SF Plate Voltage Reduction for GZ34

[lain]

I just A/B'd a SF Princeton Reverb with a RCA 5U4 and a Mullard GZ34 rectifier. With the 6V6's biased to 20ma, the Mullard is so much more fun to play. The plate voltages with the Mullard is 440 volts however, compared to 420 volts with the 5U4. I was reading in G Weber's book about reducing plate voltages with a zener diode on the B+ centertap. So, a few questions:

Has anyone done this?

Is it a legitimate/safe way to drop the plate voltage 20 or 30 volts?

What might be a good voltage to obtain for NOS RCA 6V6's? The schematic calls for 410 and that's way over the design max already.

Any problems with this method?

Any advice would be appreciated—that GZ34 sounds soooo good.

[Diablo]

Yes you can use a Zener to drop the B+ voltage. Lots of folks do that. However, the reason you like the sound with the GZ34 simply may be that you like the way amp sounds at higher voltage. If you lower the voltage with a Zener, it probably will sound like the amp does with the 5U4GB. The GZ34 will have less voltage sag under a load, so it acts closest to a solid state rectifier.

[lain]

Ahhhh

Less sag indeed, I keep forgetting. I did some tests last night. The 5U4 feels a little spongier, not bad with hotter bias though. I really like the GZ34 with a colder bias. I'm gonna try the zeners and see what happens. Weber's book recommends 56V 50W zeners. Does that seem right?

[JoeM]

I agree with Diablo. I dont see the point in using the GZ34 and then dropping the B+ back down to 5U4 voltages.

[MWJB]

What brand of 6V6 are you using? Before the modern hi voltage variants became available, biasing sturdy NOS 6V6s hot (30mA per tube) usually pulled the plate voltage down to 425vdc (410v would be usually unattainable with the factory specified rectifier). Modern 6V6s can take more voltage but not all can take a lot of current (JJ can). If you check the 65 RI schem at Fender.com, you'll see that realistic voltages are 440vdc @ 23mA with EH 6V6 (though some RIs seem to run a lot more than this & plate voltage needs checking) - this is ball park PR voltage.

Yes, the zener would work, but 20-30v drop might be more realistic (Gerald was referring to a Vibrolux if I remember right)...as Diablo & Joe have said, you say you like the sound of the higher voltage, so why drop it?

[lain]

Maybe I'm reading too much G Weber; I just read all four books—can't get enough. I'm using NOS matched RCA 6V6GT from KNOS tubes. I have a set of JJ's here as well. According to T Webers bias calc, 6V6's @ 440V and 70% plate dissipation should be biased at 19ma. Biasing to 23ma is 85% plate dissipation. How long would an old RCA last like this?

You mentioned biasing to 30ma. How is this possible. I'm know I'm a newbie, but the math seems incalculable.

I only considered dropping the voltages because Weber and Bruce (mission amps) both caution against GZ34's in SF Princetons. Weber says don't do it if the plates go over 430V, and Bruce says that PT's fail on those amps all the time at higher B+. Hands down though, the GZ34 adds something sweet, whether it's higher V or less sag, or both—I love it. I just thought that if I got the B+ down to 420, I could enjoy the GZ34 sag.

[pdf64]

How about lowering your line voltage, for one method see
Vintage Voltage Adapter
To check whether this would be an option, check your line voltage and heater voltage. Then work out what your heater voltage would be if you lost 6 / 12V from the line. As long as it's not much below 6V (GE spec 6.3V +/-10%, ie 5.7V minimum), then it's worth looking into.

[MWJB]

"Maybe I'm reading too much G Weber; I just read all four books—can't get enough. I'm using NOS matched RCA 6V6GT from KNOS tubes. I have a set of JJ's here as well. According to T Webers bias calc, 6V6's @ 440V and 70% plate dissipation should be biased at 19ma." Gerald and the late Ted are different people, the bias claculator is generic, you are muddling different approaches...I wouldn't run RCAs at 440v, you obviously haven't read all the G Weber, you''ll find that he recommends biasing up until plate voltage drops to 425vdc (usually happens at 30mA)...if you use JJs then there is no need to get the voltage any lower than 440v (they'll take 500v+). Fender run the EH 6V6 at 11.2W, RCAs at 425v will be around 12.75W (and take it)...the JJs will take considerably more than 70%, as much as the RCAs and at least as much as the EH...you can run the JJs where you like within that range (or colder/higher voltage if you still like the tone).

