AC 30 build bizarre hum

[EG78]

Hello, long time away, new handle, same problems it seems. ;-) I just completed an AC30/Rocket build, not my first by far, an odd hum like something too close to a PS or plate lead, controlled by the treble pot. With treble all the way down, big hum, all the way up, gone. Grounding the middle lug (out) increases the hum. It only stops when I ground the downstream side of the input cap to the PI, junction of the cap and 1M resistor. I have a 220k series resistor before the .022 input cap, grounding at the junction between the two increases the hum, treble pot has no effect. Grounding the input to the treble pot (50pf cap) increases the hum, more so with the treble fully up. I have pulled the preamp tubes including the CF 12AX7 when doing this. The PI ground is at the ground lug of the 32/32 radial cap that feeds the Pi and CF. Preamp grounds are to the chassis. The plate leads and B+ to the OPT are under the board, fixed to the bottom of the chassis. Raising the board doesn't mitigate the hum. Piggybacking a cap onto the first supply cap reduces the hum somewhat but not a game changer. Any help would be very much appreciated.

[pdf64]

Have you checked that the weird hum isn't oscillating?

[EG78]

Unfortunately I don't have a scope to check but I don't think it is. The layout and components are almost identical to one I just built 6 mos. ago, no problems there. If it is oscillating, what is the diagnostic procedure scopeless?

[Mick Bailey]

When you kill the hum by grounding the PI input cap/1M resistor are you grounding it to the tone stack ground, or a different place?

What happens if you lift the ground to the tone stack - does the hum stop?

I'm thinking you could have a ground fault.

[Chuck H]

We will need a full layout for your wiring scheme and grounding to assess this. If you've had this problem before (as indicated?) then it might help to recognize what you're doing that continues to fail and stop doing that. My guess is that you have the tone stack grounded in two different places or dually grounded causing a loop. A picture of what you have or an ACURATE drawing of your layout with a clear reference to the grounding scheme are needed at this time.

[EG78]

I always try and not connect my ground wires from each stage until I fire the amp up and ground them in different configs to find the quietest way. I missed the hum because the treble pot was probably half up where there is the least hum.
With PI to PS 32/32 cap ground (CF and PI), treble down-most hum. Mid pot and tone stack chassis ground with preamp.
PI and tone stack to cap ground, treble down-least hum, up-most, 2/3-none.
PI, mid pot, tone stack to cap ground, treble up-most hum.
Pushing the raised board down nearer the B+ and plate wires underneath increases hum when minimum, no difference with hum highest. It doesn't seem quite the same hum, not as intense, separate issue solved with distance and/or shielding.
Please remember only the PI tube is installed when this goes on.
This stuff drives me nuts.

Chuck H, Preamp, PS cap and input jacks grounded to chassis lug. CF and PI cap (32/32) grounded to main cap (50/50), in turn to AC ground, PT centre tap. Heater centre tap on power amp cathode.

[EG78]

Extreme L green wire-jacks to board, 2nd-preamp to lug, 3rd-tone stack, 4th-mid pot, 5th-PI.

[Mick Bailey]

"With treble all the way down, big hum, all the way up, gone."
"I missed the hum because the treble pot was probably half up where there is the least hum."

Just to be clear, which is it? A pot that nulls out noise mid-point usually has the same noise but out of phase on each end of the track so it cancels out in the middle.

[Jon Snell]

I would return all separate grounds to the single ground point, circled . That will illiminate 100/120HZ smoothing hum.
Don't forget to apply some multicore fluxed solder and a hot iron, to your solder tag to ensure a good connection. (Don't use Lead Free solder, unless a professional is applying it).
A different technique is required.

[Mick Bailey]

The main filter caps should each have their own ground lead in preference to being daisy-chained. Even that short length of connecting lead can give rise to a ground loop. A photo showing the pots would be useful.

[Chuck H]

Thank you for posting a picture, but it with little exception it doesn't demonstrate the ground scheme. Which is what we need for diagnosis. Mick is right about the PS rail filter grounding as well as his inquiry about the treble pot position.

Where are you hiding your cathode bypass caps. So much of your wiring is under the board that trying to pidgeon through the layout is complicated and incomplete. It looks like you may have some daisy chained grounds and shared ground nodes that aren't ideal as well as grounding at the pot cases or a pot case buss (which we can't see?). The possibility of ground loops is there and could relative to pot adjustment in given circuits. I'll grant that a more ideal ground scheme can look untidy in a build but the way I see it once the amp is in the cabinet and being used function is more important than aesthetics.

[Chuck H]

I would return all separate grounds to the single ground point, circled . That will illiminate 100/120HZ smoothing hum.
Don't forget to apply some multicore fluxed solder and a hot iron, to your solder tag to ensure a good connection. (Don't use Lead Free solder, unless a professional is applying it).
A different technique is required.

While I agree that taking all grounds to a single point would likely be better than what is there now I will politely disagree about locating a star ground scheme with the PS at that end of the chassis with the safety ground. The safety ground should be independent and the signal input end of the chassis is a better location for a star point. Though I do usually ground PS main filters at the PS end of the chassis. I know your suggestion is relative to what's been put on the table to work with and I mean no disrespect.

[EG78]

Thank you gentlemen, lots to digest. Just to reiterate, the green ground wires in the picture were unconnected and I put them in different combinations to illustrate how the hum changes from one side of the treble pot to the other (Post 6).
Cathode bypass caps are not installed until I start "tuning" the amp.
No buss wire on the pots.
The last build essentially the same, issue free. I will try and implement what has been suggested.
Unfortunately my upload limit has been reached.

[Jon Snell]

While I agree that taking all grounds to a single point would likely be better than what is there now I will politely disagree about locating a star ground scheme with the PS at that end of the chassis with the safety ground. The safety ground should be independent and the signal input end of the chassis is a better location for a star point. Though I do usually ground PS main filters at the PS end of the chassis. I know your suggestion is relative to what's been put on the table to work with and I mean no disrespect.

Absolutely, I didn't notice the safety earth connector was there as they are normaly crimp connectors for safety reasons.
As long as the ground point is a single point, with the safety earth on its own lug, all should be fine.
Thank you Chuck H for your insight.

[EG78]

Should I separate the "safety earth" ground (AC ground?) from the first PS cap? Separate lugs?
Progress: I un-daisy chained the filter caps. I ran the PT CT and power tube cathodes to PS 1st 50/50 cap, thence to AC ground on chassis lug. 2nd (preamp) 32/32 cap to chassis lug near preamp side of chassis, all grounds from the board to there, EF86, 12AX7, CF 12AX7, PI 12AX7, mid pot ground. Quietest amp I've ever built gentlemen. Many thanks to your generosity in sharing your knowledge, very much appreciated!