What are the Heater Current Limits on big Fender amps?

[bob p]

I'm thinking about changing the output tube compliment on a Fender Super Twin, but before going there I thought I'd ask if other people have experience in using tubes with high heater current demands in the big Fender amps.

Unfortunately, this amp runs the B+ voltage at about 530 VDC which is really hard on a 6L6. I've been thinking about switching over to a tube that's more tolerant of the high plate voltages, but the problem is that they all draw a lot more heater current. And though the difference may be negligible if you're running a pair of tubes, the Super Twin is running 3 pair of tubes, and I'm worried that the excess heater current requirement will add up to trouble.

For example, the 6L6 is spec'd for a 0.9A heater current rating. In an amp using a pair of tubes, that's 1.8A. Moving up to something like 6550 would take the heater current requirement up to 1.6 * 2 = 3.2A. That's quite a jump, but I've heard that people have gotten away with doing it.

Compare that to the Super Twin, which uses a sextet, so it's PT is delivering about 0.9 * 6 = 5.4A from the heater windings. Switching over to something like 6550 is going to put the heater demand in the range of 1.6A * 6 = 9.6A. That's probably enough beyond the amp's 5.4 A spec that the extra 4.2A is going to cause a problem.

So I thought I'd ask if anyone's ever approached the problem with Fenders running either a pair, a quad or a sextet of higher heater current tubes and how it worked out for you. I think you could probably get by doing this with a pair of tubes, but I think that the heater demands of the sextet are just going to cause a heap of trouble if I don't split the heater load and add a supplementary transformer.

TIA.

[Tom Phillips]

...I thought I'd ask if anyone's ever approached the problem with Fenders running either a pair, a quad or a sextet of higher heater current tubes and how it worked out for you...

Many years ago I built a couple of amps using stock Fender Showman/Twin Reverb PTs and a quad of high current power tubes. (One with 6550s and one with KT-88s) I know, of course, that the PTs were under more stress and would have run hotter than with 6L6 power tubes. However, in response to your question, I can tell you that I kept in touch with one of the amps for about 30 years and there was no PT problem during that time. That's all the personal experience I can offer.
Cheers,
Tom

[Enzo]

And the simple answer is this: find out. The base pinouts are the same, so plug the 6550s or whatever into the amp and power it up. Leave it in standby. Measure the heater voltage. If it is still up to a useful level, then the transformer can handle it. If it sags it to 5.8v or something, it is too much to handle.

[bob p]

And the simple answer is this: find out.

That answer might have been helpful if I had a sextet of 6550 lying around for experimentation. But consider this -- If I had the sextet of 6550 lying around, would I have needed to ask the question?

I agree that the simple answer is to find out. So I'm finding out by asking.

[R.G.]

This question reduces to " how much power can my power transformer put out?"

The answer is "until it overheats or until the voltage sags too much". Unfortunately, that's a hard question to answer, because "too hot" varies depending on the power transformer's internal insulation ratings.

The real way to find out is to run the transformer at the higher current load, pausing every 15 minutes or so to disconnect the leads and measure their resistance. Copper's resistance increases by 0.393% for every 1C its temperature rises. So if you measure the resistance of each winding at room temperature before it is powered up, then every few minutes under load, you can actually plot the internal temperature by computing the hot resistance divided by the cold resistance and dividing by the % per degree C. This tells you quite accurately what the internal temperature is.

Most commercial transformers have insulation rated to 105C, so you take the temperature rise at full load above your cold starting point and add to it the rise above your ambient that you think stages will get to - 38C is not a bad target. If that's below the insulation rating or 105C if it's not listed on the transformer, it won't overheat.

If the voltage sags too much, that's another matter.

In any case, playing safe is putting another transformer in for just the added filament current.

[Enzo]

Don't have 6550s? You want to know if the transformer can supply 4.2 more amps than 6L6s draw. Clip a 1.5 ohm resistor in parallel with the heaters. That will draw 4.2 more amperes. That is slightly less than 10 watts.

[Mick Bailey]

Would you not want to lower the B+ and stick with 6L6s?

[bob p]

The real way to find out is to run the transformer at the higher current load, pausing every 15 minutes or so to disconnect the leads and measure their resistance. Copper's resistance increases by 0.393% for every 1C its temperature rises. So if you measure the resistance of each winding at room temperature before it is powered up, then every few minutes under load, you can actually plot the internal temperature by computing the hot resistance divided by the cold resistance and dividing by the % per degree C. This tells you quite accurately what the internal temperature is.

Now that's a great idea. You're right -- internal temperature is going to ultimately be what decides whether or not my transformer enters a failure mode, and that's a good way to find out the answer that I really need to know. IMO that type of answer is a lot more valuable than just asking how much the voltage will sag when the transformer gets overloaded with an abusive load.

I haven't ever measured the resistance of leads on a PT before, but I'm thinking that any PT that's worth it's salt would have to be a low-Z supply, which means I'd probably need a 4-wire meter to take accurate resistance measurements. Or can this be done with the garden variety 2 wire meter?

[bob p]

... but I think that the heater demands of the sextet are just going to cause a heap of trouble if I don't split the heater load and add a supplementary transformer.

This question reduces to " how much power can my power transformer put out?"
...
In any case, playing safe is putting another transformer in for just the added filament current.

Thanks for your insights, RG. Being a 6x6L6 amp, it doesn't use a commonly available power transformer. Actually, I'm not sure who might make a suitable replacement. I know that Hammonds are out, and that Fender no longer offers this PT in their supply chain -- I checked. I'm not sure, but maybe Mesa might offer something different, but capable, now that they're getting back into making big tube bass amps.

Because this amp doesn't use a garden variety transformer, I'm very reluctant to hack at the project by just piling on load, checking for sag, checking temperatures, and hoping for the best. I'm thinking that the best answer to this problem is going to be to use a supplementary filament transformer rather than abusing the original transformer that might be impossible to replace. I'm thinking that my best bet is going to be to split up the heater loads, perhaps powering 3x6650 off of the stock PT, and 3x6550 off of a 4.5A add-on filament transformer.