Valliant II PS

[W9BHI]

Hello all,
I have just finished replacing all of the electrolytics and I am planning to replace the LV and bias rectifier tubes in my Valiant II with diodes.
I already have a pair of 1N2637's to replace the 866's and I know that the voltage increase is minimal on those.
I have read on this forum that the DC voltage will be quite a bit higher on the LV and bias supplies using diodes.
Does anybody know or remember what value and wattage of resistor to use on the output of the supplies to bring the voltage down closer to the original voltage?

Thanks,
Don W9BHI

[WA1GFZ]

I changed the filter on the low voltage supply to choke input in my V2. This dropped the voltage down to a safe level. I moved the first cap in parallel with th eoutput cap of the LV filter.

[WD5JKO]

I changed the filter on the low voltage supply to choke input in my V2. This dropped the voltage down to a safe level. I moved the first cap in parallel with th eoutput cap of the LV filter.

  I have done the same thing before with good results. The voltage ends up a little less than stock, but then combined with today's higher line voltage, the voltage comes out at about stock level.

   This approach does leave a couple of issues however. One is the diodes reverse recovery time which causes a brief (a micro-second or so) back emf pulse every half cycle coming from the power transformer. With a choke input filter setup, these pulses at a 120 hz rate go right through the L-C filter. Depending on the speech amplifier B+ decoupling scheme, you might get a nasty 120 HZ buzz in the audio that comes from the SS diodes.

   So I like to add to what Frank said, and follow the L-C choke input with an R-C before the B+ goes to any low level stage. The 'R' could be 100 to 300 ohms or so (current draw dependent), and the 'C' could be one of the original pi (C-L-C) capacitors, or double the 'C's' as mentioned and add a third 'C' which might need to get bypassed with something like a .1 uf polypropylene cap. Once done, those spikes are suppressed.

   The other issue is the higher line voltage typical of today which will result in the LV transformer running hot, and the filament voltage to the tubes might approach 7 volts. This issue should not be ignored. As has been mentioned here on this forum many times, controlling the ham bench line voltage is a good option. Alternatively in the Valiant, specific to the LV transformer, use the 5V winding to 'buck' 5V off the primary. This will throttle down the filaments, and lower the heat in the core from being near saturation. This also gets the LV B+ off the 5V winding, and this eliminates one failure mode of these old transformers.

Jim
WD5JKO

[WA1GFZ]

My LV transformer runs nice and cool with both 5 volt windings in series with th eprimary. My line is sometimes over 124VAC. The choke should isolate the trr diode noise but a small cap or snubber across each diode stack will absorb any niose it it is getting into the audio. I don't have that problem with 1N4007s. A fast diode could generate some hash.

[ke7trp]

The 2637s increase the B+ pretty good. I measured it last year after someone said that there was no increast in B+ going Solid state over 866s.  The peak out put power of the transmitter goes up nicely.

C

[W9BHI]

Thanks for the information.
I will install the diodes and see what the voltage is.

Don W9BHI

[The Slab Bacon]

Keep in mind that the MV rectifiers in Valiants were used for their current handling ability, NOT for their high voltage capacity. The HV in a valiant is only around 650-700v. This is not even enough voltage to arc over on a cold fire up. But the 500mills or so that it sucks on voice peaks is just way too much for standard high vacuum rectifiers like 5R4s

Why pay big bucks for the "drop in" 866 replacements when a couple of 5408s seriesed up under the chassis will do just fine for this application and they are $.50 (or less) each.

[ke7trp]

Frank, That is very true.  I like to keep things stock so the plug ins work for me. The one and only time I really scored at a ham fest was buying a STACK of 2637s. for cheap.  Real cheap.  I dont hoard crap so I took 2 each for my transmitters and gave the others way to the guys who wanted them. 

C

[The Slab Bacon]

Either way, if you solid state the MV rectifiers in a Valiant (or any other transmitter) The real gain is to increase the reliability. So, be sure to disconnect the 2.5v filament winding from the rectifier sockets. (Disconnect both leads from the transformer and insulate them off) This will eliminate the HV on the 2.5v winding of the LV power transformer giving you one less place for something to fail in a 50+ year old rig.

[Pete, WA2CWA]

When I solid-stated the rectifiers in my Valiant, I made a quadruple dc supply on a perf board, using the AC filament voltage, to drive a 12 DC muffin fan to exhaust heat generated from the tubes. Fabricated some metal brackets from some stuff I had kicking around in the junk box. Did the same thing to the Apache.

