Trouble-shooting hum in a Fender 65 Twin Reverb

[Billy R aka DynaFreak]

Assuming you have normal 120-125 VAC line voltage, 426 volts on the plates is kinda low. I work on Lots of these Fenders. That makes me think something is pulling the voltage down and that might have something to do with your buzz/hum. I would take a good look at the power supply. You can start by just measuring the VAC on the plate supply filter cap with the amp at idle. More than 2 VAC is usually indicative of a problem. I would do the same for the screen voltage supply. VAC should be way low on that, in the order of 100-200mA

[g1]

I was thinking the same thing about the B+. My gut feeling is it's still biased pretty hot. Don't get hung up on the 70% thing. There are many who suggest 60% max. or even 50% when plate voltage is in the 420 or more range.


The schematic shows 440 plate volts, you might try aiming for that with your bias adjustment.
And on the bias topic, it was stated earlier that the power tubes were biased at 38watts. If that is the case and it was like that for long, I would be concerned about the kind of shape those tubes are in.
Matching can have a big impact on hum, but your numbers sound fairly close. In the silverface twins with the "output matching" rather than bias controls, the setting of the pot had a big impact on the hum level.
Also, there are notes in the schematic about the settings when adjusting the hum pot. (open plug in J3 etc) If you didn't use those settings, give it another try.

[AMPREPAIR]

“DynaFreak” makes me think of the great Dynaco products.
“g-one” make me think of aviation. You must be a pilot.

I set the voltage supply to 120 VAC with my trusty Variac when making bias adjustments/measurements. On a separate note, I will check the voltage at pin 3 with output tubes removed (to see if they are excessively loading the P/S). I didn’t think 426 VDC would be considered low when compared to 440 VDC. But I trust you guys and this variance may be considered large for those with good valve experience.

I have often wondered how much AC is acceptable in a guitar amp power supply. Would it be safe to use a “rule of thumb” anything over 0.5% could be considered to much AC for the plate supply and .05% for the screen supply?

I agree, biasing output tubes at 60% of their max value may be a good idea when dealing with plate voltages over 420 VDC.

Because these tubes were previously biased to 38 Watts each (well above the maximum dissipation) , they may have been compromised and will be replaced.

Excellent catch! An open plug at J3 is used when adjusting the hum pot. I need to go back and verify the pot is adjusted correctly.

[g1]

“g-one” make me think of aviation. You must be a pilot.

Nope. g1 is that part of a tube commonly used as the input. I'm fairly surprised no one else here uses tube nomenclature for their user names. One of the best laughs I had was when someone on this forum called me g-spot. Not sure if they were joking or preoccupied.
I think ripple percentage will be much higher than that, maybe 5 to 10 percent? Maybe others here will have more exact numbers for max ripple at idle and at full power.

[Enzo]

I got no ripple rules, but certainly a lot higher than 1/2 a percent.

In a push pull output stage, B+ ripple tends to cancel, so the initial B+ node, which usually only goes to power tube plates, can be kinda sloppy. 450VDC? In percents, maybe a percent or two, so 5-10VAC maybe. Unless I have cause, I don;t usually measure it. By the screen node, I pretty much expect clean DC. Almost no ripple on the screens. And certainly none at all on the small tube nodes. The "filter caps" on the nodes after the first two are really there more as decouplers than filters. I reserve the term "filter" for ripple smoothing duty.

[AMPREPAIR]

Mystery solved!
C35 (a filter capacitor in the negative 65VDC supply) had one leg that was not soldered. Once the capacitor was properly soldered, the amp stopped “buzzing”. Thanks to everyone for their helpful comments!

[Jazz P Bass]

Now why didn't I think to ask that?
Glad you got it.
That's what counts.

[g1]

All this talk of bias and ripple and I couldn't put the two together and ask "did you check the bias supply for ripple?" Doh!
Thanks for the update.

[Jazz P Bass]

Must of been napping in that class
Rule #1: Check the power supply.