6G3 bias mod

[Mars-Hall]

Wanting to add a bias pot to my Brown Deluxe.  The smallest value pot I have on hand is 50k. What value resistor should I use in series or should I opt and buy a smaller size pot? Will the larger value pot affect the operation of the tremolo?  Can I use a setup similar to my Concert Amp?

[jjasilli]

The schematic (Concert) does not match your amp (Br Deluxe).  The Concert bias shunt R shown above is is 56K; Br Deluxe is 22K.  What circuit do you propose to modify?

[Mars-Hall]

6G3 is the circuit I'm modding.  I only used this diagram for placement reference.

[shooter]

going by what Doug supplies, 10k and 50k you're in the range.

MAKE sure you check your "range" at tube socket WITHOUT tubes.  Make sure that range falls in the normal bias range for your type PA tubes.

[Latole]

Your mod should work, pot is ok .
When finish mod before put output tubes in, be sure you can read close to - 30 dc volts at output tubes screen to be safe.

It is safe to put a 200K resistor in case of pot's wiper bad contact.

I do bias mod very often on different amps

[Mars-Hall]

Thanks for all the replies.  Basically I'm just replacing the 22k ohm resistor with the 50k pot and 10k resistor in series.

[shooter]

200K resistor in case of pot's wiper bad contact

as long as you're soldering, this is a good safety addition.  not sure on 200k, but I self bias 

[sluckey]

Take a look at this bias circuit. It's the typical circuit found in most Marshall amps, but I'm using it in a Fender amp. If the pot fails the bias voltage simply goes to some maximum level. No danger of losing bias due to pot failure with this circuit.

If you will be getting ac source voltage from a 50vac bias tap change the bias range resistor to a much lower value, ie, 470Ω to 1KΩ.

     http://sluckeyamps.com/phoenix/phoenix.pdf

[sluckey]

Thanks for all the replies.  Basically I'm just replacing the 22k ohm resistor with the 50k pot and 10k resistor in series.

I think that's the simplest way to do it...

[Latole]

I'll put a 100 mfd 100volts filter caps in a bias circuits. Old Fender amps use too low values

[66Strat]

IMO, the schematic that sluckey posted is the best/safest way to implement the bias pot. Should the wiper connection fail, the bias becomes more negative/colder.

[Latole]

IMO, the schematic that sluckey posted is the best/safest way to implement the bias pot. Should the wiper connection fail, the bias becomes more negative/colder.

I agree,
 2 X 10 mfd are too low IMO

[Mars-Hall]

I'll put a 100 mfd 100volts filter caps in a bias circuits. Old Fender amps use too low values

Yes already have a 100uf/100v cap for this purpose. Thanks

[sluckey]

A 2-pole filter using 10µF caps is really all you need for a well filtered, low current bias supply. Big caps in the bias circuit are unnecessary and can even be detrimental if they charge slower than the B+ rises at turn on.

[Mars-Hall]

A 2-pole filter using 10µF caps is really all you need for a well filtered, low current bias supply. Big caps in the bias circuit are unnecessary and can even be detrimental if they charge slower than the B+ rises at turn on.

I should use a lower value than 100uf/100v cap there, to help protect the tubes?

[Latole]

A 2-pole filter using 10µF caps is really all you need for a well filtered, low current bias supply. Big caps in the bias circuit are unnecessary and can even be detrimental if they charge slower than the B+ rises at turn on.

I should use a lower value than 100uf/100v cap there, to help protect the tubes?

I desagree. 100 mfd for bais circuit come from some Gerald Weber books if my memory is good
There is no tube in bias circuit.

We all know that there is not just one recipe for making amps.
And this is what is causing the GAS   

[66Strat]

A 2-pole filter using 10µF caps is really all you need for a well filtered, low current bias supply. Big caps in the bias circuit are unnecessary and can even be detrimental if they charge slower than the B+ rises at turn on.

I should use a lower value than 100uf/100v cap there, to help protect the tubes?

The law of diminishing returns come into play here. Sixty cycle AC ripple with a single pole 25K ohm /10 uf RC filter is -45.5 db with an RC time constant of 10 milliseconds. Bumping the capacitor up to 100 uf attenuates the ripple -65 db, but increases the RC time constant to 100 milliseconds. With the change to a 100 uf capacitor, DC ripple is decreased about 50% with a ten fold increase in the RC time constant. In a push/pull output circuit the change in AC ripple is audibly negligible. However, the adverse change in the time constant could cause the tube to be without bias during warm-up.

[Mars-Hall]

A 2-pole filter using 10µF caps is really all you need for a well filtered, low current bias supply. Big caps in the bias circuit are unnecessary and can even be detrimental if they charge slower than the B+ rises at turn on.

