Deville bias question

[arcman]

I picked up a 2002 HRD a few days ago that had distortion in the clean channel. Inspection found the 470 ohm screen grid resistors had been hot and one was open. Replaced both with higher wattage ratings and the amp sounded fine again.

But upon checking the bias it was over 100mV at the trimpot's lowest setting. The amp came with a set of 6l6EH's. I put a set of Mesa's in it and it was easily biased back to 68mV. I tried one each of the EH's and Mesa's and both times the voltage was way high.

Is it possible that the EH's are ok and can be brought into spec by making other changes to the circuit or should I just toss them? I see a couple of posts here mention some tubes are hard to bias and recommend removing the R83 resistor or substituting other values.

Thoughts?

Thanks.

[Leo_Gnardo]

But upon checking the bias it was over 100mV at the trimpot's lowest setting. The amp came with a set of 6l6EH's. - - - Is it possible that the EH's are ok and can be brought into spec by making other changes to the circuit or should I just toss them?

Just toss them sounds good to me. Yes you could alter the bias supply so that these EH's will settle into an acceptable bias zone, but if they're that out of whack they might continue to cause trouble. All in all I'm not much in favor of using EH 6V6's in amps that put more than say 350 volts on the plates. Same for TungSol (Russian) which appear to be EH with different paint markings.

Mesa? Those are re branded tubes. Not necessarily a bad thing, but Mesa does not make tubes. They have some outfit test them, match them, paint "Mesa" on the glass, then sell them at a rather high price reflecting the exclusivity of the Mesa product line. The idea projected to the consumer is that they are buying a superior product, and that notion is dubious at best. There are several other re branders around, the most famous is Groove Tube, also Ruby, Tube Amp Doctor (TAD), ARS, and in the past Audio Glassics & The New Tube (TNT).

The toughest of 'em all these days is JJ. Their 6V6's seem to put up well with high voltages.

Re scaling your bias supply also means when you do put in tubes that behave normally, you'll have to alter the bias supply again to accomodate them. What a hassle, we can do without that, right?

68 mV still seems high to me, that's the total bias current for both tubes so 34 mA per 6V6. With the usual DRRI hi voltage somewhere in the 425V zone, I'd say 20 to 25 mA per tube would be plenty.

Also welcome aboard!

[arcman]

I am aware Mesa does not make their own (like so many other rebranded tubes).

6l6's not 6v6's.

I guess that this is not the norm for EH 6l6's to act this way? In other posts it sounded like some tubes could be out of the standard range of the bias control.

Not a problem tossing them I just thought I should ask first.

Thanks for the quick reply.

[Leo_Gnardo]

I am aware Mesa does not make their own (like so many other rebranded tubes). 6l6's not 6v6's. I guess that this is not the norm for EH 6l6's to act this way? In other posts it sounded like some tubes could be out of the standard range of the bias control. Not a problem tossing them I just thought I should ask first.

Woops I misread, thought this was a RI deluxe. DUH! Now I get to wear the dunce cap, oh boy.

No, EH 6L6's are good tough tubes, generally they behave themselves. But if yours are zooming out of control, and it may be just one of the pair, I'd still be cautious and replace them. You could install one at a time, check that test point see if it's only one running way hot. Then mark that one or discard it, if the other one's reading something reasonable, keep it for a spare. Keep in mind, testing one at a time you're certain to find the bias current running higher than half of both at once, that's to be expected.

For 6L6's, 68 mV on the test point is parfectly good. That test point sums the cathode currents from both output tubes. Most of that is plate current, probably about 32-33 mA per 6L6, and a small slice for the screen grids, just 1 or 2 mA.

Vacuum tubes statistically show a sort of "bell curve" when tested for emissivity. Most are in the wide middle of that curve . Some re branders divide that zone into a number of slices. Before Fender bought Groove Tube, Groove numbered their power tubes 1 thru 10. To save money, the Fenderized GT now differentiates them by 3 colors, red white & blue. Mesa assigns more colors, ROYGBV plus grey. How to interpret these, who knows, they don't tell us, just put up your cash and buy 'em please. That's marketing for you. Other outfits label the tubes with some test data, emission current and sometimes gain factor too. Ruby, Antique/CE and TAD are good with this. Test conditions are different for each outfit, so the numbers from one company don't translate to another's. No matter. What you are asking about is tubes that are way off on the lower or higher "tails" of that bell curve. In days of olde, they would have been sent to the rubbish heap. Currently, somebody will sell them anyway, and they can be used, and even work reasonably well BUT you do have to alter your amp's bias circuit so rasonable emission current can be set. Then sometime later when you swap 'em out for tubes that are somewhere in the middle of that bell curve you may have to alter your bias supply again. So I'm not saying DON"T, just be prepared to spend more time faffing with your amp's innards if you go that route.

[Enzo]

Did you test each EH6L6 individually? All it takes is one bad one to make that bias test point voltage soar.

The burnt screen grid resistors is a clue the tubes were not healthy. Those burn up when tubes fail.

[arcman]

I guess I didn't do too good of a job explaining that I did try both tubes individually with one of the Mesa's and also another 6l6 I had lying around. So both EH's have the same issue.

I'll chuck 'em.

Thanks again for the informative reply.

Lee Hagen

[Enzo]

Fair enough. What I meant was using just ONE tube, leave the other socket empty. That way your test point can measure the current of one tube rather than a pair. I use that method more with quads of tubes, but it is still valid. I pick one socket then test each tube in it, that guarantees the same conditions for each tube. Then once I know the tubes are all similar, I pick one tube and try it in each socket, one by one. That tells me if the sockets have any differences.

