# Trying to make a more accurate bias chart

Discussion in 'Amp Tech Center' started by peteb, Jul 17, 2017.

1. ### petebFriend of Leo's

Apr 25, 2003
The main question I want to ask, does maximum plate power dissipation stay constant over all voltage ranges or does it increase as plate voltage increases?

Existing bias charts ( weber vst, Jim jones), and bias calculators use one value for max plate dissipation for all plate voltages. The tube spec sheets show the max plate dissipation rising with the plate voltage.

I did bias charts both ways, here is the sliding scale I came up with:

The 6v6 is a seven watt tube at 300 volts (matches the spec sheet) rising to a 14 watt tube at 450 volts (matching reality)

The 6L6 is a 22.5 w tube at 400 volts rising to a 32.5 watt at 500 volts.

Here is my argument for a sliding scale. The tube can handle more power at higher voltages. At lower voltages, the current would be unpracticable to reach the full power like a 6v6 putting out 14 watts at 300v.

Here's where my argument breaks down. For a 6v6 to be a 14 watt tube at a low plate voltage like 300 volts, it needs 46.7 mA of current, and 28 mA at 60 % idle. It doesn't sound unlikely, it doesn't sound like too much current to work.

Here's my samples for the 6L6, unfortunately I don't know how to show data on this forum, so it's just photos. The first is the 6l6 with constant power out no matter the voltage, the second is the sliding scale.

Mar 26, 2014
Northern Germany
How much heat a tube tolarate, and how much it can dissipate remain constant (unless unaided through forced cooling)
How much heat is generated by a voltage is a product of the current. Mr. Ohms sort of made a law about it, which we don't want to go breaking, or the four laws of thermodynamics.
So maximum plate power dissipation does stay constant over all variations of voltage, and over all the amounts of current, as the two interplay with eachother.
But as to which dissipation rating a tube would sound best at, might well vary with the voltage, depending on other factors as well maybe.

Last edited: Jul 17, 2017

3. ### RLee77Friend of Leo's

May 15, 2016
Silicon Valley
I think perhaps you are confusing max plate dissipation with typical output power. The former is an upper limit of what the tube can handle; the latter is how much power to expect from a typical amp with the specified voltage/current and typical biasing.

4. ### jazzguitarTele-Afflicted

Mar 17, 2003
Germany
As they say, max power dissipation is a characteristic of a particular tube regardless whether the voltage is high or low. Like a speed limit, regardless whether your car has small or big tires.

To be more precise: the maximum dissipation that occurs in a particular amp indeed depends on the voltage (and signal size i.e. volume) but the allowable maximum for a particular tube type remains the same.

That's basically correct, but still most of that is bunk. There is not any one correct setting valid for all amps with a particular plate voltage and tube type.

I have written a tech paper on that

http://www.blueglow.de/TechPaper/Debunk.html#bc

I hope it is not too technical?
There is good info at the Aiken amps web site, but that's even more technical.

Nope. The 6V6 is a 12W or 14W tube, depending on make and particular batch, the 6V6GTA are said to be 14W, but that has to be taken with a grain of salt.
The 6L6GC is a 30W tube (earlier variations such as the 6L6, 6L6GB are different).

These tubes are designed to handle up to as much power as you want provided their plate dissipation - and other maximum values given by the manufacturers - are not exceeded. There is absolutely no need to keep any of these power figures in mind when biasing other than you need to make sure they are not exceeded.

All of these are dissipation (or maximum current or voltage) figures, and not directly connected to how much power output your circuit delivers. With conventional circuit design however, the output power figures will be in the same range as the maximum dissipation figures of the power tubes.

Nope. There are also current limits so there will not be as much power output with lower voltages, but the maximum the tube could handle remains the same.

The idea that you should aim for 70% (or 60% or whatever) to bias a power amp correctly and that is true for all amps is just nonsense that came up on the internet, a classic urban myth.

It is true that these figures are more or less a safe limit for many amps, but they're not target values. I would set the bias on an individual amp depending on what the designer aimed at as far as I can tell, and from experience.

Oh, and how much power a particular amp can put out mostly depends on its power supply, whether this can as the name says supply that power. Now you choose a power tube type that will work well in that amp, eg a 6L6 whether you have a 30W or a 75W amp.

I hope this helped ....

