Cathode Bias EL34 - Output Wattage?

[pdf64]

[ATTACH=CONFIG]43543[/ATTACH]

Can anyone comment on the configuration described in the above attachment? It claims 40W, but those current figures seem kind of high to me.

Thanks

It happens that I've been trying a few months ago to get an old hifi amp up and running that looks to have used this as its model.
The big problem is that unless a very well regulated HT supply is used, to achieve 450Vdc at full load, HT sag will cause it to rise above 450 at idle, thereby pushing the idle dissipation up above 100% of plate rating. Barely get 30W output with good Winged Cs even then.
I've put it to one side for a while, with the plan to implement fixed bias when I go back to it.
The upside is that it's got an RS Deluxe OT

[Chuck H]

I thought the cathode resistance looked a little high for the current figure. But I didn't question it because I don't have anything like it on the bench to check. I do know that rapid rising screen current only gets worse with greater grid drive and can be a real problem. In fact you were part of my revelation on that matter, Pete. Increasing screen supply impedance/resistance and reducing grid drive to the sub meltdown level was the answer. That amp was a single ended, class A el34 in fact. I didn't see a voltage rise at idle though. That would be indicative of push/pull class bias of some kind for sure.?. Current decreases and voltage goes up.

jbefumo,

I wasn't necessarily saying you CAN'T run a hot bias in a small combo. If that's the tone you like and a small, uber powered combo is what you want then yes, the heat factor needs to be considered. That doesn't mean a hot bias or even true class A to 24 watts can't be done. But you may need to be creative with venting, a fan or something. Sans that, a cooler operating class helps mitigate the problem already. And if you've ever tried to fit a speaker into a crowded combo (as I have many times) you'll appreciate any room you can make.

[pdf64]

Yes, those Mullard operating conditions are definitely AB1.
PC case type fans, eg powered via the heater supply (when rectified & smoothed), can help in keeping combo temperatures reasonable.
A bit of air flow around the PT etc seems to go a long way.

[jbefumo]

Yeah, I considered a fan or two -- I have several on hand and they're amazingly quiet.

It took a bit of time but it finally dawned on me (I think) why cathode biasing should result in additional heat .... It's because they're typically biased way closer to the maximum dissipation, right? So if one were to bias them at 70% or so, like a typical fixed bias, then it would be a wash, but the peak power figure would be lower because of the negative feedback introduced by the cathode resistor ... ?

[jbefumo]

If my hypothesis (above) is correct, that would explain why the mighty 6550, in cathode bias, is making the same power as the EL34 figure I posted initially -- here's it's biased far more conservatively, and is consequently developing way less distortion, right?

That really fits my objective a lot better -- beyond just uber-power in small package, I'm trying to come up with something that old farts like myself can easily grab in one hand, with guitar case in the other, stow in a modern compact vehicle, and show up somewhere and be able to play along with a drummer without driving into extreme distortion (because there's no such thing as a 'clean pedal').

[Zozobra]

We've got to be careful to make sure we're comparing apples and apples here. The 6550 datasheet you posted states a 5k primary. I'd guess the EL34 example will be a lower primary (~3k5) as posted earlier?

[jbefumo]

correct. Trying to get this all correct before buying the transformer and then trying to make everything else work with it .... as usual ... ;^)

[Chuck H]

I used an Edcor XSE10 with a 4k primary for an el34 silver Champ conversion. Even at 4k I had to increase screen resistance and decrease grid drive to keep the screens from over dissipating when fully clipping. I might sacrifice a watt (or less) of efficiency and go with a 5k primary just to mitigate this problem. These are guitar amps after all so their operation has a slightly different set of requirements than what the data sheets account for. JM2C on that.

FWIW the Edcor XSE sounded fantastic. Obviously it would be better than the cheesy stock OT, but the guy I built this amp for calls it his little *umble. And that IS what it sounds like. The amp also got a new 10" speaker and a little re voicing, but the XSE OT was part of a winning package. The XSE is available in several primary and secondary options and I would definitely use it again.

[jbefumo]

I REALLY like Edcor iron. Quality is excellent and prices are hard to beat. I was just playing around in Multisim and thinking that 5k a-a would be the way to go.

Spent the past few days beating on my 5881 prototype (8k primary), and the sound is great, and no overheating problems so far. Attached is the full schematic with all the measured values included.JPB-24-VB.pdf

I used an Edcor XSE10 with a 4k primary for an el34 silver Champ conversion. Even at 4k I had to increase screen resistance and decrease grid drive to keep the screens from over dissipating when fully clipping. I might sacrifice a watt (or less) of efficiency and go with a 5k primary just to mitigate this problem. These are guitar amps after all so their operation has a slightly different set of requirements than what the data sheets account for. JM2C on that.

FWIW the Edcor XSE sounded fantastic. Obviously it would be better than the cheesy stock OT, but the guy I built this amp for calls it his little *umble. And that IS what it sounds like. The amp also got a new 10" speaker and a little re voicing, but the XSE OT was part of a winning package. The XSE is available in several primary and secondary options and I would definitely use it again.

[Chuck H]

Yep, the XSE is a single ended OT, so not what you need for this project, but I'm looking forward to trying their other OT's in the future. The 16xx off the shelf Hammonds (that I love) have become a bit pricey. FWIW I read on another audio forum where a guy tried the XSE clipped in side by side with two other OT's with the same specs, including another of Edcor's more expensive offerings and the XSE produced an extra watt over the others. That was for a 6W application so an extra watt would be a lot. Very efficient I suppose. That's what you want. An efficient transformer and an efficient speaker. It can make a lot of difference when you want to get louder.

[jbefumo]

Actually, I just double checked and the configuration I am looking out for EL34s uses 6.5k load:

[jbefumo]

I built a single-ended KT120 amp a while back using an Edcor OT (CXE series -- way heavy!) -- Worked out very well. Current owner absolutely loves it.