[Amps] GS-35b heat sink

[Glen Zook]

I got an E-Mail from someone with photos of the both tubes, etc.
 
It does look like it would be very easy to make an adapter ring for the GS-31b to use the GS-35b heat sink.
 
Looking at the original Russian specification sheets for both tubes there are 3 areas which are different.  
 
The first is that the GS-35b has a mutual conductance of 25 to 40 whereas the GS-31b has a minimum mutual conductance of 22.
 
The heater current of the GS-35b is 2.65 to 3.25 amps and the GS-31b is rated at 3.1 to 3.7 amps.
 
The rated plate dissipation of the GS-35b is 1500 watts and the GS-31b is 1000 watts.  Most people seem to agree that this is due to the heat sink.
 
I don't see, in the specification sheets, that there is any real difference in the grid to plate capacitance.
 
The remaining parameters are, for all practical purposes, the same.
 
The input capacitance of the GS-35b is 18 to 24 pf and the input capacitance of the GS-31b is 19 to 24 pf.
 
The output capacitance of both tubes is 0.12 pf, etc.
 
I have 3 of the GS-31b tubes in hand.  Of those I am in the process of building a linear using a pair of them to cover 160-meters through 10-meters.  With about 3000 volts on the plate obtaining the maximum 1500 watts output allowed by the FCC would allow the tubes to basically "loaf"!  
 
The 3rd tube got me to looking at the remains of a Hallicrafters HT-33A that I acquired several months back.  The HT-33A actually has all of the circuitry of the HT-33B.  Unfortunately, someone had started converting the amplifier to include 160-meters and had taken loose quite a number of the parts.  In doing so, a number of those parts were badly damaged.  Also, the PL-172 tube was destroyed.  I do have another PL-172 but I am hesitant to even use it because the PL-172 tubes are notorious for being leaky and this tube has not been "lit" for quite a number of years.
 
Since the HT-33A is basically a "parts unit" it does lend itself to conversion.  Going to a capacitor input high voltage supply allows the filter choke, which is almost as big as the high voltage transformer, to be removed, the large 10 mfd oil filled capacitor to be removed, and the large filament transformer to be removed.  This will reduce the weight by 30%, maybe as much as 40%.  The plate heat sink of the PL-172 is exactly the same size as the GS-31b and only 5mm larger than the GS-35b and therefore the chimney will work fine with either tube.
 
Actually, I was not really intending on converting the GS-31b to the GS-35b heat sink.  However, the various discussions got me to wondering if doing the conversion was practical.  That is why I am considering getting one of the GS-35b heat sinks.  If the conversion works then I would probably use it in the HT-33A.  If not, then the relatively low cost would be worth the experience.
 
Glen, K9STH
 
Website:  http://k9sth.com
 
 --- On Fri, 5/20/11, paul at greenrover.demon.co.uk
<paul at greenrover.demon.co.uk> wrote:
 
Hi Glen, they are different tubes, the GS35 has quite a good machined taper fit whereas the GS31 exchanger sits on a stud. Either way whilst I'm sure you could adapt the heat exchangers to fit either way round it won't make a GS31 into a GS35.
  
Heater currents are different and grid/anode capacitance is different.
  
The trouble is that both incredibly robust, I sacraficed a couple in the early days and effectively destruction tested them. You can exceed the published ratings by a huge amount before they start to flashover. I think this has given some people the impression that because you can easily push a GS31 to give 2000W out that it must be the same as a GS35.
  
If you run enough HT, have enough drive and are willing to ignore the grid dissipation they just go on for ever.....
  
http://www.greenrover.demon.co.uk/50MHzamp.htm has some notes on the operating conditions of a couple of tubes at 50MHz. The same amplifier At 5Kv anode and 120W of drive produced 3Kw *without* over dissipating the tube. Intermod did not suffer significantly and was no worse than the drive source.