I should use a lower value than 100uf/100v cap there, to help protect the tubes?

The law of diminishing returns come into play here. Sixty cycle AC ripple with a single pole 25K ohm /10 uf RC filter is -45.5 db with an RC time constant of 10 milliseconds. Bumping the capacitor up to 100 uf attenuates the ripple -65 db, but increases the RC time constant to 100 milliseconds. With the change to a 100 uf capacitor, DC ripple is decreased about 50% with a ten fold increase in the RC time constant. In a push/pull output circuit the change in AC ripple is audibly negligible. However, the adverse change in the time constant could cause the tube to be without bias during warm-up.

Ok, especially without a standby switch, something closer to the original value would be ideal.  I might have a 50uf/100v cap but probably should just order a 25uf/100v if available.

[66Strat]

The bias supply would not be impacted by the standby switch. That's a whole nuther can-o-worms.

[Mars-Hall]

The bias supply would not be impacted by the standby switch. That's a whole nuther can-o-worms.

Not wanting to start a theories debate, just want to know what's best practice for the longevity of my amp.

[66Strat]

The 50 uf cap should be fine. With a slow warm-up rectifier, the 100 uf cap should be OK. It just doesn't make that much of a difference with regards to hum. In a silicon rectified circuit, the RC time constant with regards to bias could be significant. A standby switch could, if used properly, mitigate this risk in silicon rectified circuits. In tube rectified circuits, hot switching concerns associated with the standby come into play.

[jjasilli]

I disagree with the use of a 50K pot in this circuit, which is the basis for my Reply #1 above.  Your bias circuit requires a shunt R of only +/-22K.  A 10K R in series w/ a 50K pot = 60K of shunt resistance.  You'll be able to use less than 1/2 the rotation of a linear 50K pot.


IMHO this circuit calls for a 10K or 25K linear pot with a commensurate series R.  This will give much more usable pot rotation.  Also, tiny changes in the shunt resistance yields large changes in the bias voltage.  A 50K pot may be difficult to dial-in, compared to a 10K pot.

[PRR]

> 2 X 10 mfd are too low IMO

While it is cut-off in that snippet-pic, the feed resistor is hundreds of Kohms (not dozens). Even 10uFd is a "LOT" of capacitance when fed from so many Ohms.

Not trusting my crayon I asked an idiot. PSUD usually gets good answers if the question is phrased properly. Two 100uFd caps makes a v-e-r-y long rise time. As high as I would like to see. Is it necessary? Two poles of 10uFd makes 20 milliVolt ripple which is plenty good for push-pull (and maybe for some SE work). And few-Second rise time.

Voltages are inverted cuz PSUD only goes this way.

[DummyLoad]

5th TC for the 6G3 bias circuit with 100uF is about 5 seconds. tube warm-up is typically around 10 seconds. 

5th TC for the same circuit with 25uF it about 1 second; with 47uF about 2.5 seconds.

with 10K fixed R and 10K pot - range is about -14V to -26V


with 15K fixed R and 10K pot - range is about a-20V to -33V

those done on a SIM - your results should be similar, but not exact.

suggest the following:  100K 2W > 1n4007 > 47uF@63V > 15K 500mW > 10K-L 250mW pot. 


--pete 

[shooter]

5th TC for the 6G3 bias circuit with 100uF is about 5 seconds

everyone should try adjusting the 5th TC to 1% without messing up the other  4    (one of many crazy cals from long ago :)

[jjasilli]

5th TC for the 6G3 bias circuit with 100uF is about 5 seconds. tube warm-up is typically around 10 seconds. 

5th TC for the same circuit with 25uF it about 1 second; with 47uF about 2.5 seconds.

with 10K fixed R and 10K pot - range is about -14V to -26V


with 15K fixed R and 10K pot - range is about a-20V to -33V

those done on a SIM - your results should be similar, but not exact.

suggest the following:  100K 2W > 1n4007 > 47uF@63V > 15K 500mW > 10K-L 250mW pot. 


--pete

Nice work, pete! Yes, use a 10K pot.  I like to use a multiturn pot for more precise dialing; but a regular pot is good enough.  For a precise circuit:  set the linear bias pot to 1/2 rotation.  Jumper it temporarily in series with with another 10K pot.  Adjust the temporary pot (add a 5K fixed R in series if necessary) so that your desired bias point is hit with the permanent bias pot at 1/2 rotation.  This gives max adjustability, either way, in the future as tubes age or get replaced.  Permanently use a fixed Resistance = test pot rotation value + the value of any test fixed R.  If the permanent fixed Resistance is an unusual value, compromise and use the closest standard value fixed R; or combine 2 fixed R's in series, or in parallel, to achieve the precise value.