adjustable NFB

[Leevi]

I have an issue with Marshall 1959 type clone. I have added an adjustable NFB pot (250K)
which causes noise maybe because of a DC offset. The noise can be eliminated by adding a cap in series with the pot
but then there comes an low freq. oscillation when the pot is turned off. I have added a 10K fixed resistor in series with the pot
but it doesn't help. Is there any way to get the NFB control  noiseless?


/Leevi

[shooter]

I'm looking at this schematic;
https://el34world.com/charts/Schematics/files/Marshall/Marshall_1959_jcm800_super_lead_power_amp_100watt_year_1988.gif

I believe presence is a form of NFB?
If you're adding another loop that will probably be problematic.

[Leevi]

The NFB wiring is like in the linked schematic. Presence control does not cause any noise. The noise of the NFB pot comes only when turning the pot.


Leevi

[PRR]

You don't want to make R24 much less than 100k, Marshall already dialed-in a lot of NFB and more will be unstable.

Yes, it scratches because there is +25V(?) at one end and zero at the other. The DC current in the wiper "scratches"; and you are upsetting bias on V3.

Use a 47k end-stopper and 0.5u-0.68u series cap. That gives about all the NFB it can stand, blocks DC, and is flat to 5Hz which *may* be enough to avoid instability. (10uFd 50V e-cap is sure to be flat to sub-Hz, and may be cheaper.)

[shooter]

I'll NEVER argue with PRR's assessment 
what is your "intent"?  are you trying to tame a Marshall enough to be a Fender ? 

[sluckey]

The NFB wiring is like in the linked schematic. Presence control does not cause any noise. The noise of the NFB pot comes only when turning the pot.

Is that really a problem? Most of the earlier Marshalls and Fenders have a scratchy presence pot only when turned. But the Presence pot is mostly a "set it and forget it" type control, so no big deal. I suspect your variable NFB will also become a "set it and forget it" control in about a week. I'd just let it be.

[Leevi]

Use a 47k end-stopper and 0.5u-0.68u series cap. That gives about all the NFB it can stand, blocks DC, and is flat to 5Hz which *may* be enough to avoid instability. (10uFd 50V e-cap is sure to be flat to sub-Hz, and may be cheaper.)

That works fine, thanks PRR. I used a 680n cap and the end-stopper can be even smaller than 47K (22K) and works fine.
It's maybe good to mention that the amp is using 2 power tubes instead of 4 which probably has some meaning in this.

Is that really a problem? Most of the earlier Marshalls and Fenders have a scratchy presence pot only when turned. But the Presence pot is mostly a "set it and forget it" type control, so no big deal. I suspect your variable NFB will also become a "set it and forget it" control in about a week. I'd just let it be.

I agree, it is not a problem for me but there are people that don't accept any scratchy pots


/Leevi