removing excessive AC hum in a Sound City 50 plus

[methodofcontrol]

what are the obvious things to do or to check?

[txstrat]

Tubes? Caps?

[methodofcontrol]

new tubes. only electrolytic caps i didn't replace are the 100uf on the bias and the 200uf's near the standby switch.
i guess i was looking more for info on how to track down AC hum to isolate where it may be coming from.

[pdf64]

Have you checked the bias? The bias supply is marginal (a FWB or voltage doubler can resolve this) and so the power amp may just be running very hot.
If ok, one approach would be to remove all pre-amp tubes and check for hum. If so then investigate around the power tubes, eg bias supply cap, reservoir cap.
If not then fit phase splitter and continue the process. Pete.

[txstrat]

...eg bias supply cap

+1 on this:
an improperly filtered bias voltage can induce hum directly into the grids of the power tubes. Some forget to change this cap in an ordinary cap job.

[methodofcontrol]

yeah, that bias supply cap is on the chopping block. and i'll check the pre tubes for AC on the grids. will get back soon!

[pdf64]

'i'll check the pre tubes for AC on the grids'
I doubt that will help - the voltages are unlikely to be significant enough to identify over the background, and the meter leads on a high impedance circuit will induce or receive all sorts of noise.
The key thing is to locate where the problem is occuring then work out the root cause.

[Tom Phillips]

...one approach would be to remove all pre-amp tubes and check for hum. If so then investigate around the power tubes, eg bias supply cap, reservoir cap.
If not then fit phase splitter and continue the process. Pete.

Along those same lines, tell the group if/how the hum level changes with different settings of the volume and tone controls. This troubleshooting technique helps isolate the cause of the problem.
Tom

[methodofcontrol]

i replaced that bias cap w/ no appreciable difference in AC hum. i did notice that when i turn the bias down (read: negative voltage up) on the power tubes, the hum definitely went away. weird thing, is it went away more so when i adjusted one power tube vs the other. does that point to something weird in the PI?
i'm about to throw my hands up on this one and let someone else handle it. very frustrating.

[Tom Phillips]

...when i turn the bias down (read: negative voltage up) on the power tubes, the hum definitely went away. weird thing, is it went away more so when i adjusted one power tube vs the other. does that point to something weird in the PI?...

That sounds exactly like what one would expect to happen when the bias balance was adjusted. When you say that "adjusted one power tube vs the other" does that indicate that the amp has a separate bias control for each power tube? Or does it have separate bias adjust and balance controls? If so maybe nothing is wrong at all. Maybe the amp just needed the bias optimized for the installed power tubes. Can you measure and post the actual values of the plate voltage and the plate current you set when you adjusted the bias?

[methodofcontrol]

the plate is about 450vdc. each power tube has it's own 10k pot w/ the wiper tied to a 100k resistor to pin 6. then there is a 10k resistor tying pin 6 to pin 5. this is the same on each EL34. so with both tubes adjusted for -36vdc the amp hums excessively. i'm not citing this as CAUSATION but CORRELATION. if i adjust the power tube on my left to have more negative voltage... the hum diminishes considerably. whereas if i only adjust the power tube on my right and leave the left at -36vdc, there's not a lot of difference.
does that make sense?

[pdf64]

Are you setting the bias purely by the negative grid voltage, rather than plate dissipation?
In case you aren't aware of the preferred method/rationale, see linked document.
Idle Current Biasing - Why 70 percent?
Have your el34 tubes recently been confirmed as 'matched'?
See relevent schematic
http://www.webphix.com/schematic%20h...city50plus.pdf

[Tom Phillips]

Thanks for posting the Schematic pdf64.

Methodofcontrol,
Bias is set to achieve a target power dissipation value. This is done by adjusting the bias voltage which affects the plate current that flows. The product of the plate current and the plate to cathode voltage = plate dissipation. This is explained in detail at the web link posted by pdf64. The -36V shown on your schematic is only a guideline. It was probably the average or safe setting when the amp was designed and for the tubes that were then available. The fact that you have a separate bias control for each tube shows that the designer allowed for different settings for each tube. The reason is to allow the bias current to be balanced even when a mis-matched set of power tubes is installed. Therefore, it seems that all could be good with your amp if the bias is adjusted correctly. You would end up with equal current flowing in each tube but with unequal grid voltage. Some tubes could be so mis-matched that they could not be balanced with the current set up. I can’t tell if your tubes are that bad or not.

In order to properly adjust the bias you need to measure the plate current at idle. (zero signal condition). The most common way is to measure the voltage across 1Ω or 10Ω power tube cathode resistors and then calculate the current. Does anyone know why the power tube cathode resistors shown in the Sound City 50 Plus are one 68Ω and one 22Ω? Is that really what is installed? (If it were my amp I’d change them each to 1Ω or 10Ω bias sense resistors). With 450V on the plates I’d set the idle current within the range of ~28ma to 45ma. I know that’s a wide range but final setting depends on the tone characteristics that the user desires.

Regards,
Tom