custom bias circuit issue

[xtian]

My bias circuit is not working right. It charges right up to -58vDC, but when turning the bias adjust pot, the voltage changes VERY slowly. Schematic attached.

When I added a load (470R 10W) between -bias and ground, that solved the issue--voltage tracked closely with turns of the bias pot. But even without a load, shouldn't my voltage be draining thru the bias pot and 10K resistor?

[sluckey]

Remove the diode. It prevents the cap from discharging through the bias pot and resistor.

[xtian]

Thanks, will try. I'm off to play services now! Merry Christmas!

[xtian]

Yes, that solved it. Thank you!

[sluckey]

I've never seen a diode used like that in a real circuit. But Merlin talks about using the diode and a parallel resistor. Makes sense to me but you need the resistor to have any benefit. Mesa Boogie would probably like this idea.

Using only the diode allows the bias cap to charge up quickly just as if the diode were not in the circuit. But once the cap is charged to the voltage applied from the wiper of the pot the diode becomes reverse biased. If you try to decrease the voltage with the pot the diode just becomes more reverse biased. The only discharge path for the cap is it's own leakage resistance. That cap may hold it's charge for hours, even days. Putting the large resistor across the diode will allow the cap to discharge slowly and make the pot usable.

[xtian]

You're right on, Steve. I used Merlin's circuit, but removed that bleeder 100K resistor, without removing the diode.

[SilverFox]

I had built an amp that was working fine until I flipped the power on and off rather suddenly and the HT fuse blew. But so did one of the rectifier diodes. Took me a while to find the bad diode since it still read 200 ohms- in both directions. But that makes sense as the bias circuit I used was just a typical design.

silverfox.