B-52 Biasing Problems

[dpm309]

Working on a B-52 AT100 that had a blown screen grid resistor (470 Ohm, 2W) on V9.  Replaced the burnt resistor and installed new tubes but when I checked the bias (using Doug's bias probe) I am getting a reading of over 55ma with the trim pot turned all the way in one direction.  This is way over the recommended 40ma.  I am suspecting a problem with the bias circuit.  I believe that this caused the screen resistor to blow and actually fall out of the board in the first place.  I am reading 455 VDC on all of the power tube plates.  I am not seeing any scorch marks (except for right under the burnt resistor) on the tube socket.

Should I suspect the trim pot or bias supply filter cap?  I have attached the schematic

Thanks,
Dan

[HotBluePlates]

Which position of SW5 did you use when making your checks?

One of the positions is cathode bias, with 0v supplied from the bias supply. The remaining settings have a full-bias and 91%-bias settings. So can we assume you're in the highest-voltage setting of SW5?

I would be surprised if a cap in the bias supply is bad, as the amp is not a 40 year old model.

If you can't get a large enough negative voltage with VR18 swept all the way over to C78 (the 220uF cap), then the easiest way to get more negative voltage is to parallel R125 (100Ω) with a ~1k-10kΩ linear pot.

Use a wire from the wiper of the pot and another from an outer lug, and adjust for maximum resistance before soldering in place. Adjust the bias pot (VR18) to its mid-point. Turn the amp on and slowly reduce the resistance of the added pot across R125 to get enough bias voltage to idle your tubes where you want them. Turn the amp off, and carefully remove the pot without disturbing the setting. Measure the resistance between the two wires, and find a standard resistor with the closest value to what you measured. Install this resistor across R125 permanently.

[dpm309]

HotBluePLate, thanks for the reply.  I am using the solid state rectifier position which is what B52 recommends.  Not sure which is the highest voltage setting.  The 3 settings are A, A/B, and solid state.  With the wiper all the way to the right, I am getting -40VDC on pin 5 of all 4 power tubes, while the bias probe reads 55ma.

[HotBluePlates]

...  With the wiper all the way to the right, I am getting -40VDC on pin 5 of all 4 power tubes, while the bias probe reads 55ma.

And with the wiper all the way to the left, I assume you get more idle current?

... Not sure which is the highest voltage setting.  The 3 settings are A, A/B, and solid state. ...

Flip the switch and find out.

I assume the "A" position is the cathode bias indicated in the schematic, but the schematic doesn't have switch labels.

Regardless, for the positions which you have fixed bias, the method I mentioned will work to correct the bias. But you will have to flip among the settings to make sure you don't have issues with one of the settings you don't already have it in.

[dpm309]

The solid state rectifier position gives the highest voltage.  Soldered in a 10Kl pot and I can't get a reading below 48ma with the trim pot all the way to the left.  This is a drop of only 8 +/- ma.  Tried a 25Kl and only dropped to 46ma.  Don't know if I should try a higher value or look else where for the bias problem.

Thanks,

Dan

[HotBluePlates]

... Soldered in a 10Kl pot ...  Tried a 25Kl ...

Do these mean you replaced the existing 5k pot (VR18) with these new pots? If so, that is not what I was describing, go back to the original 5k pot.

Any pot added in the procedure I mentioned is in parallel with the 100Ω resistor as an adjustable way to reduce its value. And, the added pot is only for temporary use to experimentally find the value needed.

[dpm309]

No, I didn't replace the original pot.  I installed the pots in parallel with 100ohm resistor.

[HotBluePlates]

Both should have had the same minimum current, as you should have been able to adjust either to be a dead-short across the 100Ω resistor. It might have been difficult to tell that though, as anything over ~1kΩ max would have seemed like all the effect was down around zero on the pot.

If you truly can get no more bias voltage with the 100Ω resistor shorted out, you should keep the 100Ω in place as-is, and instead make R128 (22kΩ) bigger. I might try 47k-56kΩ as a start, and see how mouch more bias voltage I can get.

What you should notice is that R125, VR18 and R128 form a voltage divider across the total of the available bias voltage. VR18 can only go so high and so low in that total voltage. So if you need more bias voltage (because you need to reduce tube current), then you have to either make the 100Ω resistor smaller, the 22kΩ resistor bigger, or both.

[Jack1962]

replace r124 with 220 ohm and VR18 with a 25K ohm this should allow you to lower the idle current on these tube . I have biased a ton of these I have never used one of Dougs bias probes to do so I have always used the shunt method , I have found that biasing this thing in A/B mode with solid state recto is your best bet , but switch it over to tube recto and double check the bias , I will almost bet you have to slightly adjust it .