"You mentioned biasing to 30ma. How is this possible. I'm know I'm a newbie, but the math seems incalculable." It might be incalculable if 30mA @ 425v exceeded 14W but it doesn't (though 5E3s often run in excess of 14W cathode biased, wouldn't recommend it for fixed bias). Your 70% calculation is generic, it is a "rule of thumb", the Princeton sadly is unaware of this rule. The 70% "law" does not take into consideration the differing abilities of the tubes mentioned (...if they were all the same why buy different brands?) nor the fact that the amp simply cannot comply. What IS incalculable is running a RCA in a stock PR at 70% dissipation.

Feel free to drop the B+ with a zener, it's a good idea, but you probably don't want to drop to below 410vdc at a reasonable current draw (25-27mA). However, I still feel that if you like the sound now, why change it, it won't sound the same. The amp only exists to make a good sound, it's not there to adhere to a mathematical ideal.

[lain]

Great advice everyone, and thank you very much for the knowledge and ideas. I am definitely gonna build a vintage voltage adapter, just for the fun of it.

MWJB—Where does G Weber write about biasing till reaching acceptable voltage? I must have missed that very important idea. That helps a lot. The bias hit 28ma@425V. I guess I like the cleaner tone with 20ma@435V. I set it 24ma@430—it has a nice clean at 4 and a raucous bite at 9. The weird things is that after playing hard and taking the bias back and forth for about an hour, my matched set of RCA’s started performing very different from before. Maybe it has to do with burn in. One tube was biased at 21.5ma and the other at 26ma. They spread apart, and one tube was much hotter to the touch. For the last week they were running within 1-2ma of each other.

I’ll leave the B+ as it is, at least for a bit. My Allen OT came today and I’m gonna do the PI voltage and bias mods. I should probably do something with this reverb as well, it’s a bit noisy and gainy.

[MWJB]

"MWJB—Where does G Weber write about biasing till reaching acceptable voltage?" It's in the Q&A section of one of the books, maybe the last one? I think he mentions it in the section on the BF Deluxe in "A Desktop Guide..."

Not a lot of useful info can be got from touching your rectifier, power tubes (or the 12AT7 reverb driver in a Fender)...mostly you'll just learn that they are damn hot! Where did the RCAs come from? I would expect some drift over time, but 4mA isn't a major issue as long as you are happy with the fidelity of the amp (I would think a greater mismatch would be required to negate the value of having them).

[Bruce / Mission Amps]

What kind of flipping nut case would tell you to bias a power tube so that the plate voltage drops to some predetermined level!?
You NEVER use the idle current to lower plate voltage... that is crazy.
Nor do you ever use the plate load resistor to drop or increase B+ to a tube's plate... gads.

If you need to lower plate voltage then you do it at the PT or in the B+ rail BEFORE the output transformer or at one of the B+ dropping nodes.... do it in any way needed that doesn't use up too much power (heat).
The magic numbers for me, with respect to a good sounding PR, is B+ around 410-420 vdc max and the power tubes idling at about 9-10 watts each...

[lain]

Screw science

Upon ear testing—Cabernet influenced. An over biased Princeton does indeed sounds like "A thousand bees in a paper bag" which pops after the tubes come off of clipping and move toward normal functioning. Biasing down to 425 ma doesn't work here. The zeners are going in tomorrow.

[StratTone]

I agree with Diablo. I dont see the point in using the GZ34 and then dropping the B+ back down to 5U4 voltages.

I agree with Diablo and JoeM. Been down this road with my super and played with the voltages and didnt see the point of all the changes I made when all I had to do was change the tube and rebias. I ended up with a blackface spec board but with 5U4 and left the doghouse stock.

I'm actually thinking about putting it back to Silverface specs at the moment. The amp for sure has the blackface grind but is still kind of thin sounding. The trannies in the '74 range supers just don't seem to compare to the mid to late 60s no matter how much I try and kid myself. haha