biasing to 100% with cathode bias...your opinion?

[j-po]

...Theoretically, 100% dissipation is within the tube's limits, so it should be fine...

100% at idle is not fine! What happens when you break out of "class A"? You exceed 100% dissipation. Just because the plate is not glowing red does not mean it is within limits.

[Alexo]

It's pretty normal to exceed maximum plate dissipation over the course of a big signal, especially when it enters the class B portion. At least in guitar amps. As long as the average dissipation over a full signal swing is under 100%, it's usually fine.

Besides that, the bias setting doesn't affect what happens when you enter class B operation, only how soon you switch over to it.

[iknowjohnny]

Ok, i'm the dumb guy here, knowing very little about theory. But while i could be wrong it would seem to me that if the tube could swing to 200% when biased to 100% at idle, then even playing it "safe" biasing at say 80% would still allow it to swing to 180%, no? So that would suggest you's have to bias it near zero to keep it from ever exceeding 100%. So i think Alex must be right on this one, but go ahead and debate it some more because i learn a lot listening to you guys duke it out.

[jjman]

I'll play. If it's biased at "100%" of the gospel max wattage it is likely in Class A. Assume this is the case and it is 15watts. Let's say the idle current is 52ma. And the current swings from 1ma to 103ma during the signal.

This is an output of 15watts, no?

[Ronsonic]

This is why I really do not like the semi-technical discussions of class of operation. There is so much misinformation, some deliberate, and even more simple and honest misunderstanding of what is a very, very complex phenomenon that it becomes almost impossible.

Here's an example: "Think it through and you will see why... both tubes are at 100% at idle, when one is at 125% the other is at 75%, the total current draw remains constant, the bias voltage doesn't change"

If the amp is truly operating in Class A, then there IS no bias current change on either tube at any time*. That is the basic and fundamental definition of Class A. And here we get to why it is kinda silly to even argue about class of operation here, we are regularly pushing the amp into an overdriven condition. There are no formally defined classes of operation at that point, there is only what the amp is doing.

The biggest mistake people make is talking about "biasing an amp in class A" rather than operating the amp in class A. This isn't just about how much current you run through the quiescent tube. It's about transformer behavior and the reflected impedance of the speaker (which changes radically through its frequency response), the screen supply, the driver circuit and its output crossing the zero point. You can't "bias" an amp as Class A, you have to actually "operate" it in Class A.

Cathode bias, does not equal class A.

And no, hotter isn't always better.

</rant>

*Okay, actually at max power output more current is going to the speaker and DC current goes down very slightly. This is negligible and we are going to ignore it.

[Jana]

Here's an example: "Think it through and you will see why... both tubes are at 100% at idle, when one is at 125% the other is at 75%, the total current draw remains constant, the bias voltage doesn't change"

If the amp is truly operating in Class A, then there IS no bias current change on either tube at any time*.

Isn't that what I said?

There is so much misinformation, some deliberate, and even more simple and honest misunderstanding of what is a very, very complex phenomenon that it becomes almost impossible.

Yup, there is a lot of misinformation, and a lot of people take it as gospel.

a very, very complex phenomenon that it becomes almost impossible.

Ah, guy stuff, that explains it. Anybody know where I can find some pink tolex?

[Alexo]

Hey Jana, with all due respect, I think the point that's being made is that biasing to 100% dissipation does not neccessarily mean that you are "biasing to class A operation." The bias setting does come into play when you're designing a class A output stage, but that setting alone is not completely responsible for the class of operation your output stage uses. In many cases, you can bias up to 100% and still have a class AB amp, sometimes you can bias to 90% and have a class A amp (as in a single ended amp, for example.)

[iknowjohnny]

Ok, then excuse my ignorance, but i'm just trying to understand this...what IS it then that makes an amp class A? What thing or things have to be in place in order to have a class A PA?

[Alexo]

Check out the valvewizard.co.uk site and read up on drawing load lines. Then you'll see what you need to make an amp that never crosses into the class B portion of the load line, usually it requires a low plate voltage, a hot bias and setting the screen voltage and load just right. Some old datasheets also have circuit examples for class A amps.

[iknowjohnny]

Well, i tried to get load lines before and my brain just shuts down when i see things like that. But i understand what you're saying now. It's a matter of voltages and current and all these specs being in a certain balance. Thats more than i ever understood about it before.

[Alexo]

I know that feeling. It took me a long time to digest a lot of my electronics learning and my brain hurt on numerous occasions. But each time you come back to it, it makes a little more sense and pretty soon, things start falling into place.

[j-po]

It's pretty normal to exceed maximum plate dissipation over the course of a big signal, especially when it enters the class B portion. At least in guitar amps. As long as the average dissipation over a full signal swing is under 100%, it's usually fine.

Besides that, the bias setting doesn't affect what happens when you enter class B operation, only how soon you switch over to it.

How can avg dissipation stay within 100% in an AB amp when its biased at 100% idle? Only as long as both tubes conduct, i.e. up to 200% / 0%. When the other tube goes to say 250% (normal class AB operation) then drops down to clip at 0% you certainly have a average over 100%. During a long chord it's it's stuck at over 100% for quite some time and even after the chord it only gets to rest at 100%.

By the way, our amps never operate class B. Or at least sound horrible when they do. Class B means 0% idle current.

[j-po]

...In many cases, you can bias up to 100% and still have a class AB amp...

Correct. You could even bias at 110% idle and still be in class AB. Only now tube lifespan is even shorter.