Dropping Voltage at PT

[MWJB]

Traynor & some Boogies run EL84 right at the top of the practical voltage range...it's normal for those designs but can have a bearing on tube brand choice.

I might have dreamed it, but I seem to recall someone (Lord Valve?) suggesting the Russian mil spec tubes can take 500vdc?

[ThermionicScott]

Voltage by itself isn't the problem, it's dissipation. There are EL84 designs out there that run the plates at or over 580V. (Of course, they're not cathode-biased. )

Looking at the Traynor schematics, the cathode-biased ones had a 250-ohm cathode resistor (rather than the usual 130-ohm), and the fixed-bias version used -16V. They're running hot, but probably not as hot as you'd think just given the plate voltages.

- Scott

[captntasty]

How far off would my calculations be if I don't figure in screen current to the equation? I know there isn't alot, buuut.... it might make a difference in this situation. Normally I wouldn't take it into account so the tubes would be biased conservatively..

[MWJB]

Don't sweat it, maybe a little out, may not, depends on current at idle, you shouldn't be setting plate current on such a knife edge that screen current makes it "Make or break".

[Chuck H]

JMHE, but... If your considering leaving the plate volts over 400, I wouldn't. EL84's are tough, sure. And Vp can be above max (whithin limits) as long as total dissapation is observed. But EL84's are kind of tender. While they may perform and sound good with very high Vp, and they really do actually, their life span can be shortened to weeks instead of years. I've had two EL84 amps that ran over 400Vp. as a caveate I should confess I ran both of them fully over driven much of the time. Biased modestly both amps would burn through the tubes in about eight to twelve weeks. My most recent design using EL84's has the tubes at 355Vp biased almost to class A and I get more than a year of service from the same tubes under the same playing conditions.

Again, JMHE, take it as you like.

Chuck

[captntasty]

That's the kicker... I wouldn't be setting plate current on a knife edge if I could avoid it. It seems this amp is just designed to be there. With the added AC voltage of modern supply it's being pushed past design specs - a hard one to balance.

[Steve Conner]

Chuck H: Does high plate voltage burn them up even if you lower the screen voltage? I think you could get away with running EL84s at 400+V on the plate, if you only put 250 on the screen.

If you put 400 on the screen too, then the poor tube ends up passing way more current than it's supposed to and the cathode wears out prematurely. It's the electronic equivalent of putting a brick on the gas pedal. My guess is that that's what's going on. But a low screen voltage would keep the peak cathode current under control.

[Chuck H]

+1 Steve. I never tried to lower the screens on either amp so I can't say for sure. But that makes sense. My big problem wasn't so much that the tubes would get tired. They would become very microphonic and rattle terribly. Probably also an effect of overall current as opposed to strictly high plate volts.

Chuck

[captntasty]

I'm thinking this would be a candidate for screen grid resistors - not a total fix, but might help a bit to prolong tube life... ? This particular amp is >400v on both plate and screen. It was designed this way but the "spec" (I use that term loosely) voltages with lower AC supply were just under 400 - 395v on plate, 390v on screen. Here's a link:

http://www.schematicheaven.com/bargainbin/traynor_bassmate_yba2b_manual.pdf

it's the 3rd/last schematic

would there be any benefit running it off the the next node in the power supply?

[jhelm_waterw]

Captntasty,
I just came out of my garage (working on my possessed yba-2) to my computer to find an answer for just this question.
Mine eats tubes and will not idle.
Getting desparate, I even pulled the phase splitter coupling caps to narrow down the issue.

I CANNOT get the tubes (tried 8 diff tubes) to bias down to a reasonable current (they seem to be in runaway mode or something)
I even installed a variable bias (down to -22 !!!!) and even cranked down to -22, they are running at %90.
Then... I realized the amp was running at 410V B+.
Even the schematic says 390V !
I thought maybe this circuit was meant for 7189s??? But that is not what the schematic says.
I also considered changing over to cathode bias.

Why would they design this SO close to limits that even a fluctiation in line voltage could catastrophically damage the amp???

I am ready to do a zener drop on the power supply... but can someone help me understand how this amp could have been designed so poorly?
(I wonder if it ever worked)

Thanks
JH

[captntasty]

Hey JH,
It would seem that in most guitar amp applications the el84's are typically run up to their limits. Makes for great sound but short tube life. I'll go back and find the exact stuff tonight, but taking into consideration that chose not to go with the zener (due to need to drill extra hole in chassis, heatsink, etc...) I reworked the power supply and tweaked things, gave it adjustable bias and found a happy compromise. Still, the tubes were going to have short lives compared to other types (6v6, 6l6gc, etc) - maybe 1 to 2 years depending on if it was being used often playing out or not as often as a home amp. The 7189's or new production EL84M (and their Russian equivalent 6P14P-ER or EV which could be had for less $$) could have been one compromise that would have given longer tube life - the client didn't want to go that route.

I'll post the details after I can fish out the file. Have you checked all components/replaced filter caps?

[jhelm_waterw]

Capn,
Thanks for response.
Short life? yeah.. in this case... 5 minutes !
I basically disconnected everything from the output stage by pulling the phase coupling caps.
I put in new filter caps a few years back when I took my first swipe at this amp (and shelved it due to it being possessed)
So... I have solid PS feeding the OT/Output Tube cicuit with the bias circuit fully stable and engaged (at a hefty -20 volts)
That's it ! There is no way I should have issues getting this thing down to idle.
But I did. Which leads me to the only conclusion that the 410 rail is just eating the
84's alive.
I, too, do not want to go to a "less than readily available" tube in the military 7189.
I am going to do one of two things...
a) put in the zener (already ordered a few) and drop it 30
b) do the inline 12v transformer drop (and put it inside the case)

One of those solutions should put me well within the EL84 "safe" operating region.

[captntasty]

Hmmm, even with B+ at 410Vdc 84's shouldn't fry in five minutes... when I got this amp I was seeing 430VDC at the plates on the output with bias v at ~ -18VDC. Worked out to 37ma(~15.5w plate dissipation) and 40ma(~17w plate dissipation) per tube. That didn't fry the tubes and it "worked" - not the optimum conditions, but... If you're frying tubes and bias current is runaway, I would think there are bum components. Check the polarity of caps? I would suggest listing voltages for all the tubes here and we might see something. Also, what are the 84's current measurements?

[jhelm_waterw]

captntasty,
Sorry for misunderstanding, I need to be more careful with my terminology.
I was seeing 40ma per tube. (similar to you)
That works out to be a nasty 16.4 watts @410+V (for a 12.5 watt tube)
When I say "frying", I mean the tubes were working and the amp sounded great but the plates started glowing red after a few minutes.
I was not comfortable letting it run like that shut it down before I saw smoke.
Maybe the amp could have operated for a bit longer but I was not taking chances.
If I let it cool and try it again, the tubes still do work (so they are not really fried)
but clearly this is an unhealthy operating region.
All my caps are solid.
I will remeasure my voltages and current and post back later.
Thanks

[tubeswell]

I've used zenering to great success on 3 amps to drop the B+. 5W 9V-15V zeners are all you need (if you are going to mount them on tag strips away from the chassis). Just run more than one in series off the CT of the HT winding to get a range of drops. All you are doing is dropping the peak current x the voltage of the zener - i.e.: (say for a 5F6A) 200mA* x 12V = 2.4W - so 5W rating is ample

(* the assumed max current draw rating of the HT winding)

However if you are going to use chassis bolt-on mount zeners, you will want a higher power handling, because they soak up the natural heat from the chassis anyhow