Peavey 5150 Combo Adjustable Bias

[dorrisant]

I have one of these on the bench and need to add this mod to it. I see a few places where the bias mods are laid out for the original 5150 and the 5150 II but nothing for the combo, which is different, of course.

Below is the schematic for the bias supply. I had initially replaced R1 (47k) with a 50 k pot, but thought I should hang a resistor off the pot to ground for a "hottest bias" limit. The problem is that even with all of that I was only getting the bias to move between -50 and -52v. I want to get more of a swing, say from -42v up to about -52v. What would you guys suggest? I was hoping for some insight here because it is a real PITA to get this board in and out of the chassis. I'm hoping to change/add a few parts and button it up for good. If it doesn't seem feasible then I would probably move on to install a Merlin type bias supply and abandon the onboard circuit.

Btw... changing all of the .047µF signal caps to .022µF will get this amp to sound pretty much like the original 5150s. This along with a slope control makes for a much better amp.

[pompeiisneaks]

Basically I think you'd need a slightly larger pot, but you can adjust the resistor to ground to be a bit different up or down to adjust the range stopping points. Experiment up and down a few k to get where you need it. Also don't forget to tie the wiper to ground so that if somehow the wiper comes off, you still have the full path of the pot to ground so the tubes don't lose bias voltage and go into runaway.

Usually what I've seen is for something like a 47k resistor, replace it with a 10k pot and say a 39k resistor below, so you have a very large safe resistance range beneath it, to keep the bias from going too hot if something goes 'shorted' on the pot, and then you have the 10k swing range.

~Phil

[pompeiisneaks]

I could even add that you could mess with the 47k resistor to get it to the min negative voltage you'd want, and then put a pot on top of that. I.e. maybe the lower resistor could be a bit hot at say 72% or 75% whatever you'd hope is the 'hottest' you'd ever bias it, and then add the pot on top of it so you can lower the negative bias with more resistance. Hope that makes sense?

~Phil

[dorrisant]

I get what you're sayin' Phil. I tried all kinds of combinations of components to get some swing to it... Last thing was a 100k pot with a 8k2 resistor between it and ground. Each tube is pulling 47mA at 486v. That's a bit too hot for what the customer wants... and the bias won't go any lower (more -voltage). I think I'm gonna pull off of the HT and build a supply like one seen in Merlin's book, swap the pot for a 10k and call it good. I know how that one works.

[Synchu]

In order to get a larger swing of this voltage divider you would need to lower the value of the resistor you're hanging to ground of the pot.

Suppose you have 50V coming in with 470R and 2K7 to ground you will get approx 42V out with 50K resistor at 0.

You may want to play with this calculator - http://www.ohmslawcalculator.com/voltag ... calculator quite useful

Hope this helps.

Niki

[pdf64]

To cool off the operating point, the available bias supply voltage should be increased; to do that, need to increase the values of C65 and R112.
There's very little voltage normally being dropped across R112, so increasing the value of R114 (eg with a 50k pot) can't ever make much difference.
You can't get more out than you put in
Increasing C65 will let more Vac to the rectifier, hence bias supply voltage will increase; if C65 is doubled, then the voltage should about double, hence R112 would need to be increased some.
Make sure that the cap voltage ratings aren't exceeded.

[martin manning]

You have about -50V available with the circuit as it is, so there is no lower (more negative) voltage to reduce idle plate current. The two caps and diodes are a doubler circuit, not a capacitive voltage divider, so you can’t lower voltage that way. Have you committed to some other bias circuit yet? This is a PCB amp, so your options are kind of limited, no?

[Leo_Gnardo]

Might think about selecting a different set of output tubes, ones with lower emissivity. Could be the ones you have are at the high end of the emissivity range.

[dorrisant]

No, I haven't committed to another circuit yet. Project is in a holding pattern until resolution and the customer is aware. I would be willing to modify this circuit, but every time I remove the board to do changes, I'm taking a risk on breaking a connection somewhere and I really don't want to chase that problem. I was hoping someone here had done this before and had a proven solution. Short of that, someone who could run a Spice sim or something like that could give a definitive answer so I could minimize the handling of the delicate PCB arrangement.

Thanks for the advice guys... I'm all ears.

Also, this is a brand new set of burned in and matched JJ 6L6GCs... The customer wants this mod so he can tube roll. Going for a different set of tubes that would happen to bias correctly would not really fall in the parameters of the customers request. Besides, this looks like a good time for some Tony learnin'.

[Synchu]

After the doubler circuit, there's a voltage divider I was speaking about, formed by R112 and R114.

A quick and dirty simulation with a bunch of assumptions is attached. Green at the top is X = 2k, violet at the bottom = 47k

Niki

[sluckey]

A voltage divider will not help. He needs more negative bias voltage, not less.

[pdf64]

...The two caps and diodes are a doubler circuit, not a capacitive voltage divider...

Doh, of course, thanks for the correction

[martin manning]

What you could do is isolate the doubler supply grounds and take them to the -25V at the bottom of C71. You could try this out by lifting the ground ends of CR23 and C69, and run a jumper from those leads to the -25V. That would pull the minimum bias voltage down 25V, and then you could add a pot to make the bias adjustable. The easiest way to do that would be to replace the 470R with a 25k trim pot, shorting the wiper to one end, and replace the 47k with a 22k.

[pdf64]

Link to full schematic for this amp (so many 5150 versions)?

[martin manning]

https://www.electronica-pt.com/esquema/ ... mbo-28123/