[Collins] 32 s 1 self oscilation

Sun Apr 22 11:43:25 EDT 2012

In my original 32S-1 manual (pencil notation says for sn 2868 there is a 
table (4-1) of signal levels, probably measured with an HP 410B. On the 
schematic which it says is current as of 8-24-59 (Note 8) there are two 
voltages indicated on most control grids, the one in parentheses is AC 
or RF (note 9). The scan of the 10th edition 32S-1 manual on that other 
site is missing pages 11 and 12 where that signal level table would have 
been. The schematic as of 10-27-61 shows AC and RF voltages on grids. I 
wonder if the maintenance of the schematics and voltage charts have been 
the same through history. The voltage and resistance charge is not 
always perfection with major differences that should not show between 
the two 6146s and in 75S-x a failure to keep schematic and voltage 
charts up to date for power transformer and audio output tube bias 
changes. Some how the instruction manual department didn't detect all 
the MCN. When I revised the '64 75S-3B/C manual I presumed the tech 
writer took care of those. He didn't. Makes me wonder about other such 
details.

The 410B uses a diode in the probe but a high impedance circuit for a 
half wave rectifier, not the low impedance of a typical RF probe from 
ARRL handbook or Heathkit. Its peak reading calibrated in RMS. So the 
diode produces a voltage about the peak, and then the DC side of the AC 
scale tosses in a 0.7 multiplication factor. It would be more modern to 
measure those voltages with a 100 MHz scope, like a tek 465 or more modern.

In most tube mixers, low oscilllator drive results in low mixer gain. 
That phenomena is far less pronounced in diode ring mixers, like 
eventually used for the double balanced modulator in Collins transmitters.

In general for class A operation the positive peak grid voltage should 
be considerably less than the negative grid to cathode bias. E.g. if the 
grid is -10 volts with respect to cathode, the positive going RF drive 
peak at the grid should be probably not more than -5 volts to keep the 
tube very linear as expected in class A. For class AB1 likely the 
operation mode for the driver and I'm sure the designed operating mode 
for the PA, the peak RF voltage can never exceed the bias voltage, e.e. 
can't drive the grid positive to cause grid current because that clips 
the drive and introduces distortion. That's how the grid circuit ALC 
works detecting grid current from the RF drive exceeding the bias 
voltage and using that to reduce the gain of the gain controlled stage(s).

73, Jerry, K0CQ, Technical Adviser to the Collins Radio Association

On 4/22/2012 5:56 AM, David Knepper wrote:
> Never rule out, as Carl and Dr, Jerry indicate, low grid drive.
>
> I wonder if you two illustrious gentlemen could come up with the correct
> R.F. levels on the grids of say the two mixer circuits, the first and
> second drivers,
> and the final amplifiers in the KWM-2 and 32S transmitter stages.
>
> Weak drive can cause all kinds of problems as we all know.
>
> Great job, nevertheless, guys with this discussion.
>
> Thanks
>
> David Knepper, W3CRA/W3ST
> Collins Radio Association - Join Today
> www.collinsra.com
> Editor of the Collins Journal
> Secretary of the Collins Radio Association
> See us on Facebook: Collins Radio Association
>
> ----- Original Message ----- From: "Carl" <km1h at jeremy.mv.com>
> To: <geraldj at weather.net>; <collins at mailman.qth.net>
> Sent: Saturday, April 21, 2012 8:21 PM
> Subject: Re: [Collins] 32 s 1 self oscilation
>
>
>>
>> ----- Original Message ----- From: "Dr. Gerald N. Johnson"
>> <geraldj at weather.net>
>> To: <collins at mailman.qth.net>
>> Sent: Saturday, April 21, 2012 7:11 PM
>> Subject: Re: [Collins] 32 s 1 self oscilation
>>
>>
>>>
>>>
>>> On 4/21/2012 11:54 AM, Carl wrote:
>>>> long with the other suggestions Ive tamed an unruly S1 and others
>>>> with a
>>>> 22
>>>> Ohm 1/2W carbon resistor right at the grid pin of V-7 and a 100 Ohm
>>>> 1/2W
>>>> right at at Pin 8&3 with C58 at the tube side and C59 at the other and
>>>> both
>>>> with very short leads.
>>>
>>> Those can be most effective at preventing VHF parasitic oscillations at
>>> some possible cost of gain on 10 meters.
>>
>> That is only one of the benefits. It can also tame a stubrorn low
>> frequency
>> oscillation or a not perfect neutralization. Very common in many HF
>> circuits. Unless a circuit is already drive starved it shouldnt affect
>> 10M
>> at all.
>>
>>
>>
>>
>>>>
>>>> For the finals be sure R119/120 and R43/48 havent drifted high and as
>>>> long
>>>> as you are at it check the cathode resistors.Next install other 100 Ohm
>>>> 1/2W
>>>> carbon resistors at pin 3 of both tubes with added .01 discs at the far
>>>> end
>>>> as with V7.
>>>
>>> On my 32S-1 schematics, R43-48 ARE the cathode resistors of one of the
>>> 6146 with R49-54 the cathode resistors of the other 6146. I can't find R
>>> 119-120 on the schematic or parts list. I find them in the 32S-3
>>> schematic, part of added grid parasitic suppressors not shown in my
>>> 32S-1 manual or the 1959 edition of the yellow book.
>>
>>
>> I used the 32S1 schematic on collins.org
>>
>>
>>>>
>>>> Why Collins didnt use some of those basic stabilazation components
>>>> across
>>>> all S Lines is beyond me.
>>>
>>> You'll have to ask Warren Amfahr W0WL about that.
>>>>
>>>> Also dont overlook a gassy 6AH6 or 6CL6 which would likely be running
>>>> hotter
>>>> than normal.
>>>
>>> Also tubes with grid emission will counter act the fixed or ALC
>>> generated grid bias and draw more current and in pentodes the gain goes
>>> up as the plate current goes up.
>>
>>
>> No different in a receiver where it overides the AGC voltage and in
>> extreme
>> cases drives the grid positive with severe signal overload and sometimes
>> oscillation the result.
>> I see this in receivers, transmitters and transceivers on a regular
>> basis.
>>
>> Carl
>> KM1H
>>
>>
>>>>
>>> 73, Jerry, K0CQ, Technical Adviser to the Collins Radio Association
>>>>
>>>>
>>> ______________________________________________________________
>>> Collins mailing list
>>> Home: http://mailman.qth.net/mailman/listinfo/collins
>>> Help: http://mailman.qth.net/mmfaq.htm
>>> Post: mailto:Collins at mailman.qth.net
>>>
>>> This list hosted by: http://www.qsl.net
>>> Please help support this email list: http://www.qsl.net/donate.html
>>>
>>>
>>> -----
>>> No virus found in this message.
>>> Checked by AVG - www.avg.com
>>> Version: 10.0.1424 / Virus Database: 2411/4950 - Release Date: 04/21/12
>>>
>>
>> ______________________________________________________________
>> Collins mailing list
>> Home: http://mailman.qth.net/mailman/listinfo/collins
>> Help: http://mailman.qth.net/mmfaq.htm
>> Post: mailto:Collins at mailman.qth.net
>>
>> This list hosted by: http://www.qsl.net
>> Please help support this email list: http://www.qsl.net/donate.html
>
>