1967 Fender Bassman AB165: Oscillation on Normal Channel w/ bright switch engaged.

[cluster]

Hello Everyone, I need some schooling on oscillation. I have a 67 Bassman AB165 which is oscillating on the Normal channel with the bright switch engaged. The oscillation comes on when the volume is above 3 and the treble is turned up past 5. No oscillation with bright switch off. Earlier I did remove the suppresser caps across the plate resistors as well as restoring the PI and power section to AA864 specs (47k/47k plate load resistors and 100k/100k bias splitters-feed). I also removed the 470k feedback resistor (plate to grid) on the 3 rd tube where the normal and bass channel meet. However, I did put shielded cable on the input jacks and volume pots (thought that would help).

Can anyone help to figure out where the oscillation could be coming from? Note: If I touch the hot side of the brightness cap connection to the volume pot (from the bright switch), the oscillation disappears.

[nickb]

All of the things you did have significantly increased the gain, especially at higher frequencies. There is always going to be some tiny signal from the output looping back to earlier stages. By increasing the gain you increased this loop gain enough to make it oscillate. Screening the input leads was a good thing and that (should) reduce the amount of fed-back signal and so lowered the loop gain.

Your body is a bit like a capacitor to ground. So, when you touch the bright cap you shunt HF and reduce the loop gain enough to stop the oscillation.

You can start to change things to reduce the loop gain again. One way would be to put capacitor(s) back across the plate(s), increasing the value until the problem goes away. If that kills the highs too much you'll need to figure out how the signal is getting fed back or replace the 470K - try a higher value and see what you can get away with. Start at the most sensitive place i.e. the input. With no instrument plugged in, does it still happen - this shorts the input to ground. If no, then the pickup is in the input wiring. If yes, then proceed to look at the tone stack wiring and see how you might clean that up.

[Justin Thomas]

I think that feedback loop on V2 is part of the fun of this model...
Everything nickb said is true. Also, this is the beginning of Fender's sloppy lead dress practices and other SF shenanigans (that bias circuit & plate feedback on the 6L6s?). Not the late SF years, but the start of it all. Check your amp against the layout closely - I own a slightly earlier example than yours (cloth wire in mine) and have learned the hard way that wire placement matters. Small changes can have big results... Good luck!

Justin

[cluster]

Thanks so much for both your answers. I had a 330k lying around and put in the plate to grid feedback position where the 470k used to be. Problem solved. I'll get back to it and slowly raise the resistance to where I think sounds good. I appreciate your input very much. I'll follow up when i get back into that amp.

PS: I removed those 220k feedback resistors coming off the output tube plates a long time ago. Funny thing is that this amp had a mixed fix/cathode bias as stock. I converted it to a fix bias and converted the balance bias to an adjustable one. This is a fun amp to work on. There is an unused triode just waiting to be used ;-)

[Justin Thomas]

Sounds like you had yet another example of an infamous Fender "hybrid" - you had a mashup of an AB165 & an AC568. The "mixed bias" is the giveaway for the ---68 series. You also seem to have a very rare type of wire in yours, the really skinny cloth stuff - it's different from the usual. I know some don't worry about the small stuff; I kinda enjoy it.

And, nice amp!

Justin