Need schematic and advice for fixing vintage AC30

[Mick Bailey]

Back when we made AC30s, 1N4007 was not around, we had a choice of Mullard BY127/BY133 or RS REC53A diodes. We didn't fit capacitors across the diodes as it serves no purpose in this design.

Were you on the design or production side? Presumably this was in the JMI days, as the 1N400x series was around in the mid-60s.

[Helmholtz]

Using AC30 schematic voltages the EL84 idle current is 50mA with a shared cathode resistor of 50R.
With a B+ of 330V this results in a plate dissipation of 16W or 133%.
Some models with a SS rectifier even have a B+ of 360V resulting in a plate dissipation of 155%! In these amp the EL84s are redplating at idle.
Consequently the tubes get very hot (and the top of the amp gets too hot to sit on ).

As a result the tubes don't last very long. They soon start to rattle and need to be replaced.

Nevertheless an AC30 is a class AB amp.

[Mick Bailey]

The EL84M holds up a lot better, but JJs aren't too bad. I see them as a consumable and the price to pay for getting that definitive sound. Modern tube sets can end up with one tube hogging the current in AC30s and I suppose the original matching is only good for so long at the high dissipation. I had a one-owner 63/64 transition model in for restoration that had all its original tubes from new. The EL84s had that aperture fogging and brown gravy stains on the inside, but were still working fine.

[Helmholtz]

The EL84M holds up a lot better, but JJs aren't too bad. I see them as a consumable and the price to pay for getting that definitive sound. Modern tube sets can end up with one tube hogging the current in AC30s and I suppose the original matching is only good for so long at the high dissipation. I had a one-owner 63/64 transition model in for restoration that had all its original tubes from new. The EL84s had that aperture fogging and brown gravy stains on the inside, but were still working fine.

So you verified that increasing to 68R (EL84 datasheet recommendation would be 130R/2 = 65R) noticeably changes sound?

The EL84s in my 1970 AC30 never lasted long before starting to rattle. This was in the 70s. Don't know about modern replacements.

[rockman627]

Vibrato switch jack had a wire disconnected. It's ok now. I'm only going to replace the 32/32uf cans and the big green ones. I think the 25/25 bypass caps too.
Plate voltage is 335 and 333 volts. Cathode is 10 volts. I'm going to get a 10 watt 68ohm resistor. I need to preserve tubes as much as possible.

[rockman627]

Hey I still have 2 new JJ el84s. If I order 2 more, how matched do they have to be since they are self biased?

[Helmholtz]

Hey I still have 2 new JJ el84s. If I order 2 more, how matched do they have to be since they are self biased?

I think that self-bias with a shared cathode resistor does not improve balance (rather the contrary) with mismatched tubes.

But just try and if there's not excessive hum or redplating it should be ok.

[Mick Bailey]

If it hums try changing the order of the tubes. Sometimes swapping them round gives a good improvement if they're not too far off. JJs can vary a fair bit when taking a random sample so you take a chance on what you'll get.

So you verified that increasing to 68R (EL84 datasheet recommendation would be 130R/2 = 65R) noticeably changes sound?

I didn't say anything about changing the cathode resistor

[The Dude]

I think what Helmholtz is getting at is that there is an idea being presented in this thread that you shouldn't mess with the original circuit even if tubes are biased at 133%.
Further: Has anyone verified that biasing at a more reasonable level "noticeably changes the sound" of the amp.

I tend to agree with his inference. I can't imagine that biasing past 100% gains you anything but red plating and shortened tube life. I don't see any reason to bias beyond normal biasing recommendations, which wouldn't even be 100%. Reliability must be factored in, too. I don't know about the rest of you, but I don't want my amp flaming out in the middle of a gig.

Edit: Oh, and sorry if I misread/misunderstood your point Helmholtz.

[Helmholtz]

I didn't say anything about changing the cathode resistor

Sorry, I misinterpreted.

[Greg Robinson]

I usually replace the stock 50Ω with 68Ω to rein in plate dissipation - always letting my clients know and offering to revert it if they prefer. I've never been able to hear a difference, and no one has ever asked me to change it back in many dozens of AC30's, so presumably no one else has either, or if they did, the trade-off of improved tube life was worth it to them.

[g1]

I'd be interested to know what the % dissipation was at the time the originals rolled off the assembly line. I'm guessing lower due to vintage line voltage, but I could be wrong.
I do see that the AC30CC2 had a switch to select either 'warm' or 'hot' bias. 82 or 53 ohms respectively. B+ voltage approx. 340V.

[Helmholtz]

I'd be interested to know what the % dissipation was at the time the originals rolled off the assembly line. I'm guessing lower due to vintage line voltage, but I could be wrong.
I do see that the AC30CC2 had a switch to select either 'warm' or 'hot' bias. 82 or 53 ohms respectively. B+ voltage approx. 340V.

I don't think actual UK mains voltage changed much since the 60s. UK guys?
If at all it would be lower today.

[catalin gramada]

https://www.voxac30.org.uk/

circuit diagrams on the bottom of page please.
Dallas 1974....? The sticker looks unconventional.

[Mick Bailey]

UK mains voltage is quite variable, depending on where you live. In the late 60s our house supply was 250v. Where I live now it's 248v most of the time (sometimes higher), though we get periods of lower voltage which with LED lights isn't so noticeable. With incandescents you really saw a dip in light output. We're supposed to be on the harmonised EU voltage of 230v +/- 5%, but that doesn't seem to be tightly applied.