Last edited: Jul 17, 2017

5. ### robrobPoster ExtraordinaireAd Free Member

Dec 29, 2012
United States
The maximum plate dissipation for a 6L6GC is 30 watts.

Watts = voltage x current so the higher the plate voltage the lower the plate current must be to stay under 30 watts.

For a Class A amp with one 6L6GC max idle plate current would be 100% of 30 watts.

400v x 75 milliamps = 30 watts and 100% max plate dissipation.

300v x 100ma = 30 watts and 100%.

500v x 60ma = 30 watts and 100%.

24 track likes this.

6. ### alnicopuFriend of Leo's

Oct 3, 2009
georgia
This^^^. A watt is a watt. So, in theory, it doesn't matter how you get to it, low voltage/high current or high voltage/low current. As long as the math adds up to the same result. I say theory, because everything has its reasonable limits. You probably wouldn't want to use 300 amps x .1 volts to get 30 watts. Basically a dead, short circuit. Materials have physical their limits. I would be running the secondaries of the transformer that are feeding my entire house to power a tube. But we're not going that far down the road here.

7. ### petebFriend of Leo's

Apr 25, 2003
Thanks for the replies.

I agree with everything written above.

Here's the problem, tubes are being considered cold biased when they may not be.

The standard bias calculation will tell you how close you are supposed to be to red plating a tube because it is exceeding maximum plate power dissipation, but it tells very little about the operating point of the tube.

I Agree it's all about setting an upper limit.

Bias is supposed to be about tone, and the operating point of the tube, where the idle current is in relationship to saturation and cut off. When People throw out comments like they like 60 % bias, when 60 % in one situation is a far different operating point than 60 % in another.

Here's my case, made up, it's not even as extreme as my real case:

Suppose a 2x6v6 amp idles at 14 watts. The standard Formula says those two 6v6 tubes are idling at a cold 50 %, 14w/28 w. It's a cold bias because it's 50 %.

It turns out the amp is not biased cold at all. The amp has 12 watts maximum output, and it's biased to idle at 14 watts.

What I'm after is a bias method that says more about the operating point of the tube in the amp.

Is the answer to use the maximum power of the amp and not maximum power of the tube if one wants to get an idea of the operating point of the tube in the amp?

Nobody uses the amp output power when they calculate bias do they?

What's the relationship between redplating and tube saturation? Not too much but they are related. You set the bias to be below redplating, and to get enough tube saturation.

I learned something about amp design thinking about this question. When comparing 12 w and 22 w 2x6v6 amps, I was thinking the 22 watter will saturate but how will the 12 watter ever get there? Then I realized, a 22 watt 2x6v6 amp could be a total clean machine and never reach tube saturation, while a 12 watt amp could reach tube saturation really easily, it depends on the rest of the amp design.

My conclusion. To get a better idea about the operating point of tube in an amp,dividing the idle power by the max amp power is more informative than dividing the idle power by the max tube power.

8. ### RLee77Friend of Leo's

May 15, 2016
Silicon Valley
There are some good articles online that detail the parameters and design considerations of a power stage, in more accurate detail than I could explain here.
Remember that tubes have internal resistance, and a lower supply voltage will yield a lower max attainable output power, but the max allowable power dissipation remains the same. Also, it's average power dissipation that you want to keep under the limit; it's acceptable to exceed the power rating at various intervals of the signal, as long as the average is at or below it.

Also, saturation has little to do with max output power. A saturated tube is fully 'on' and not dissipating much (low plate-to-cathode V), just as a tube near cutoff (max plate-to-cathode V) is not dissipating much. It's the linear portion in between where max power is seen.

You're right about the amp design determining the ideal idle bias points; hotter does not equal better in all cases. Too hot or too cold can limit the clean headroom.

Last edited: Jul 18, 2017
peteb likes this.

Mar 26, 2014
Northern Germany
But it does tell you all you need to know when biasing an amp that has been designed to use those tubes in the circuit that has been designed around them.

Then you must know and understand the concept of drawing up load-lines on the characteristics charts of the tube in question set to the screen grid voltages designed into the amp, with the use of the imparted load from the output transformer and speakers, with the supply voltages to the plates...these define the operating point of the tube.... then you can choose and vary the bias setting of the control grid at idle setting... defining the dissipation, and the A Class biasing as it drops into the cut-off point at cross over, and the paralell effected cut-off and cross-over to the B Class part of the cycle.

