Bassman 100 grid leak and bias Q's

[clyde1]

I'm in the process of changing the dreaded bias balance system and noticed a few things that I'm wondering about. Grid leak resistors-Why 68K? In an earlier thread it was established 220K was the best for 6L6 and 100K for EL34.
I replaced them with 220K values but won't be changing the plate resistor values or cathode values of the PI. There's also a 220K resistor coming from one of the output grids that goes to a .01 cap and thence to the 100R resistor on the bottom tail of the PI, and on to ground. I've not noticed this before in any other Fender amp. Purpose? Thanks.

[pdf64]

'In an earlier thread it was established 220K was the best for 6L6 and 100K for EL34'
Not so sure about that, every data sheet I can remember seeing for either tube specs 100k as the max grid circuit resistance in fixed bias.
The bias balance circuit is great - it just needs an additional overall bias control adding. Either way, grid feed resistors of 100k or less are fine, it makes no difference to the bias voltage. Values less than 220k will just reduce the drive level a little to the power tubes, which is no bad thing, especially if you want to keep things fairly clean.
The 220k + 0.01uF between the 6L6 grid and feedback point is just an extra feedback loop, in addition to the one from the output transformer secondary. I don't see how it could cause a problem if it was lifted, so try it without to see if you prefer it.

[clyde1]

Thanks. I thought that might be another feedback loop and if I recall correctly, that's the only amp I've seen it in. It's disconnected with components in place so we'll give it a listen. I guess if you have to choose between bias balance and level, you have to go with level and hope that the output tubes are matched. As far as grid leak resistors go, .1 megohm is maximum indicated but why did Fender (and others) use 220K on almost every amp except later models with the different bias system?-less distortion? Thanks again.