[Randall]

Man, you gotta love this guy! ^^^ I have learned more cool stuff here from Enzo than I did in Electronics class 35 years ago.

[Mick Bailey]

FWIW I like to keep screen resistor wattage low and preferably use flameproof, fusible types. They're added insurance against a shorted tube causing damage, especially in cases where there's no HT fuse.

[arcman]

Enzo,

It didn't even cross my mind to try each tube alone in one socket. But, if after trying each EH alongside a known good tube and both times the voltage was above 100mV, I assumed they are pulling too much current.

Just out of curiosity, do you have any idea as to how much current these EH's are pulling vs a good tube at 60-60mV?

Also, any thoughts on why both EH's went this way? From what I've gleaned is the tubes are usually the culprit and then the screen grid resistors overheat and fail.

One last thing... Jeez, is this thing loud. From my limited electronics knowledge, the Deville has a audio taper volume pot on the clean channel and it goes from silent to deafening by 2. Would a linear pot tame this down so it has some kind of usable range?

Second last thing.....Weird that this actually had a pretty good distorted sound with the one failed screen resistor. Could you explain what happens to the signal in this situation?

Thanks again guys.

Lee

[Enzo]

First, tubes are not generic, meaning we cannot assume all EH6L6s draw this much current. I have no idea what to espect, maybe you simply have a hot set of tubes there. Yours are presumably idling at 50ma a tube, 100ma for two. That is part of testing them one by one to see if we have 50+50 or 60+40 or whatever. That ios more than the 30-35ma they are going for but not outrageous.

The test point is a 1 ohm resistor, so each millivolt represents a milliamp of current. 100mv is 100ma of current, shared by two tubes.

An open screen resistor makes the tube not conduct, might as well not be there. SO if you had an open screen resistor, your amp was running on one tube anyway. This is a typical classs AB amp, which means each side only goes into cutoff on upper regions of the wave. So until you get to a certain loudness there is no clipping. It sounds pretty clean. At louder levels, you do get into the class B part of AB, and it distorts, bu tnothing says you can't like teh resulting distortion.

[Leo_Gnardo]

the Deville has a audio taper volume pot on the clean channel and it goes from silent to deafening by 2. Would a linear pot tame this down so it has some kind of usable range?

No, just the opposite. Fender's thinking was the old "the amp that gets louder faster in the store showroom is the amp 'guitar johnny' will be more impressed with, and he'll buy that one." They might as well have put in linear pots.

You could select preamp tubes with lower gain, say 12AY7's for V1 and V2. Then you wouldn't be tempted to have to rip out the circuit board and modify. FWIW I've done mods on these amps too, to pad the volume, tame that gain.

Second last thing.....Weird that this actually had a pretty good distorted sound with the one failed screen resistor. Could you explain what happens to the signal in this situation?

With one form of distortion already present, it's hard to tell from the other form. And might add up to something your ears like. Consider one of the legends of distorted guitar being deliberately added to a recording. One of the Nashville session guitarists was worried when he knew his tone wasn't right at a late 50's record date. The producer & engineer pressed on anyway, then afterwards the muso found one of his output tubes had simply fallen out of its socket. When he was called in for another session with the same producer, his amp was working right - both tubes operating - but the producer told him "we were looking for that sound you got on the last session." Hmmm, I bet he said "OK boss!", pulled out a handkerchief, popped that tube out, got on with the recording and kept getting more studio calls.

Without that screen grid resistor (burnt open-circuit in place, same thing) your tube isn't adding anything, it might as well be out of its socket. If you applied a sine wave test signal to your amp's input, as we do on the test bench, you'd see only half a sine wave at the output, the other half effectively chopped off, a rather distorted waveform. Same would happen for the signal out of your guitar.

[Enzo]

I don't think he is going for less gain, I think he wants deafening to be up more like 6 or 7 on the knob rather than 2, so a different taper can address than. He wants useful range on the control instead of only being able to use 0-2 and it being real touchy as to setting.


Marketing: They want Johnny to think, "If it is this loud on 2, i bet it gets REAL loud all the way up." Johnny never thinks that at 2 it already is all the way up, essentially.

[arcman]

Installing one tube at a time showed 60 and 69mV. That's with the trimpot full CCW.

I did try two AT7's out of my other amps a few days ago and it did make the volume control a bit more usable. Are AU7's or AT7's just less gain or do they change the dynamics or tone in any way?

Enzo, Exactly, I would like to see 'loud' somewhere around 5 instead of 1.

Instead of buying more tubes, would a linear pot flatten out the rapid volume gain? Would a change of resistor value on the input stage do something similar?

[Justin Thomas]

Arcman,

Yes, the 12AT7s are "less gainy." In reality, they have a lower amplification factor. A 12AX7 can <potentially> amplify a signal up to 100 times, though usually it's much less, around 40-50. So, if you put 1V in, you could potentially get 100V out. The 12AT7 has a factor of 60, and an AU7 is about 20.

This is all very simplified, of course, nd the actual circuit the tube is in has a lot to do with how the tube will behave. So, yes, if you change nothing else in the circuit besides the tube, the tone may change, for better or worse. Neither the 12AT7 nor 12AU7 were intended to operate in the exact same circuit conditions as an AX7, and a better tube fit for an AX7 circuit would be a 12AY7, amplification around 40 or so. There are a few others, as well.

I build very simple amps, usually just two preamplifier tubes and 2 output tubes. I do find the AY7 to be better-sounding in the same circuit than either AT7 or AU7.

Again, all very simplified...