• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Servo Bias Boards

I'am using the bias board from

Module AB-Q for 4 tubes, PP & PPP & Single Ended amps, requires 6.3VAC & bias supply from the amps circuit. Not suitable where the 6.3VAC filament supply is referenced to ground, use the ABF-Q below. ...

for the amp I am building (class AB).

I think its a clone of
bias.png

The resistors and output circuit seem to match. The first time I tried it I connected the negative supply straight from the rectifier cap. The result was the bias got modulated by the 50Hz component there causing some modulation. The second time I added a 2k + 470uF to smooth the negative supply. However it takes about 18ma and this was enough to cause all the grid voltages to become too positive. So I settled on 470R. However the circuit seems to offer little rejection of the negative supply.

So I can see the voltage on the grids change if the mains dips (which mine does). Maybe I need to stabilise the negative supply, but I was expecting the grid voltages not to be a function of the supply.
 
Sorry for your troubles.

With all that differential gain of those op amps, a very small amount of ground loop hum voltage goes a long way (causes problems).
The problem may be more than just the negative supply ripple.

Now you can see why I use simple circuits.
 
Last edited:
I'am using the bias board from

Module AB-Q for 4 tubes, PP & PPP & Single Ended amps, requires 6.3VAC & bias supply from the amps circuit. Not suitable where the 6.3VAC filament supply is referenced to ground, use the ABF-Q below. ...

for the amp I am building (class AB).

I think its a clone of
View attachment 822633

The resistors and output circuit seem to match. The first time I tried it I connected the negative supply straight from the rectifier cap. The result was the bias got modulated by the 50Hz component there causing some modulation. The second time I added a 2k + 470uF to smooth the negative supply. However it takes about 18ma and this was enough to cause all the grid voltages to become too positive. So I settled on 470R. However the circuit seems to offer little rejection of the negative supply.

So I can see the voltage on the grids change if the mains dips (which mine does). Maybe I need to stabilise the negative supply, but I was expecting the grid voltages not to be a function of the supply.

AB requires about 18 mA current from the BIAS supply. It must be taken into account.
Pavel
 
Thanks you are the guy from audioamp.eu - many thanks for your help getting it working.
I had a look on ltspice. There may be a quite simple mod. Basically removing the original integrating cap (22uF to ground). Then replacing with an active feedback circuit from the output. The cap needs to scale with the gain (x20) to 1uF but now needs to be 100v so a metal film will do. I need to add a small cap to ground for HF stability and a protection diode. Now if the negative supply changes the op-amp will hold the output at the same voltage. Think that works. I will modify a single channel to check.

Screenshot from 2020-03-06 15-31-58.png
20200306_141044.jpg
 
Thanks you are the guy from audioamp.eu - many thanks for your help getting it working.
I had a look on ltspice. There may be a quite simple mod. Basically removing the original integrating cap (22uF to ground). Then replacing with an active feedback circuit from the output. The cap needs to scale with the gain (x20) to 1uF but now needs to be 100v so a metal film will do. I need to add a small cap to ground for HF stability and a protection diode. Now if the negative supply changes the op-amp will hold the output at the same voltage. Think that works. I will modify a single channel to check.

View attachment 822832
View attachment 822833

ok
I see you use the HVin pad :)
 
I think I will leave boards as is - its not actually causing any real problem now I have filtered the supply - I don't want to break something as its all working. I could add a negative regulator but don't have much excess voltage to play with. Its just I seem to share the mains supply with the cesspit pump at the bottom of the track where I live.

20200306_140505.jpg
 
I think I will leave boards as is - its not actually causing any real problem now I have filtered the supply - I don't want to break something as its all working. I could add a negative regulator but don't have much excess voltage to play with. Its just I seem to share the mains supply with the cesspit pump at the bottom of the track where I live.

View attachment 822842
it's ok, I would leave it that way
 
Screenshot from 2020-03-09 16-24-24.png
Added a negative regulator to sort out the mains dips. I have about 8v across the reg to cope. If this drops to zero then the output will just follow the input. Output is about -63v. Don't want to go below this as bias boards need about >4v drop and 6550 needs about -53v for 38ma.