[Collins] kwm-2 fan and outputs

[Dr. Gerald N. Johnson]

While on active duty in about 1968 I worked on a 32S for the NSA MARS 
station. Needed finals, so I dropped over to supply and picked up a 
couple 6146 made by the Electronic Tube Company. The only thing that 
looked like a RCA 6146 was the base, the glass and the plate cap. The 
guts were a totally different design. In the 32S I couldn't set idle 
current, and they didn't make RF output. I made a trip to the radio 
store and got real RCA 6146 and they worked fine.

Warren Amfahr, W0WL, worked on the final of the KWM-1. He says because 
he was running a home brew mobile with 6146 that sounded good and Art 
heard it during Warren's daily commute. He told me that in the design 
process that the linearity wasn't as good as Collins desired, so they 
made a conference call with RCA tube designers about the 6146. RCA 
agreed to change the tube design for better linearity and call it the 
6146A. So by that the 6146A ought to be more linear than the 6146 and 
RCA ought to claim that as a positive. I've not found that claim. I've 
also not found any changes in CPN for PA tubes from 32S-1 to Rockwell 
32S-3B/C. Warren spent most of his working career in sales.

What I've been seeing until lately about 6146B neutralization is that 
they won't neutralize in the older transmitters with the 8-50 ceramic 
trimmer for PA neutralization. What you are saying is that they will but 
the neutralization drifts. The air trimmer certainly will handle much 
more RF current and having much less capacitance is more easily trimmed. 
Is it that the Q of the neutralization circuit is so high that with the 
8-50 the dip is missed? And so easier to hit with the bandspread of the 
air trimmer? Or is it that the circulating current is so high that the 
trimmer and some of the fixed capacitors including the sometimes 
troublesome feed through are drifting with the heat from that current?

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

On 10/11/2010 10:43 AM, Glen Zook wrote:
> The original RCA specification sheets on the 6146, 6146A/8298, and the 6146B/8298A show basically identical capacitances.  The dates on the sheets are May of 1952 for the 6146, May of 1963 for the 6146A/8298, and February of 1964 for the 6146B/8298A.  The differences in the capacitances being that in the original 6146 the input capacitance is 13.5 pf whereas the input capacitance of the 6146A/8298 and the 6146B/8298A is 13.0 pf.  Also, the grid to plate capacitance of the original 6146 and the 6146B/8298A is 2.0 pf whereas the grid to plate capacitance of the 6146A/8298 is 2.4 pf.
>
> The ICAS power ratings of the 6146 and 6146A/8298 are 90 watts input CW, 85 watts input for SSB, and 67.5 watts input for AM.  The power ratings of the 6146B/8298A are 120 watts input CW, 120 watts input for SSB, and 90 watts input for AM.
>
> Because of this, RCA originally said that the 6146B/8298A was completely backwards compatible with both the 6146 and 6146A/8298.  Because of that the military changed the specifications of the 6146W from those of the 6146A/8298 to those of the 6146B/8298A.  However, the nomenclature on the newer tubes was not changed to reflect the changes in the construction.  Depending on the tube manufacturer, since the change was done when a new contract was issued, the 6146W tubes produced as the 6146B/8298A equivalent happened sometime during the last half of 1964.  Therefore, any 6146W tube with a code date of June 1964 or before is going to be the 6146A/8298 equivalent and any 6146W with a code date of after January 1965 is going to be the 6146B/8298A equivalent.  Unfortunately, any 6146W with a code date of the last half of 1964 could be either version.
>
> Unfortunately, the "backwards" compatibility of the 6146B/8298A to the earlier tubes did not "pan out".  In many cases the 6146B/8298A is very unstable when used in place of the earlier versions.  Why, I have no idea!  However, there is something in probably the physical "layout" of the 6146B/8298A which contributes to this.  But, the specification sheets do not tell us the difference in the construction of the tubes.
>
> Often the 6146B/8298A tubes will not properly neutralize in transmitters designed for the earlier versions.  In some cases they will neutralize but, for some reason, the neutralization will not "hold" for very long.  Then, in other transmitters, the 6146B/8298A neutralizes fine and the tube works very well.  I have experienced situations with 3 transmitters of exactly the same type, with serial numbers very close to each other, in which the 6146B/8298A worked fine in 1 but would not work in the other 2!
>
> There are a few amateur radio operators who place the blame on the designers of the transmitter because the 6146B/8298A has problems in those transmitters.  Those people say that the designers should have "known better"!  I certainly disagree with this.  The transmitters were designed with the known performance of the 6146 and 6146A/8298 and they worked fine.  At the time the transmitters were designed the 6146B/8298A was unknown and therefore there was no way possible that the designers could have designed the circuits to work with an unknown quantity!
>
> There are those amateur radio operators who swear that they have had no problems at all replacing the earlier versions with the 6146B/8298A.  I believe them!  It may be how the component tolerances "add up" as to whether or not the newer tubes will work in the older equipment.  But, I am always VERY cautious when replacing the earlier versions with the 6146B/8298A.
>
> For those who have not read the article on the 6146 family of tubes which has been printed in several sources, go to
>
> http://k9sth.com/uploads/The_6146_Family_of_Tubes_1.pdf
>
> There is quite a bit information in the article which tells of the problems, history, and so forth of the 6146 family of tubes.
>
> As for getting more power out of the Collins S-Line and KWM-2- series equipment using the 6146B/8298A:  Frankly, the 516F-2 power supply was not designed to provide the higher current required for more output.  Therefore, I definitely do not recommend doing so.
>
> Glen, K9STH
>
> Website:  http://k9sth.com
>
>
> --- On Sun, 10/10/10, Dr. Gerald N. Johnson<geraldj at weather.net>  wrote:
>
> 6146B are decent tubes, just they have more capacitance than 6146 and 6146A so the neutralization circuit has to be changed. I would have thought there would be a service bulletin or service information letter about that, but I can't find any. Indeed the revision 7 Rockwell KWM-2 manual that lists many of the revisions doesn't admit the PA neutralization trimmer was ever changed. IT WAS!
>
> 6146B are rated at more output than 6146 or 6146A. I'm of the opinion that the Collins output tank (especially the load padders) can't carry the extra current when tuned to load the 6146B to full ratings and will have short lives. They are very hard to get to for replacement and not a common everyday computer part so are hard to purchase these days. For that reason, I suggest running the 6146B or W at 6146 ratings which will put the extra capability into a much greater tube longevity.
>
>
>
>