Mixing brands of output tubes

[GlennW]

An amp that I'm working on uses 4 6L6GC output tubes. I have two pairs of tubes made by different manufacturers - they're supposed to be matched pairs, but not to each other. I don't have a tube tester.

Should I put the pair of Brand A in the inside sockets and the pair of Brand B in the outside sockets, or Brand A in the left two sockets and Brand B in the right two sockets?

I plan on getting a matched quad, but want to wait until the amp looks like it's ok.

Thanks.

[MWJB]

First you need to identify how much current they are drawing in the amp, using bias probes or by installing 1ohm resistors between the power tube cathodes & ground.

What I would do is install Brand A, one tube in ONE of the left 2 sockets, the other tube in ONE of the right hand sockets, lets put them both in the outer 2 sockets, (double your speaker load in ohms if you are concerned about an impedance mismatch) record the bias current of the 2 brand A tubes.

Now replace Brand A with brand B, what's the difference in current? If you can bias the amp so that neither brand is running more than 35mA, or less than 25mA WITH ONE OF EACH BRAND in the left 2 sockets, and ONE EACH of each brand in the RIGHT 2 sockets you should be able to evaluate the tone of the amp, at least to the point of establishing that it functions OK. If for instance brand A runs at 35mA and brand B runs at 15mA this will dertrimentally affect the tone.

Don't run both brand A on one side and brand B on the other if there is a mismatch of more than 5mA between the brands, this will also affect the tone.

There probably aren't that many different manufacture tubes that will bias up close enough to do this sucessfully. Are you sure that they are from different manufacturers, rather than just different "brands/labels" (even if they were different grades from the SAME manufacturer this could still be a long shot)?

This isn't a perfect scenario and it would really be easier to buy a matched quad, even if you don't use them on this amp you'll be able to on something else (e.g. 2 new tubes for a 2 output tube amp, plus a spare pair that you can drop in knowing that they'll bias up OK).

Don't fit any of them without checking the bias.

[Enzo]

Assuming your unidentified amp is like most, tubes 1-2-3-4 will be working like this 1 and 2 work on the positive side of the waveform, and 3 and 4 on the negative. Or exactly the reverse depending on which way we are looking at the amp.

The point of balancing the tubes by matching them is that both the positive and negative halves of the waveform get more or less equal treatment. So we want the pair of tubes in 1 and 2 to be essentially the same as 3 and 4. With two matched pairs that don't match each other, then you want to split each pair across the divide. So with pairs AA and BB, we want to wind uo ABBA, or BAAB. And really, since the sockets are all in parallel, ABAB, and BABA would also work. In all those cases there is an A and a B on each side.

Lots of amp makers don't bother with matching the tubes, and we tend to make a larger deal out of it than it merits, so random tubes won't harm the amp or anything.

But if you want the most out of your tubes, if AA are a lot hotter than BB or something, then setting the bias so it makes sense for both pairs of tubes might be tough. Set the bias for the hot pair, and let the other pair run cool, I guess. Just make sure none of them are red plating and you are close enough.

On the other hand, AA-BB won't hurt anything either, and if the tubes are not very similar in performance, maybe the imbalance will sound appealing to your ear. FUnky.

[Enzo]

Geez, it is 6:30AM and there are TWO of us trying to answer this.

[MWJB]

There's a 6:30 in the morning now? I'll take your word for it, it'd take some sort of early morning cataclysmic disaster for me to witness that!:-)

It's just gone midday here in the UK.

[GlennW]

Thanks a lot for the detailed responses.

It's an Earth Sound Research B-2000, I've read that it's very similar to the Fender Bassman 100. The pots date it to 1974.

http://www.schematicheaven.com/barga..._superbass.pdf

I've never biased an amp before, and I don't have the resistor mentioned earlier. I've tried to learn, but to be honest I get confused on the various types of biasing. I'll try to learn. I did guesstimate the centering the bias pot.

These tubes are all gems from ebay:
One pair of Svetlana? Made in Russia, 30 1610, 27 1560 are written on the tops
One pair Westinghouse, England, 82, 83 written on them
One (not pair) GE that came with the amp (ebay), the other 3 make a noise when shaken (not stirred) sorry...
3 Sylvanias that came another old amp

I'll try it tonight if some parts are delivered today. I'm not going to run it very long, just enough to see if it'll work.

I'm not sure of the impedance, but have been told it's 4 ohms, with paralleled speaker jacks. When I was working on it yesterday I noticed an unused white wire coming from the OT. I'd like to find what that is, maybe an 8 ohm tap. I found a couple of articles through Google on how to check that, but I'll try to do that later.

[MWJB]

Your amp appears to have been conveniently fitted with 1ohm resistors (brown/black/gold), you can easily read the bias with a voltmeter (set to <200mV) at pin 8 (to ground) of the 6L6 socket. Reading will be the value for 2tubes (either the 2 left or the 2 right) so matched tubes are a very good idea, as you won't know what each individual tube is drawing.

Centering the bias pot guarantees nothing. I would be inclined to turn the pot until you get the most negative voltage at pin 5 of the 6L6s, with no tubes installed, before trying to install the tubes and set the bias.