Red plating is red plating, the tube can only take so much. Saturation is a far more complicated mater...for a start saturated when...saturated where..saturated with or by what....then one can answer maybe saturated why. Don't c make it confusing by adding transformer saturation into the equasion just yet...that is a separate matter. The bias point in my mind has little to do with saturation...that is happening down the other end of the load-line, and being determined by other factors far more than the bias point. Abrupt or gradual saturation? Limited?..if so how?

Okay...now you're mixing up so many factors that should be dealt with seperatly....because they have nothing to do with each other in the context of your supposition.
Trying to draw an analogy between rated amp output and plate dissipation is like determening the weight of CD's by the type of music recorded on them, and concluding that Heavy Metal music weighs more than Light Opera.

For a start, the maximum potential wattage output of an amplifier...this is in itself a complex concept. (What part of the amplifier?) For starters, it has nothing to do with accoustic sound presures, potential volume, efficiency of any components.

The heat given off by the tubes plate is wattage.
The heat given off by the tubes screen-grid is wattage.
The heat given off by the tubes cathode is wattage.
The heat given off by the cables leading to the output transformer is wattage.
The heat given off by the output transformer is wattage.
The heat given off by the speaker wire is wattage.
The heat given off by the speaker is wattage.
The heat given off by anything the is wattage.
Each must be handled seperatly. Some will be effected by the same thing, others not. Some will go up and down together, some will go down when the other goes up.

The bigger picture is perhaps the reason that charts like you want to have do not exist, because they cannot, and they serve no purpose.
Every amp must be taken for what it is. You could analyse the gain stuctures of each amp, corrolate the voltages and currents....but in the end you are probably best to just bias it to 70% to start with, and then adjust up or down to suit your ear and playing style....which might well be different than that which other people might choose.

Last edited: Jul 18, 2017
peteb likes this.

10. ### robrobPoster ExtraordinaireAd Free Member

Dec 29, 2012
United States
This is what's confusing us. You're talking amplifier output power and the no signal idle power (plate heat).

A 12 watt 2x6V6 is more than likely a cathode biased amp. A 20 watt 2x6V6 is a fixed bias amp.

Maximum plate dissipation rating is a heat limit. Flow too much current and the plate gets too hot and things will melt.

If you set the idle no signal current flow through the tube (the bias) too high when an AC guitar signal is applied to the tube grid during the positive voltage half of the AC wave more current will flow and you're more likely going to bump into saturation (max current flow) and clip the signal.

For a Class A amp you can set 100% dissipation as the bias current. This doesn't mean you're right up against tube saturation. The positive half of the guitar AC audio signal can flow more current but because the negative half will reduce current flow it averages out around 100% and the tube is safe.

24 track and peteb like this.

11. ### petebFriend of Leo's

Apr 25, 2003
Good info all, thanks.

The brown black and silver princetons are all 12 watt, fixed bias.

I'm looking first at class AB fixed bias, also class A, but Not cathode biased class ab yet.

Ok, so the question comes down to this:

How can a Princeton reach power tube saturation if it's full output power is only 12 watts? which is 43% of full output for the tubes.

jimytheassassin likes this.

12. ### robrobPoster ExtraordinaireAd Free Member

Dec 29, 2012
United States
The Princetons put out lower wattage due to their smaller output transformer and cathodyne phase inverter (vs LTP in other blackface amps). The blackface PR layout shows 410v on the 6V6 plates so they're running near full power--the tubes aren't being run at 12 out of 20 watts. The PR is biased about the same as all the other blackface amps.

Last edited: Jul 18, 2017
peteb likes this.

Mar 26, 2014
Northern Germany
Ahh....just trying to keep up the suspence!

I'm still writing at it, but haven't been at it for the last week or so.....all will be revealed in the near future....but not in the next installment...but the one following that.

24 track and peteb like this.

14. ### petebFriend of Leo's

Apr 25, 2003
PR=17W, P nonverb=12W

heres a bit of my latest bias chart. I think it shows the operating point of the tube best out of all of the charts.

I have four charts for class ab 6v6 fixed bias alone.

Original - the standard calculation
Version 2 - the output power of the amp steps down at each voltage level
Version 3 - the same currents, 50,60,70,100% are applied at each voltage
Version 4 - 100% power is now 12 w for all voltages, this for a 12 w amp.

