# excessive plate dissipation causes redplating - does this power dissipation from AC or DC?

Discussion in 'Amp Tech Center' started by peteb, Jul 20, 2018.

1. ### petebFriend of Leo's

Posts:
4,208
Joined:
Apr 25, 2003
Location:
dc current and ac signal current flow across the plate but in the case of a single ended class A amp, the plate dissipation is constant at all times, resulting from the constant DC current flow, and is supposedly unaffected by AC signal flow on the plate.

Why doesn’t the signal current contribute to the plate dissipation?

On class AB, signal on the plate will increase the DC current flow and cause AC current to flow. But the dissipation that will cause redplating, is it again, only the DC that matters?

2. ### elpicoTele-Afflicted

Posts:
1,154
Joined:
Sep 14, 2011
Location:
Vancouver BC
The plate dissipation in a class A amp is not constant at all times. It’s highest at idle then falls when the tube starts passing signal. I already wrote this in one of these threads.

We have to design the amp so the disipation remains within the tube’s limits under any normal use. Well that’s easy with class A, the tube will never be hotter than when it’s idling and doing nothing, so in this particular case we are interested in the idle dc power. That’s not true in class AB or B however, it’s irrelevant in those amps.

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

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

Can you provide evidence? I’ve measured the DC voltage on the plate and on the cathode throughout extreme playing conditions, maybe you can get those voltages to move, I can’t, they hold rock steady. In all practical measurements the dissipation in class A is the same at idle and full output power.

Aiken amps:

“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).”

I have confirmed this. Class A amp run all out all of the time.

Further evidence is I have played different amps thru a power monitor in all playing conditions measured to th hundredths of amps.

Class AB amps will draw more power from the wall under heavy load but a champ won’t. Once a champ gets turned on and had a half a minute to warm up the tube, the current draw then is the current draw you are going to get for as long as the champ is plugged in.

People who say plate dissipation drops in class A in signal conditions need to accept that the amount is very small or try to quantify it. How much does the dissipation drop? I would like to know.

4. ### petebFriend of Leo's

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

We agree on the most important part about class A

We know AC generates heat, just feel the heat from any AC powered light bulb.

The current flow on the plate is DC with an AC component. The DC current is given credit for causing the plate to dissipate energy, why isn’t the AC current component also going to add heat?

I used to think I had this figured out. The champ signal is the whole cycle, positive and negative. Because it has equal amounts of positive and negative, the overall DC current is unchanged. In class AB each tube always sees either a positive signal or it is in cut off (0), so the signal increase the overall flow of DC.

This explains why single ended has constant DC current flow and AC has increasing DC current due to signal.

But the question is, why doesn’t the AC current component add to the plate dissipation?

I guess the answer is that it just doesn’t.

5. ### elpicoTele-Afflicted

Posts:
1,154
Joined:
Sep 14, 2011
Location:
Vancouver BC
I can’t make a chart to show how it works right now, need my computer for that and I’m away from home until after the weekend. The quote you provided from Aiken also agrees it drops though. You might want to read that more closely:

“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).”

This has nothing to do with the current draw from the wall or the dc supply voltage, those aren’t expected to change. As you turn up the signal in a champ some of the disipation that was occurring in the tube gets shifted to the speaker. The total amount of power being dissipated SOMEWHERE is unchanged, but now some of it is happening in the tube and some of it is happening in the speaker. If you start with 15w at idle, then at full output you divert 6w into the speaker, that means there’s only 9w left in the tube side.

6. ### petebFriend of Leo's

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

We agree on the important concept, apply signal to a class AB amp and the plate dissipation rises.

Apply signal to a champ and the plate dissipation does not rise.

Thanks for putting numbers to it.

The plate dissipation is the DC plate current times the DC voltage difference between plate and cathode.

Let’s say Vp=400, Vc=25, and Ip=40 mA at idle

15 W = 375 V * 0.04 A

What I am saying, from personal experience, if the plate and cathode volatages on this class A amp were monitored thru all playing conditions, the plate will stay at 400 VDC and the cathode will stay at 25 VDC. I just had to think this thru, do I know the plate current stays the same? Yes. If the cathode DCV holds steady, so does the cathode DC current.

This is enough evidence. The plate dissipation, at the plate, will hold at 15 W.

This makes me want to remeasure at the OT, but there is no need. One end of the OT is the plate, and it doesn’t change. The other end is the B+ and it doesn’t change, there is no sag in the power supply of a champ.

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

Posts:
3,941
Joined:
May 10, 2017
Location:
Once banned always banned
The Class-A reduction in plate dissipation is caused by the linear plate load line becoming further away from the upward curving constant plate power-dissipation curve on the tubes Eb-Ib curve...look for yourself.

