Grid clamping/bias excursion calculator

[octal]

Has anyone seen this calculator for grid bias excursion in power amps?

Bass Response and Grid Bias Excursion Calculator

I'm wondering what I should be aiming for... I would guess that a short/fast bias recovery time would be good, the bass cutoff frequency calculator is obvious enough, but I'm less sure about the bias excursion ratio target and bias excursion time.... anyone?

Nathan

[Chuck H]

Yes, but what does it sound like?

Point is, crossover distortion, which is a common side effect of grid loading, is a significant part of the tonal character of overdriven guitar amps. The "recovery time" and phase shifts that occur, mostly due to the lower frequencies (larger coupling caps) can create "swirl" which is also considered desireable by some players. By reducing the coupling cap size or drive voltage you can reduce the unwanted side effects but you are also reducing the desireable side effects. Guitar amps (IMHO) aren't about designing the most linear or efficient amp, but rather a combination of those signal distortions that naturally occur which creates a pleasant effect instead of an unpleasant one, if you get my meaning. This can include intentionally operating an amp in such a way as to cause grid loading, to some degree, in combination with other signal distortions.

I personally don't favor crossover distortion. But I have heard amps with significant crossover distortion and even grid blocking which, in the hands of an inspired player, sound great. I;m just saying that the grid loading effect has been a common "malady" and a fact of life in many guitar amps for a long time and some players consider it, in moderation or even extreme amounts, part of the formula for good tone.

You may be one of these players that likes a little (or even a lot of) crossover distortion and grid blocking in the OD tone of your amp. So I encourage that you build, scope and listen to see for yourself if you find this effect offensive before taking steps to eliminate it from a design. Some of the most notorious amp designs OD the piss out of the power tubes (usually resulting in significant grid loading) with relatively low amounts of preamp OD. Designing an amp with the same OD capacity without driving the power tubes as much would mean the need for more preamp distortion, which most players find thin, scratchy and undesireable (just my observation, not my opinion).

Now I know you were looking for a hard factual way to determine how to use the info in the link, But I can't really quantify it due to the above considerations.

Chuck

[Loogie]

If you're in a position where you can experiment it might not be such a bad a idea to calc some values and hear how they sound. Maybe go from one extreme to the other; try to induce grid blocking and try to avoid it. Hear the numbers...

[octal]

Chuck,
I totally agree with you- with guitar amps, if it sounds good it IS good. I built two amps with essentially no crossover distortion/blocking (one using special high voltage op amps as drivers and one with cathode followers DC coupled to the outputs to allow AB2 operation.) The clipping/ overdrive was so soft, it sounded more like gentle compression, had no bite or feel to it at all. I ended up modifying both amps to allow a touch of x-over distortion.

I guess what I'm wondering is if the online calculator might be useful for predicting what might sound good, rather than having to go though many iterations the old fashioned cut and try way, but in order to do that, I need to know what might be "optimum" (in terms of guitar tone) rather than "optimum" in a Hi-fi sense.

Nathan

[Chuck H]

That's a tough call. Many trusted circuits have different combinations of distortion components that make for a unique sound. IMHO a little (or none for me personally) grid loading is OK, but even a little more than what most Marshalls have sounds bad. I suppose you could plug in "Marshall" type parameters to get a bench mark figure, but that may even be out the window when you compare to a Matchless, or whatever. The biggest mistake I see in noob high gainers is that they don't accoun't for bass attenuation throughout the circuit (big decoupling caps, PI or otherwise) and they don't account for the added drive to the PI and power tubes. The big coupling caps and lack of attenuation before the PI makes for a huge amount of bottom end (high current stuff) creating the square wave and this makes for mushy tone. Try to push enough top end to balance this and the whole thing starts to sound boomy and harsh. And at it's worst grid blocking occured and makes the distortion choppy and farty.

I have to stick with "what sounds good is good" here. I haven't had much joy designing blocking distortion, or even moderate amounts of crossover distortion into my builds. So I use circuits that I've fine tuned to give me a good square wave without the added hash. But a 1959, JTM45 Marshall or a tweed Bassman are pretty much the benchmarks and they all exhibit some grid loading. And some of the best sounding examples are the biggest offenders too. I'm still learning how to incorporate the right amount of flawed and failing amp tone into my designs too. So any hard numbers are beyond my design philosophy so far I guess. But I might use a Marshall model as a guide and experiment.

Chuck