[WA1GFZ]

why not just use an ac fan and not load the transformer.
nothing quieter than an induction motor

[Pete, WA2CWA]

why not just use an ac fan and not load the transformer.
nothing quieter than an induction motor

I had the DC fans; I don't like AC wires running around the top side of the chassis. The 12 volt DC fans I have will run with as low as 9 volts. Can't even hear that they're running even at roughly 12 volts.

[W9BHI]

Hello all,
After doing some experimentation I have figured ot the proper resistor values to make the solid state rectifier output voltage about the same as the tubes.
For the bias rectifier (6BY5) I used two 1N5408 diodes in series along with a 160 ohm 10 watt wire wound resistor in each leg of the AC (pins 1 and .
For the 5V4 I used two 1N5408 diodes in series along with a 250 ohm 14 watt wire wound resistor in each leg of the AC (pins 4 and 6).
The voltage output of the SS rectifier assemblies was within 2 volt of what the tubes were.
The resistors are the square cinder block type from Mouser elec.
I hope this information is helpful to those wishing to SS your Valiant PS.

Don W9BHI

[W9BHI]

I noticed an interesting observation, the HV with the 3B38's was 680 volts.
I replaced them with the 1N2637 SS units and got 676 volts. 
I thought it would go up a little bit not down.
Must be more voltage drop in the diode stack that is being used.

Don W9BHI

[ke7trp]

Yes. The Diodes are close to the 3b28. However, There was around 100 volts different from 866.  If the rig is out of the case, Test the peak modulation power on scope or watt meter.  You gain a nice amount of mod capability.  I think this is because there is little sag.

C

[W9BHI]

Has anyone replaced the 0A2 in the vfo with a 150 volt zener?
If so, what wattage wound it be rated at?

Same goes for the two 0C3's, changing them to a pair of 105 volt zeners.
What wound thier wattage have to be?

Thanks for all of the helpful replies.

Don W9BHI

[WQ9E]

Don,

The 0A2 is rated for 30 mils and the 0C3 for 40 mils and the original circuit is set up with dropping resistors so that at minimum circuit load no more than 30 (40 for the 0C3) mils flows through the tube and at maximum load there would still be around 5 mils flowing through the tube to keep it "turned on".    You can use the same info to calculate the wattage that will be dissipated in a zener diode but allow for a large safety margin to allow for increased line voltage and any increase in voltage from power supply modifications you have made.

HOWEVER, VR tubes are pretty darn reliable and although they can fail it isn't a common occurrence.  I would leave well enough alone.  The VFO tube, keyer tube, audio tubes, entire exciter chain etc. is far more likely to fail than one of the VR tubes and you will probably replace dial lamps more often than VR tubes.

[Jim, W5JO]

The only thing I don't like about your solution is that you now have a 10 watt and a 14 watt heater under the hood.  This adds to the stress of other components including your chokes and transformers.  I would suggest if this is the approach you will stay with, then add a fan to pull the heat out of where you mounted the WW resistors as a couple of the guys suggested.

[W9BHI]

The bias supply uses the 10 watt resistors but they only get slightly warm.
I probably could have used 5 watters.
The 6BY6 gets way hotter than the resistors do.
The 14 watt resistors in the 300 volt supply do get pretty warm but not nearly as hot as the 5V4 gets.

Don W9BHI

[W9BHI]

I tried to find a calculator on line to figure out the correct wattage for the zener.
I know the input voltage and the zener voltage but I guess I need to know how much current the load is going to be.


Don W9BHI

[WQ9E]

Per my earlier reply, the 0A2 is rated for 30 mils and regulates at ~150 volts so it was rated for 4.5 watts, go to a higher wattage for the zener for safety.   The 105 volt octal 0C3 is rated for 40 mil service or a maximum of 4.2 watts.

Keep in mind that Zener diodes make excellent noise generators, the 0A2 circuit should be well shielded by the VFO enclosure bit if you do go ahead with this keep the possible noise generation in mind. 

For curiosity, why are you trying to replace the gaseous VR tubes?  They generate no more heat than a replacement zener diode since the heat generated in both is purely a function of how much current they must draw (and dissipate as heat) to maintain the desired voltage. 

[W9BHI]

I guess I have a bad case of tinkeritis and should just leave well enough alone

Don W9BHI