What is Bias Excursion and How Bad is it?

[SilverFox]

Is bias excursion from say, -29 to -48 and even as high as -90 while playing low frequency, (Emaj and F#) power chords, is this potentially damaging? Does it occasionally cause output section failures during live performance?

The excursion only lasts for a brief period or for as long as the power chord is played then the bias returns to the original setting. But what is overheating during the excursion? The desired, base setting is -29 on 6L6 output tubes.

silverfox.

[JazzGuitarGimp]

This is normal. The signal is symetrically superimposed about the bias voltage.

[SilverFox]

So that should be a sort of rough indication what the grid level voltages are on the power tubes?

silverfox.

[Stevem]

No harm, but it's part of nasty sounding crossover distortion!

[pdf64]

Is bias excursion from say, -29 to -48 and even as high as -90 while playing low frequency, (Emaj and F#) power chords, is this potentially damaging? Does it occasionally cause output section failures during live performance?

The excursion only lasts for a brief period or for as long as the power chord is played then the bias returns to the original setting. But what is overheating during the excursion? The desired, base setting is -29 on 6L6 output tubes.

silverfox.

Where and how are you measuring this?
I fear you may be using a low spec meter at the power tube grids?

As long as components are rated for the highest voltage / power they are subject to, eg not 90V across a 50V bias supply cap, then there should be no damage.
Bias excursion acts to run the power tubes cooler; crossover distortion and in extreme, blocking distortion can result; the latter especially if there's a prolonged recovery time.

[Leo_Gnardo]

In all my time wrenching amps & reading, never ran across the term "bias excursion." Where did this come from? What do they mean by it?

In a self biased amp, you'd expect voltage on the outputs cathode R to rise as they draw more current in high signal conditions.

In a fixed bias amp, how the #### does a bias V at plates swing up to -90V from a nominal say -45V. Wait, there goes a flying pig!

Or does the bias get on board a cruise ship for a long weekend in Puerto Vallarta?

hmmmm......

[martin manning]

Here you go: https://www.ampbooks.com/mobile/amplifi ... excursion/

[Leo_Gnardo]

Here you go: https://www.ampbooks.com/mobile/amplifi ... excursion/

Thank you Martin! The pig has flown out of sight... this article is very enlightening! I'm verklempt - clamped. Gonna have to load up the coffee pot & re read this a few times. I knew there's without a doubt some movement of bias control voltage at clipping levels. And keeping bias "serve" Rg's & coupling Cg's minimum to keep time constants short to minimize blocking. Still have doubts that it could zoom all the way to -90V. In time, I'll "do the math" & see what about that.

Ya learn something new every day!

[David Root]

I recommend the book, it's quite thorough and AFAIK the only serious discussion of the dynamics of power amp distortion around. Not for newbies though, unless you're an EE.

[JazzGuitarGimp]

Oops, it looks like I misunderstood the question. Please disregard my previous comment.

[SilverFox]

From the referred page at ampbooks-

"When the input signal amplitude is finally reduced, the excess voltage across the capacitor bleeds off slowly through the grid resistor RG in a process known as bias recovery."

The most commonly observed reading on the yard sale special BK precision meter, (cost $1.00 so yeah cheap), while connected to #8 was around -42. This value is close to the calculated "excursion" value of .61 for the parameters entered in the calculator on the ampbooks page.

Leo_Gnardo- As they say in the Shaolin Temple: Kow chi ha no liuea wo ni tee, "A good student is like a partially filled cup where there is always room for new water"

silverfox.

[Leo_Gnardo]

Leo_Gnardo- As they say in the Shaolin Temple: Kow chi ha no liuea wo ni tee, "A good student is like a partially filled cup where there is always room for new water"

silverfox.

Yes, like that. I t'ink of the brain as a sponge that can soak up more knowledge. Gets less so with age for many people, and then there are some who are sooo smart, they know it all and for them my comment is "Junior's head's as hard as rock."

I knew there was some variation of bias with signal, no problem there. And keeping time constants short as practical within the output circuit was an advantage vs blocking. With Martin's guidance we have a formula for it, how cool is that!

Thanks Martin & silverfox for bringing it up & have a great Thanksgiving holiday one and all. If you're in Canada, do it again. Can't hurt.

[pdf64]

The most commonly observed reading on the yard sale special BK precision meter, (cost $1.00 so yeah cheap), while connected to #8 was around -42

What is '#8'?
Some meters don't deal happily with Vac superimposed on Vdc, so it may be that the meter is getting confused. Can you scope it?
What happens to the bias supply output voltage, at the junction of the grid leak resistors, eg 220k, when the amp is overdriven?

[martin manning]

Note Kuehnel defines bias excursion ratio as bias excursion divided by the PI output signal amplitude; it's not a fraction of the bias voltage. This is a common problem with on-line calculators- you often don't know what assumptions are made, and they are easily (and often) mis-applied. I think the best way to use this calculator is to make some numbers for representative designs to compare against.

The -3db point it displays is just a little better than the usual 1/(2Pi CoRg) calc since it includes the Zo of the PI.

[SilverFox]

I was referring to pin 8 as the measurement point for the bias. After changing out the OT the bias seemed to have stabilized, however until I cranked the amp. I believe the situation to be normal at this point as described above. It doesn't range as far from the base value but even if it does I guess that's normal as was mentioned.

Now I see the calculator doesn't work the way I assumed it did so I've got to go with the last best measurement- Sounds better with this OT then the other.... As far as the bias voltage at the 220K resistor, I'll be looking at that and several other factors when I start a new topic on this build and how to tweak the performance. I played through it yesterday using an emulator pedal and it sounds great. Totally supports the tubes in the tubes vs. transistors controversy. But then the Mosfet amp is from the 80's so not all things equal.

Thanks all and Happy Thanksgiving Day.

silverfox.