** Collins uses a lot of sharp cutoff 6AK5's in the RF stage and the
> poor overload performance shows. Of course Hallicrafters followed suit
> with the 6CB6 and continued with the 6DC6.
> Some sharp cutoff pentodes can provide pseudo semi cutoff by varying
> screen voltage with plate current and reducing gain. Poor mans AGC
> without having to design an adequate circuit.

Trying to make up sensitive in spite of the extra noisy mixers.

It would take an added control tube to lower the screen voltage while 
lowering the plate current but then the control grid voltage could be 
left constant and with plate current directly proportional to screen 
voltage and pentode gain directly proportional to plate current the 
control range would be more linear than even a remote cut off pentode 
and likely considerably wider.
>
> The old saying was if you want a frequency meter buy a Collins; if you
> want to hear them buy a National. Others added that if you wanted an
> automatic band scanner buy a Hallicrafters.

Lots of DX has been worked with Collins so hearing wasn't all that bad. 
My dad bought us a Hallicrafters SX-96 before we passed the novice test 
in 1955. With a 50 kHz IF it had selectivity (choices from 1/2 to 5 kHz) 
but the two dial general coverage with band spread did have stability 
issues. It didn't exactly drift but even on a sturdy desk we didn't dare 
drop the pencil while making a CW contact, the impact would tune it far 
enough to loose the signal. Most of its problems were mechanical. We 
were able to improve it considerably. Fundamental problems included a 
too long coil mounting screw from under the chassis pushing up on the 
middle shield of the main tuning capacitor, the huge slots in the 
chassis for the dials, and a flexible cabinet bottom. We changed to a 
shorter screw, that might not have been in every radio shipped, we added 
braces past the bottom of the dials from chassis top to chassis front, 
and stiffened the bottom of the cabinet with a chunk of 1/8" steel 
sheet. Didn't look so nice but it did help the frequency stability.

One time in the 70s, I did better than 1ppm in an ARRL FMT with nothing 
more than my 75S-3B. Lots of care in setting the calibrator to WWV, then 
in setting the dial and I think a selected receiver with no end spread 
on the PTO, plus good slide rule eyeballing of the frequency. That was 
on 40 meters. So yes it was a good frequency meter.
>
>
>
>
>>>
>>> If I was stuck with tubes and circuits available during the different S
>>> Line production Id probably choose the 6BZ6 for RF, 12AT7's as straight
>>> mixers for early models then upgrade as better choices became available.
>>> Drake was smart to continually update the R4 series.
>>
>> I don't know about tube availability, but I suspect early post war (WW
>> II) designs were encouraged to use surplus tubes because they were
>> available or on hand in large quantities and very cheap. Heathkit got
>> into electronics that way with surplus 5BP1 CRTs and 6SN7s. I expect
>> Collins had considerable quantities of tubes used in the ARC2 and
>> other radios built during the war, which was a period of great
>> manufacturing expansion for Collins. I don't know that to be a fact,
>> though my first boss, Lloyd Winter (badge number 5) told me that in
>> the 50s they had to redesign all the early post war broadcast
>> transmitters because they had used surplus vernier reduction dials but
>> had used up all the supply and couldn't get them any more. I would
>> wonder if there wasn't a machinist in the place that could have made
>> them in the quantities needed or specials couldn't be bought. Probably
>> the cost was high. By then the style probably needed to be updated and
>> Art did like to see style changes, even when they negated his demands
>> for random S-line serial numbers. He didn't always keep his thoughts
>> connected from one week to the next it would appear.
>
> ** National built large quantities of the same receivers from before WW2
> with minimal changes. HRO, various variants of the NC-100, 101 and 200
> plus numerous strictly military designs of all octal tubes. The 1946
> NC-240D was part NC-200 and part new 240D until 1947 to 49 and all octal
> tubes as was the NC-173, NC-183, NC-57, etc.The HRO-60 was produced
> 1952-68 and used octals everywhere except the 4 miniatures in the front
> end and a 0A2. Ive no idea when they ran out of surplus tubes but they
> ran the octal since 1936. Hallicrafters ran octals into the 60's in some
> of their lower end sets.

Washington U (St. Louis) radio club in the early 60s had two receivers, 
an NC-200 and a HRO-5 (I think) with 2.5 volt tubes, not octals in the 
HRO. It wasn't horrible even on 10 meters where it was used with an 
ART-13 running on a dynamotor.
>
>
>>>
>>> In all cases the B+ would be no more than 150V except the audio amp
>>> which would be a 6V6 or 6BQ5/EL84.
>>
>> I have thought it would have been nice thermally to have had a couple
>> small (like 6AQ5 or 6AK6) tubes in pushpull so they could be run class
>> AB or class B and have a significantly lower idling current than the
>> single ended amps running class A.
>
> ** You dont wnt to run PP beam tetrodes or pentodes in Class B. It is
> AB1 for low distortion and AB2 for more peak power....and distortion as
> used in the 32V series, Johnson, Heathkit, and others. The 6AK6 was an
> obsolete pentode with little changed since the 41 of 1932.
>
>
>> Costs a tube and a socket, and an extra wire on the output transformer
>> plus s phase splitter and a couple extra coupling components.
>
> ** The phase splitter could also be the audio driver and a simple 6SJ7,
> it saves the cost of a 6SN7 and extra filament current. Just dont let
> the output stage draw grid current. There are comparable tubes right
> into Compactrons
>
> The audio quality and the heat dissipation
>> would have been detectably better by ear and hot finger.
>
> ** If you want heat grab a 6F6 in the early Super Pros, driver and PP
> output all inefficient triode connected. Or a 6L6 or 4 digit versions,
> and 5X4; all guaranteed to remove skin.

The 6L6 was more versatile than admitted in the tube manuals. Many a 
metal 6L6 ran a lot more power out at RF than rated when oil or water 
cooled. The curves for a 6L6 aren't a lot different from an 807, except 
that one is shown as constant current and the other as constant voltage.

In about 1960, I noticed the plate of a 6BG6GT sweep tube sure looked a 
lot like that of a 1625. During some boring class I converted the curves 
for the 6BG6GT to the same style as the 807 and they matched. I built a 
plug in coil AM transmitter that I used on 6m for a couple decades with 
a 6BG6GT. The main difference was that the 6BG6 didn't have the shield 
up top shielding the control grid from the plate cap wire.

Running some tests on the bench long ago, probably without a speaker, I 
noticed the screen of an octal based audio output tube running white 
hot, so I backed off the audio gain and the glow went away. The envelope 
was probably hot too, I didn't do a wet or dry finger temperature check.
>
> Carl
> KM1H
>
>
  73, Jerry, K0CQ, Technical Adviser to the Collins Radio Association