# Efficiency of class A AB and B. What does it mean?

Discussion in 'Amp Tech Center' started by peteb, Jul 17, 2018.

1. ### petebFriend of Leo's

Posts:
4,208
Joined:
Apr 25, 2003
Location:
That’s what this thread is or was trying to establish.

Theoretical Maximum efficiency for

Class B = 78.5%
Class AB less than 78.5%
Class A = 1/2 of class B

2. ### petebFriend of Leo's

Posts:
4,208
Joined:
Apr 25, 2003
Location:

I love all the numbers at the front to set up the situation.

Like maximum plate current from signal and max plate voltage from signal.

And I see these are real conditions, I recognize the numbers.

These are real playing conditions, right?

And you said that bias isn’t about power level but isn’t that what is going on here?

Maxing out the bias to max out the power.

3. ### Old Tele manFriend of Leo's

Posts:
3,941
Joined:
May 10, 2017
Location:
Once banned always banned
Tweaking the bias higher (>70%) is about squeezing more from the EbIb plate curve across the plate load Rpp (or Zaa).

4. ### Old Tele manFriend of Leo's

Posts:
3,941
Joined:
May 10, 2017
Location:
Once banned always banned
Class A begins at 50% maximum theoretical, but falls to 25-30% due to tube characteristic use (triode < tetrode < pentode) of available plate voltage.

Class A PP is what you're typing but NOT indicating.

peteb likes this.
5. ### petebFriend of Leo's

Posts:
4,208
Joined:
Apr 25, 2003
Location:
They say at full power they are dissipating 69% and 87% at idle.

I thought the full power dissipation was more than idle dissipation in class AB?

6. ### Old Tele manFriend of Leo's

Posts:
3,941
Joined:
May 10, 2017
Location:
Once banned always banned
There are TWO different power sources/causes operating inside a vacuum tube:

Idle is DC, full power is AC! ! ! !

Do not intermix them.

7. ### petebFriend of Leo's

Posts:
4,208
Joined:
Apr 25, 2003
Location:
Ok

From your numbers full signal AC plate dissipation is less than the DC plate dissipation at idle.

That means the limiting factor is how much DC power the plate can take and not how much AC power the plate can take. And the DC power at idle is less than the DC power dissipation at full power, so a more useful number would be the DC power dissipation at full power instead of the AC.

EDIT: it’s how much power of both AC and DC combined that matters and is the limiting factor and for what the tube is rated for.

Last edited: Jul 19, 2018
8. ### petebFriend of Leo's

Posts:
4,208
Joined:
Apr 25, 2003
Location:
If the maximum efficiency of a class B or class AB output section is 78.5% and this amp puts out 55 Watts then the power input to the power section would have to be 70 Watts.

70 Watts input * 78.5% = 55 Watts

35 Watts per tube

They can’t all be true

Either the amp puts out less Watts, or the max efficiency is higher or the max plate dissipation is at least 35 Watts per tube.

Last edited: Jul 19, 2018
9. ### petebFriend of Leo's

Posts:
4,208
Joined:
Apr 25, 2003
Location:

Don’t not intermix them!

Your answer really opened my eyes on the hows and why’s of setting bias in AB amps.

And it brings up questions for biasing class A amps.

Both the AC and the DC current will add to the plate dissipation.

You set the DC idle bias so that that in playing conditions, the increased DC current and added AC current, both adding heat to the plate, stay below the maximum plate dissipation.

The max plate dissipation is a good standard number to use for signal energy input into the tube. The idle dc current doesn’t start as signal current but it becomes signal current once the signal is placed on it.

Last edited: Jul 20, 2018
10. ### petebFriend of Leo's

Posts:
4,208
Joined:
Apr 25, 2003
Location:
Nice paper all the way thru.

Is the 65% plate efficiency calculated off of conduction angle?

In this case, divideng the power out 55W by the max plate dissipation of the tube type 60 W is 91.7% efficiency. High.

85 W input X. 65%. =. 55 W

Using a plate efficiency of 65% also means that if the output is 55 W then the input is from 85 Watts of plate dissipation or 43 Watts per tube.

My take on efficiency

78.5% is a good maximum for efficiency of class B and AB classes.

In order to have all known output powers of well known amps to be true, then tubes are going to need to dissipate a little more energy, or the amps will put out a little less energy.

Actually I think the reported power outputs are fine, the 78.5% max efficiency is fine, even the max dissipation figures.

The only thing needing adjusting is realizing that the max plate dissipation ratings are ratings, they are valid ratings but they get exceeded.

The questions is when they get get exceeded, does the extra power just get dissipated and wasted as heat and not contribute to more power, or does it contribute power?

I think it contributes power but just a minor adjustment, nothing significant.

Last edited: Jul 20, 2018
11. ### petebFriend of Leo's

Posts:
4,208
Joined:
Apr 25, 2003
Location:
Now let’s consider the amp with two power tubes, 6L6 or EL34, that puts out 100 Watts of power because it is running in class B.