Are any of these tubes supplied with any kind of warranty? Are they "new in box" & tested? If not, you may be making a rod for your own back, a duff tube can easily blow a screen grid resistor then you're just making more work for yourself. If you have doubts about the performance of the amp it would make sense to fit known good tubes just to eliminate tubes from the possible things that could be wrong with the amp.

Current production 6L6s are not expensive.

[Satamax]

Originally Posted by Enzo

Assuming your unidentified amp is like most, tubes 1-2-3-4 will be working like this 1 and 2 work on the positive side of the waveform, and 3 and 4 on the negative. Or exactly the reverse depending on which way we are looking at the amp.

The point of balancing the tubes by matching them is that both the positive and negative halves of the waveform get more or less equal treatment. So we want the pair of tubes in 1 and 2 to be essentially the same as 3 and 4. With two matched pairs that don't match each other, then you want to split each pair across the divide. So with pairs AA and BB, we want to wind uo ABBA, or BAAB. And really, since the sockets are all in parallel, ABAB, and BABA would also work. In all those cases there is an A and a B on each side.

Lots of amp makers don't bother with matching the tubes, and we tend to make a larger deal out of it than it merits, so random tubes won't harm the amp or anything.

But if you want the most out of your tubes, if AA are a lot hotter than BB or something, then setting the bias so it makes sense for both pairs of tubes might be tough. Set the bias for the hot pair, and let the other pair run cool, I guess. Just make sure none of them are red plating and you are close enough.

On the other hand, AA-BB won't hurt anything either, and if the tubes are not very similar in performance, maybe the imbalance will sound appealing to your ear. FUnky.

Hi everybody!

Enzo, with adjustable bias on each side, wouldn't it be better to make them AA on one side, BB on the other and adjust the bias for the same dissipation on each side? Ok i understand that would may be change the relative balance of the two sides, but i'm pretty daft and don't understand much in tubes yet!

[Mars Amp Repair]

'ABBA'... 'Enzo-Fernando!!'...I think I'd avoid that config g

[bob p]

Originally Posted by Mars Amp Repair

'ABBA'... 'Enzo-Fernando!!'...I think I'd avoid that config g

Regarding the ABBA, BAAB idea suggested earlier, i guess it would be good to ask the OP what circuit amp he's working with. although the ABBA / BAAB rule is probably going to work 99.9% of the time, there are some relatively common amps out there that won't follow the formula.

all of the simulclass Mesa amps, for example, intentionally violate the rule by not wiring the output tube pairs in parallel. so it would be good to double check the schematic of the amp before making the final decision on tube placement.

hth

[GlennW]

Thanks again for all the help. I haven't tried to learn anything about working on amps since before the website changed. It's not a lack of interest, it's a matter of time.

A couple days ago I tried to get it working after having done nothing with it for a year or so and noticed that the two electrolytic caps which are connected to the bias pot were not installed the same way. One had + to ground, and the other had - to ground. After putting in a new cap so both are + to ground the output tubes aren't glowing reddish orange as they were before.

[Enzo]

Enzo, with adjustable bias on each side, wouldn't it be better to make them AA on one side, BB on the other and adjust the bias for the same dissipation on each side? Ok i understand that would may be change the relative balance of the two sides, but i'm pretty daft and don't understand much in tubes yet!

Well, I missed the part where he said he has individual bias settings. ( he didn't) The question seemed generic, so I don't assume more complex factors.

And he did mention it was an Earth head. Real basic.

all of the simulclass Mesa amps, for example, intentionally violate the rule by not wiring the output tube pairs in parallel.

But wouldn't the ABBA/BAAB arrangement still be appropriate there? I would agree that ABAB/BABA would no longer be valid then. But the outer pair would be matched and the inner pair would be matched.

[bob p]

Originally Posted by Enzo

But wouldn't the ABBA/BAAB arrangement still be appropriate there? I would agree that ABAB/BABA would no longer be valid then. But the outer pair would be matched and the inner pair would be matched.

i guess i should have clarified things a bit more, so that what i said would actually have been right. the ABBA / BAAB rule will work in the mesa amps, but depending on your tubes they might do better in the A slot or the B slot, depending upon how each pair is matched with respect to the other pair.

what i should have said is that in the mesa simulclass amps, what's actually going on is that the inner and outer pairs of tubes are actually wired in parallel in an ABBA fashion, so the ABBA rule works. the only thing to watch out for is that the inner and outer pairs are biased differently, and the bias is non-adjustable. when someone buys a mesa amp they're supposed to buy spec'd tubes for the amp. if you're not doing that, any given pair of tubes might work out better in the A slot ("Class A") than the B slot (Class AB), or vise versa. that's one of the downsides of mesa using non-adjustable fixed bias in their amps, you have to find the "right" tubes for the application, or convert the amp to adjustable bias. converting an amp like the Mk IV to have 2 adjustable bias circuits is a bit of a PITA because of the layout.

sorry for leading us off on a tangent. i promise to be quiet next time.