This last chart, version 4, I think shows the operating point of the tube the best, and the 400 volt section of the table is below:

%. Tube. Volt. MA. Watts
50%. 6v6 400. 7.5. 3
60%. 6v6. 400. 9. 3.6
70%. 6v6. 400. 10.5. 4.2
100%. 6v6. 400. 15. 6
100% 2X6v6. 400. 30. 12

To interpret this table, an amp running a pair of 6v6 tubes in fixed bias class AB, at 400 volts, needs a minimum 7.5 milliamps of current to be biased with idle putting out 50 % of full amp power.

Most people don't RECOMEND current lower than 10 milliamps. This table shows to me that 7.5 mA is still in the middle ground, and should not sound cold. This mirrors my cold biased 6g2, which does not sound cold at all.

Thank you for reading my little investigation, and for the replies.

15. ### robrobPoster ExtraordinaireAd Free Member

Dec 29, 2012
United States
peteb, are you saying a 12W Princeton (2x6V6 fixed bias) is biased very cold to put out only 12 watts?

The non-reverb Princton has 420v on its 6V6 plates. It would have to be biased really, really, really freakin' cold to drop from 20 watts down to 12 watts output.

I still don't understand what you're trying to do in this thread.

peteb likes this.

16. ### Old Tele manTele-Afflicted

May 10, 2017
Tucson, AZ
To OP: quiescent (idle) watts and signal (output) watts are TWO different things...but both are watts.

Idle assumes DC-bias only...output assumes drive AC-signal "riding on/about" that DC-bias.

peteb likes this.

17. ### clintjFriend of Leo's

Apr 4, 2015
Idaho
http://www.valvewizard.co.uk/se.html

http://www.valvewizard.co.uk/pp.html

I'm not sure how to even start jumping into this, but give these a read and see if it clarifies anything about output total power vs each tube's operating point. It might also be illuminating to consider a 4 output tube amp like the Twin Reverb for a few minutes.

I'm not going to sugar coat it. Output stage design and pentode/beam tetrode characteristics are a deep subject. Richard Kuehnel and others have written whole books on it, and they're not small books.

peteb and RLee77 like this.

18. ### petebFriend of Leo's

Apr 25, 2003
Rob,

I'm glad to see you brought up the AA964 amp for discussion. It will be interesting.

Let's apply the standard formula to the AA964 and hopefully you will see what I am saying. I happen to be a 964 myself.

AA964: 12 watts, class AB, fixed bias, 2 x. 6v6 @420 V. Max power of the tubes is 28 W

One tube:

100%: 14W / 420 V = 33.3 mA
70%: 9.8 W / 420 V = 23.3 mA
60 %: 8.4 W / 420 V = 20 mA
50 %: 7 W / 420 V = 16.7 mA

So you set your bias at a cool 50 % bias with 16.7 mA of idle current going thru each tube. Sounds fine doesn't it?

Your at 14W out of 28 watts for 50 % of full power dissipation. Check.

Here is the problem, the maximum power out of the amp is 12 watts, the 50% idle is putting out 14 watts. The idle shouldn't be higher than the max output, and choosing 50 for a cool idle didn't turn out to be cool at all.

My point is that the standard bias calculation does not always help one understand the operating point of the tube.
Especially when the output of the amp is much less than full output of its tubes.

What I did in this thread is I rebiased my amp from a very cold 20 % to a moderate 50 % by changing the way I calculated the bias, and best of all to me is I was able to raise the bias without actually raising the current, which makes an amp heat up.

19. ### Old Tele manTele-Afflicted

May 10, 2017
Tucson, AZ
Sounds like you're confusing IDLE (DC) watts and the combined OUTPUT (DC & AC) watts.

Idle watts are the result of constant (thus, quiescent or "quiet") current and volts, ie: Ipq x Vpq = Wp.q

Output watts occurs when the output tubes are provided an AC-drive signal (from phase splitter) which causes the output tubes to draw additional current (beyond the idle current), and it's ONLY the AC-plate current the creates the useful plate current/voltage 'swing' that couples across/thru the output transformer and drives the speakers...ie: the output signal wattage.

Something else to ponder: at full output power each tube is ON and conducting slightly more then 50% of the time; but during the other 50%, it's deep in cutoff and drawing NO current, not even idle current. whatsoever.

Last edited: Jul 18, 2017
peteb likes this.

Apr 25, 2003