6G6B adjustable bias

[Greg_L]

Hello friends. I have a question....

I'm about to start a 6G6B build into an old junky SS combo I had laying around. This will be my first build from scratch not using a kit. I've already sourced all my parts and stuff.

Anyway I'd like it to have an adjustable bias. I swear I've searched and researched and read threads and seen ideas and I'm just a little confused as to which would be the simplest way to add a pot to the stock bias circuit. I've seen many different ways to do it. I think I've overloaded my brain and given myself paralysis by analysis. Should I use a Blackface style bias circuit? Or where would I tweak the 6G6B bias supply to give me some adjustment? Any help or ideas is appreciated.

I'm attaching the schematic and layout.

[vintagekiki]

6G6B adjustable bias

[Greg_L]

6G6B adjustable bias

Haha seems too simple to be true!

[stokes]

Yep, its that simple.I have done too many of these mods to count.Only difference is when I do this mod I like to use a cermet pot of 2X's the value of the resistor in the circuit.In this case I would replace the 27k fixed resistor with a 50k pot.That way it puts the setting approx. in the middle of the range of adjustment.

[Greg_L]

Yep, its that simple.I have done too many of these mods to count.Only difference is when I do this mod I like to use a cermet pot of 2X's the value of the resistor in the circuit.In this case I would replace the 27k fixed resistor with a 50k pot.That way it puts the setting approx. in the middle of the range of adjustment.

Thanks for the tip!

And sorry for the late reply...been out for a bit.

[pdf64]

... I would replace the 27k fixed resistor with a 50k pot.That way it puts the setting approx. in the middle of the range of adjustment.

A bias voltage that can be adjusted to 0 seems 'non-ideal', akin to fitting a self destruct switch!
Why not add a resistor to limit things to a safe range?

Haha seems too simple to be true!

Sorry to dissent, but I think that bias adjustment arrangement is too crude to give satisfactory results.

A closely similar arrangement is shown on a Mojo kit layout, and has been reported as having a poor range of adjustment, eg https://el34world.com/Forum/index.php?topic=27767.0

It operates by loading down the source impedance of the bias supply, the most significant part of which is probably the 1k resistor feeding the rectifier diode.
Hence the voltage won't be reduced much until the loading is significant, by which point ripple voltage will also have been increased by that same loading mechanism.
A load formed by 25k variable resistor in series with a 10k fixed resistor will have a rather limited range of adjustment, because even with the variable resistor set to 0, the 10k resistor won't load the bias supply much. When set to 25k, the now 35k load won't increase the magnitude of the bias voltage much.

It would be much better to use a proper potential divider pot, with a safety resistor, eg along the lines of the 'standard' arrangement from http://www.valvewizard.co.uk/bias.html

[vintagekiki]

Sorry to dissent, but I think that bias adjustment arrangement is too crude to give satisfactory results.

Dear colleague
Leave unnecessary theory. Had Leo and Jim deal with theory, they would never have become Fender and Marshall.
All practical solutions are outlined in vintage amps.
After all, don't believe it, check it out.
The proposed solution ( #2 ) is technically correct.

[pdf64]

Dear colleague
Leave unnecessary theory. Had Leo and Jim deal with theory, they would never have become Fender and Marshall.
All practical solutions are outlined in vintage amps.
After all, don't believe it, check it out.
The proposed solution ( #2 ) is technically correct.

But that's comparing apples to oranges, the Marshall arrangement noted has a much higher supply impedance feeding the variable loading, and 2 stages of filtering, so it works great, with plenty of adjustment and low ripple.

The AA165 arrangement shown is what I am suggesting is better than that shown in post #2, ie it uses a pot, not variable loading on to a low source impedance. A safety resistor would be beneficial in the case of the pot failing.

[Greg_L]

A bias voltage that can be adjusted to 0 seems 'non-ideal', akin to fitting a self destruct switch!
Why not add a resistor to limit things to a safe range?



Sorry to dissent, but I think that bias adjustment arrangement is too crude to give satisfactory results.

A closely similar arrangement is shown on a Mojo kit layout, and has been reported as having a poor range of adjustment, eg https://el34world.com/Forum/index.php?topic=27767.0

It operates by loading down the source impedance of the bias supply, the most significant part of which is probably the 1k resistor feeding the rectifier diode.
Hence the voltage won't be reduced much until the loading is significant, by which point ripple voltage will also have been increased by that same loading mechanism.
A load formed by 25k variable resistor in series with a 10k fixed resistor will have a rather limited range of adjustment, because even with the variable resistor set to 0, the 10k resistor won't load the bias supply much. When set to 25k, the now 35k load won't increase the magnitude of the bias voltage much.

It would be much better to use a proper potential divider pot, with a safety resistor, eg along the lines of the 'standard' arrangement from http://www.valvewizard.co.uk/bias.html

Very interesting, thank you. I like it.

[vintagekiki]

With "soft power supply", in the filtration is capacitor of smaller value (10u), so Pi filter are needed.
With "hard power supply", one capacitor of higher value (47-100u) is sufficient, voltage changes is relatively little with change load.
Question
What happens if bias voltage can't adjust with schematic shown in post #6 if C1 have more voltages than necessary.

[pdf64]

…
Question
What happens if bias voltage can't adjust with schematic shown in post #6 if C1 have more voltages than necessary.

If I understand the query correctly, I suggest that a suitable dropper resistor be added, as per R1 of the half wave circuit

[vintagekiki]

If I understand the query correctly

Yes you understand query correctly.
Suggest solution is well, only how resistor R1 affect on source impedance?
Personally, I think that slight change of P1 or R2 solves all problems.