The 6L6 tube dissipates 30 W max

The EL34 25 W max

Last edited: Jul 20, 2018
12. ### petebFriend of Leo's

Posts:
4,208
Joined:
Apr 25, 2003
Location:
Aiken Amps can answer a few questions

Here’s the first paragraph on biasing:

Why do you need to bias a tube?

Tubes have to be properly biased in order to function as amplification stages. A tube is biased by setting the amount of DC current that flows in the tube when there is no signal present at the tube's grid with respect to it's cathode. This DC bias current can be set in a number of ways. The bias point determines several things about a tube amplification stage. It determines the power output, amount of distortion, headroom (the size of input signal that can be applied before the output signal clips), efficiency of the stage (the amount of output signal power vs. DC input power), gain of the stage (the magnitude of the output signal for a given input signal), noise of the stage, and class of operation (class A, AB, etc.). The proper bias point is a tradeoff between all of these factors, and selecting the optimum bias point can sometimes be difficult, and it will vary depending on the amplification stage requirements.

13. ### petebFriend of Leo's

Posts:
4,208
Joined:
Apr 25, 2003
Location:
Does Bias affect power?

The first thing Aiken says that bias does is:

“It determines the power output”

How is efficiency calculated?

Aiken says this:

“efficiency of the stage (the amount of output signal power vs. DC input power)”

Is plate dissipation DC, AC, or both?

Aiken says this about class A:

“while true class A amplifiers generally run right at the maximum plate dissipation (the dissipation at full power is lower than the dissipation at idle in a true class A amplifier)”

Here, the topic is dissipation and redplating, not efficiciency but the focus is the DC, or the only dissipatuon is DC, why? Why isn’t there AC plate dissipation in class A that increases dissipation during full power?

IDK

On the 70% rule Aiken says this:

“A general rule of thumb is that class AB amplifiers are usually operated at no more than 70% of the maximum plate dissipation of the tube (to account for the higher dissipation that occurs under signal conditions), “

Last edited: Jul 20, 2018

Posts:
4,208
Joined:
Apr 25, 2003
Location:

15. ### petebFriend of Leo's

Posts:
4,208
Joined:
Apr 25, 2003
Location:

I think we are in agreement here.

I think we agree that power supply and dissipation ratings are measuring the same thing, and that the power supply often over runs the dissipation ratings.

16. ### petebFriend of Leo's

Posts:
4,208
Joined:
Apr 25, 2003
Location:

That says quite a lot about class B operation.

700 volts on the plates, a cold bias and low current.

What’s the current?

17. ### Bill MooreTele-AfflictedSilver Supporter

Posts:
1,259
Joined:
Feb 2, 2012
Location:
Silver City, New Mexico
I didn't bother to check tube current, the bias procedure has one set the voltage across the emitter resistors at 25mv.
(There are 2 transistors in the bias section that control the output tubes bias.)
Actually there was more than 700 volts, I just don't remember the number, MM says it could be 770 volts!

18. ### petebFriend of Leo's

Posts:
4,208
Joined:
Apr 25, 2003
Location:
Thanks Bill

I think any amp that puts out more power than it appears capable of is putting more power in that the tube appears capable of. These two tube 100 W amps must really be testing the power tubes.

The idle bias current should be zero.

Two tubes putting out 100 W must each be putting out 50 W and must each be taking in 50 W.

700 V X 70 ma = 50 W

But the tube is only working half of the time so each tube when it is on must have a DC current over 100 mA.

Is that true?

For two tubes to put out 100 W combined, each tube must dissipate in excess of 50 Watts?

19. ### petebFriend of Leo's

Posts:
4,208
Joined:
Apr 25, 2003
Location:
These are my conclusions

Bias adjustment is a power adjustment but the adjustment is very minor. You can’t turn one amp into another with a bias adjustment.

The maximum plate dissipation ratings for tubes are good and meaningful numbers. The ratings can be exceeded and they regularly are. The plate dissipation is the energy put into the tube, therefore the max plate dissipation serves as a pseudo max power out rating as well.

The efficiency rules for tube amps are real and cannot be violated. They are based on real world concepts such as you can’t get more energy out than you put in, and there will always be losses and inefficiencies. Maximum theoretical power out can only be increased by an increase in input power.

20. ### viccortes285Tele-Afflicted

Age:
71
Posts:
1,024
Joined:
Jan 14, 2017
Location:
Phoenix
On my DR I set it as follows
For 400 vdc plate voltage you can go to 24.5 ma if amp sound better to your ears.

You'll see plate voltage will go down at 24.5 and you may have to do a new bias adjustma.
Each time you adust bias you must re-check plate voltage and do a new ajustment for ma if needed.

You can keep bias colder if you ears did not see a better tone. This make tubes last longer

peteb likes this.
IMPORTANT: Treat everyone here with respect, no matter how difficult!
No sex, drug, political, religion or hate discussion permitted here.