How hot can EL84s be biased?

[Mason]

I started a thread last week about an issue I was having with my newly built ac30. In that thread sluckey pointed out I was murdering my el84s with 136% dissipation. I’ve rebaised and now they’re at 97% dissipation. Unfortunately the tone has definitely suffered and now I’m wondering how much can I safely push the tubes? Obviously the tubes won’t last as long, but am I risking anything else if I push it to, say 110% dissipation? They’re new production tubes fwiw. I should note that this is also a cathode-biased amp.

Also, is there something about the ac30 circuit that sounds better with a hotter bias? My 18 watt Marshall build sounds great with the tubes around 95%. And other amps I’ve owned I’ve never felt the need to run a hot bias.

[jim]

Present El84 crop are very inexpensively produced tubes.  If you are planning to play out much failed tubes are inconvenient at the least.  You might want to go to NOS or the Russian EL84M which I have found to be very durable but HiFi sounding tubes.  If you are a basement player than go for it--the tubes are cheap.  Jim

[acheld]

I own several Blues Juniors.  Each one came from the factory biased at around 100% plate dissipation, IIRC.  In general, the EL84 output tubes would last several months of moderate use, and then crap out.

After installing a trim pot, and re-biasing down to 70% PD, the amps sound just as good, and I don't have to replace tubes.   I have re-biased several now down to 60%, and no difference in tone, at least to these old ears.

So, my guess is that biasing up past 100% is just asking for burned out tubes.

Interesting that you hear a tonal difference.  I'd be curious what exactly is happening there -- how and why.  What type of distortion you're hearing, etc.   And, is there a way to duplicate that without burning up your tubes.

[Mason]

Present El84 crop are very inexpensively produced tubes.  If you are planning to play out much failed tubes are inconvenient at the least.  You might want to go to NOS or the Russian EL84M which I have found to be very durable but HiFi sounding tubes.  If you are a basement player than go for it--the tubes are cheap.  Jim

The amp will get a few hours of use a week at home, but I will be pushing it hard during those few hours.

I own several Blues Juniors.  Each one came from the factory biased at around 100% plate dissipation, IIRC.  In general, the EL84 output tubes would last several months of moderate use, and then crap out.

After installing a trim pot, and re-biasing down to 70% PD, the amps sound just as good, and I don't have to replace tubes.   I have re-biased several now down to 60%, and no difference in tone, at least to these old ears.

So, my guess is that biasing up past 100% is just asking for burned out tubes.

Interesting that you hear a tonal difference.  I'd be curious what exactly is happening there -- how and why.  What type of distortion you're hearing, etc.   And, is there a way to duplicate that without burning up your tubes.

This amp is cathode bias so 100% PD is fine. And it is interesting because I've not noticed this kind of difference when messing with the bias of my other amps. I would say the sound feels a bit flatter, and there's not as much midrange content.

[thetragichero]

couldn't find an ac30 schematic with cathode voltage listed, but based on https://www.ampbooks.com/mobile/classic-circuits/vox-ac30/ they're biased right around where you originally were. they're known for eating up power tubes

[kagliostro]

Give a try to the russian 6P14P-EV


https://www.hobbielektronika.hu/forum/getfile.php?id=21127


Franco

[d95err]

If you are going to push the EL84s over the limit, it could be good to have some protection. Greg Fryer describes how he does it for Brian May’s amps here:

https://fryerguitars.com/el84-cathode-fusing-on-brian-may-ac30s/

[Latole]

If you are going to push the EL84s over the limit, it could be good to have some protection. Greg Fryer describes how he does it for Brian May’s amps here:

https://fryerguitars.com/el84-cathode-fusing-on-brian-may-ac30s/

Great, thank's to sharing.

[pdf64]

AC30 are cathode biased, Blues Juniors are fixed bias; it’s a world of difference!
With fixed bias, it can be moved over its range without affecting the dynamic conditions too greatly. Lower anode current will lower the stage gain and linearity to some degree but typically there’s a negative feedback loop that will counteract that to some degree. And fixed bias is almost never used at or near class A, so the range of bias that’s appropriate can’t be particularly wide anyway.
Whereas cathode bias HAS to idle close to or at class A; attempts to cool the bias too far into AB will result in dynamic conditions further cooling / affecting the bias, referred to as bias shift, ‘squish’ by Aiken.
The key point is that in the class B region of operation, average anode current will tend to rise, hence so will the bias voltage. ie cathode bias tends towards self regulation / limitation of anode current; as such it’s inappropriate for class B operation.
For a given anode load, as HT voltage rises, to maintain class A operation / x degrees conduction angle, idle anode current and hence dissipation must also rise; beyond a certain point, the anode limit will be exceeded. Hence with cathode bias, the HT supply voltage and sag characteristics are crucial, above a certain point and cathode bias becomes inappropriate.
It’s a big topic and a single post can’t be definitive, I’m just trying to provide a taste of the constraints and considerations involved.
Also of note is the valve type rating system; if it’s fine for 6L6GC to dissipate 30W, then under that rating system, it would probably be fine for EL84 to dissipate 14W.
And neither of those is the absolute max; if testing can be used to eliminate individual valves of below average ’toughness’, then the limit might be moved up to 17W. Which is where 70’s AC30s, with solid state rectification and a HT up around 360V, tended to idle.
RCA AN174 explains the design centre, design max and absolute max rating systems http://www.one-electron.com/Archives/RCA/RCA-AppNotes/RCA%201958%20AN-174%20Design-Maximum%20System%20for%20Rating%20Electron%20Tubes.pdf

Moving from a GZ34 to solid state rectification is a rather retrograde change on several fronts, eg increased HT voltage, reduced HT supply impedance, increased power up surge cathode current.

[PRR]

What voltage, what current, what load impedance?

It is too-too easy to over-volt happy operating conditions. Over-current restores system balance but cooks the tubes square.

[acheld]

Apologies for my uninformed comment.   Any time I see EL84 --  I think Blues Junior burning tubes.  It's a Pavlovian response!   

[pdf64]

Apologies for my uninformed comment.   Any time I see EL84 --  I think Blues Junior burning tubes.  It's a Pavlovian response!   

Yes, with them, increasing the magnitude of the bias voltage so that they idle a bit cooler is pretty much all benefit, no downside, no reason not to take action.

[sluckey]

An EL84 is just a hot natured tube. I have several EL84 pp amps, mostly cathode biased, but one that is fixed biased. All of them will melt your fingerprints even if only the filaments are lit.