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

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

I have already measured it. If it exists it is so small it can’t be detected.

Do you have a number?

How much does the dissipation reduce?

9. ### petebFriend of Leo's

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

Thanks Elpico for putting it that way, I see where you are coming from.

I would change that and make it into this:

15 W of DC power applied to the power tube, this stays constant thru idle all the way to max output. The DC power never makes it thru the OT to the speaker.

Under full power, 6 Watts of AC signal power develops on the plate, this goes thru the OT to the speaker.

Idle plate dissipation: 15 W of DC power

Full output plate dissipation: 15 W of DC power
And 6 Watts of AC power.

10. ### petebFriend of Leo's

Posts:
4,208
Joined:
Apr 25, 2003
Location:
I think it is a significant difference in amp performance when one amp draws more current from the wall under load and one amp doesn’t.

This highlights one of the main differences between class A and class AB.

If class A does not draw more wall current under load
And class AB does

What do you think of the class AB amp that does not draw more wall current with signal at low volume but does draw more wall current with signal at higher volumes?

I thougt of a whole new purpose of a power monitor.

You can monitor when a class AB amp is exhibiting class AB behavior and when it is exhibiting class A behavior.

That’s cool and amazing

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

Posts:
3,941
Joined:
May 10, 2017
Location:
Once banned always banned
1) Do you LOOK at tube data sheets?

2) Do you UNDERSTAND the data they show?

3) Do you KNOW 'how' Class-A load line "lays" upon the Eb-Ib curves...and touches the rated plate dissipation line at idle?

4) Do you KNOW "how" the AC-plate current moves along the load line at full conduction?

5) Can you SEE that the linear load-line path becomes further away from the central 100% idle point at full power?

6) If the curved "maximum plate dissipation" curve represents 14W (6V6GT assumed) and the two load-line ends are BELOW that 14W curve, then points on the load-line are creating LESS than 14W dissipation!

Study the datasheets and UNDERSTAND what they're showing you.

Last edited: Jul 20, 2018
Bendyha and Modman68 like this.
12. ### BendyhaFriend of Leo'sSilver Supporter

Posts:
2,528
Joined:
Mar 26, 2014
Location:
Northern Germany
^^DATA SHEETS^^
They are the best point of reference, and the easiest way to quickly spot the desired working conditions.
If you look at them, you will see they are made up of curves, therfore nonlinearity is the norm under dynamic conditions, that is why comparing a limited set of measurment numbers from a static condition will give you no reliably valuble information for prediciting another condition in as it would be in a dynamic state...or at least nothing nearly as quick and simple and accurate as could be read off the data sheet.

Last edited: Jul 20, 2018
LudwigvonBirk likes this.
13. ### Old Tele manFriend of Leo's

Posts:
3,941
Joined:
May 10, 2017
Location:
Once banned always banned
"Pictures are worth 1,000 words"

"1,000 words are just words without the pictures"

14. ### petebFriend of Leo's

Posts:
4,208
Joined:
Apr 25, 2003
Location:
This is just basic stuff, everyone knows single ended amps run full out all the time.

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

Posts:
3,941
Joined:
May 10, 2017
Location:
Once banned always banned
That statement illustrates that you do not understand what you post nor what others have been trying to explain to you.

SE Class-A tubes conduct all the time, but their dissipation DOES change/decrease between idle and full output. There is a difference which you seemingly fail to grasp.

JD0x0 and Bendyha like this.
16. ### BendyhaFriend of Leo'sSilver Supporter

Posts:
2,528
Joined:
Mar 26, 2014
Location:
Northern Germany
Pete, get dynamic...read, or at least look at these pdf's

File size:
98.2 KB
Views:
184
File size:
109.6 KB
Views:
34
File size:
113.8 KB
Views:
36
17. ### Old Tele manFriend of Leo's

Posts:
3,941
Joined:
May 10, 2017
Location:
Once banned always banned
If he's having trouble understanding assumed-linear (basic) resistive load lines; actual elliptical (real world) inductive load lines will really trouble him.

Last edited: Jul 20, 2018
18. ### LudwigvonBirkTele-Holic

Age:
118
Posts:
933
Joined:
Aug 26, 2017
Location:
I'm afraid to click on those.

robrob likes this.
19. ### Old Tele manFriend of Leo's

Posts:
3,941
Joined:
May 10, 2017
Location:
Once banned always banned
They won't bite...but they might nibble at your mind (wink,wink).

Age:
118
Posts:
933
Joined:
Aug 26, 2017
Location: