Biasing a Leslie 122 Amp

[patlaw]

What plate current would you choose for the 6550s or KT88s in a Leslie 122 amplifier with 420 volts on the plates and 25 volts on the cathodes? I have one tube running hot.

[Helmholtz]

Amp is cathode biased, so no easy way to individually adjust bias.
These circuits need matched tubes.
Cathode voltage looks good, meaning a total common cathode current of 166mA.
Now if one tube takes more current, the other will get less.

I have one tube running hot.

Judging from what? Is it redplating?
If so, does the redplating follow the tube if you switch positions?
If not, check the coupling cap for the position where it replates.
Also check screen voltages as well as cathode resistor and capacitor.

To answer your question, the 6550 datasheet suggests a total cathode current (for 2 matched tubes) of 180mA, meaning 90mA per tube.
https://frank.pocnet.net/sheets/127/6/6550.pdf

[patlaw]

Judging from what? Is it redplating?

No, it's not red plating. It's based on measuring the current through the cathode using a Bias Master socket.

If so, does the redplating follow the tube if you switch positions?

One tube is dissipating more power than then other, and the higher dissipation moves with the tube.

If not, check the coupling cap for the position where it red plates.
Also check screen voltages as well as cathode resistor and capacitor.

The coupling cap is new. The cathode resistor is dead on the money, and the bypass capacitor is new.

To answer your question, the 6550 datasheet suggests a total cathode current (for 2 matched tubes) of 180mA, meaning 90mA per tube.
https://frank.pocnet.net/sheets/127/6/6550.pdf

One tube is running at 70mA while the other is at 100mA. I'm going to put the amp back on the bench and measure the screen, cathode, and plate voltages for confirmation. In short, it looks like there's an imbalance in the tubes, which means things are pretty normal. I asked because I rarely, if ever, hear anyone mention checking the bias on Leslie amplifier output tubes, particularly since it's not adjustable. There are a lot of technicians who just order a rebuild kit and replace everything. I doubt most of them actually verify the plate dissipation. I have a matched pair of tubes I'm going to try, as well.

I wonder if it would be worth modifying these amps for adjustable bias to get additional life out of the original tubes that many of these still have. Then again, like Uncle Doug has started saying lately, many musicians don't like the sound of perfectly balanced tubes.

[Helmholtz]

Bias balancing could be realized with a 300R/10W wirewound pot at the cathodes, wiper going to ground.

[Helmholtz]

There's some risk involved with the balancing pot, as setting to the extremes might damage a tube.
A safer solution would be a 100R balancing pot on top of a 100R common cathode resistor.
Or course, that will limit the balancing range.

You'll also need separate caps from each cathode to ground.

[patlaw]

Of course. Now I will ask a really dumb question. In this circuit, with the values shown in the schematic, assume 75ma cathode current (plate current plus screen current) per tube. How much of the cathode current is screen current?

[Helmholtz]

Datasheet says 5mA screen current with 85mA plate current, so around 6%.

[patlaw]

Datasheet says 5mA screen current with 85mA plate current, so around 6%.

Which datasheet are you using?

[Helmholtz]

BTW, why only 75mA cathode current?
Cathode biased amps are typically biased between 80% and 100% plate dissipation.
I'd go with 80% as the datasheet suggests.

[Helmholtz]

Which datasheet are you using?

The one I linked above.

[Chuck H]

BTW, why only 75mA cathode current?
Cathode biased amps are typically biased between 80% and 100% plate dissipation.
I'd go with 80% as the datasheet suggests.

Just to elaborate on this in case it helps anyone browsing topics or following...

Cathode bias amps are DRAWING their bias voltage via the cathode resistor. The positive voltage on the cathode in relation to the grid at a 0V reference relates the same to the tube as a negative voltage on the grid (as it relates to fixed bias circuits). In class AB operation (class A not being considered here, yet), when the amp begins conducting signal current is increased. In a cathode biased amp that current is pulled through the cathode resistance and greater current introduces a greater voltage drop across the cathode resistor. That is, higher voltage across the cathode resistor. More positive voltage across the cathode resistor means a higher grid negative to cathode relationship. This is a cooler bias condition than the standing bias at idle. For this reason most cathode biased amps idle at a hotter bias condition than fixed bias amps to target the desired current, analogous to cathode voltage, when conducting signal.

EDIT: Also a consideration, in cathode biased circuits the cathode voltage should be subtracted from the plate voltage to realize the actual WORKING plate voltage. The whole tube is operating under an elevated condition with respect to the cathode voltage.

[patlaw]

Another question is looming large in connection with Leslie 122. From the factory, the AC input (from the organ) has .05µF on each side to ground. It occurred to me that they should probably be replaced with safety caps. They're definitely need for noise suppression. I'm having trouble identifying which safety cap to use. There is a good write up at https://www.justradios.com/safetytips.html. One vendor is selling an X1Y4 for $3.25 each. I can't find the original vendor of that part. https://ssl.tonewheelgeneral.com/bui...item_no=AC-CAP. I'm looking for a wholesale source. I am also not convinced that the part he sells is the best one for the application. Any comments and suggestions welcome.

[g1]

For line-to-ground you want classY.
Y4 is the lowest voltage rating, Y1 the highest.
These Y2 rated (150 to 300VAC) are available at digikey and maybe elsewhere, but just an example: PHE850EB5470MB14R17
There are many caps out there that will meet the spec., but you want a package that will work for your application.