5E5-A Fender Pro 120 Cycle Hum

[Fletcher Munson]

Just finished a fresh build on a 5E5A tweed Pro and need some help tracking down a 120hz hum. The amp is built to vintage Fender specs/layout with a couple tweaks: (1) Added adjustable bias. (2) Used a ground buss bar. (3) Added virtual center tap on the heater supply. I can post a copy of my layout drawing if needed.

I determined that it's 120 cycle hum by using a signal generator to inject 60hz and 120hz signals, then matched the signal to the hum by ear. I'm pretty darn sure it's 120 cycle hum.

When removing power (either the power switch or pulling the plug) the hum stops instantly; it doesn't die away slowly.

I pulled preamp tubes one by one. Removing V1 has little effect. Removing V2 reduces the hum somewhat, but it's still there. Removing V3 stops the hum completely.

I've tried two different, brand new, 12AX7 tubes in V3 with no change. Also tried replacing V1 and V2.

My measured voltages match the Fender schematic pretty closely. About 3-5% higher, as expected, but I don't see any anomalies.

Here are the plate voltages from the first three tubes, both VDC and (looking for ripple) VAC:

V1/pin 1 - 143 VDC - .006 VAC
V1/pin 6 - 138 VDC - .01 VAC

V2/pin 1 - 132 VDC - .07 VAC
V2/pin 6 - 268 VDC - .003 VAC

V3/pin 1 - 212 VDC - .51 VAC
V3/pin 6 - 262 VCD - .49 VAC

I also measured VAC on the power rail, right on the positive side of the 3rd (16uf) filter capacitor, and at the junction of the 56k and 100k plate resistors for V3. All I see is .005 VAC. But when I measure on the other side of those plate resistors, I see .51 VAC on pin 1 and .49 VAC on pin 6.

So it looks to me like there's a problem around the phase inverter, but I'm not sure how to find it. My understanding is that 120hz hum has to come from the power supply. But what are the causes? If it were a bad filter cap, shouldn't I see some AC on the 3rd filter cap? What else should I be looking for?

[loudthud]

Probably a grounding issue. Post the layout.

[Fletcher Munson]

Here's the layout. All jacks are insulated from the chassis with shoulder washers.

5E5A Layout.pdf

[loudthud]

Why is the virtual ground for the heater supply hooked into the bias supply? Try just grounding the CT. Other than that, the ground looks OK.

[Leo_Gnardo]

Why is the virtual ground for the heater supply hooked into the bias supply? Try just grounding the CT. Other than that, the ground looks OK.

I had to look, then had a WTF moment: the virtual CT is attached to an AC reference, there's your hum source. Yes try grounding the CT, and you can also try grounding the virtual center tap instead, whichever gives you least hum is the winner.

Just buttoned up an old Ampeg B15N that had its humdinger pot wiper wired to the bias filter negative pole - theoretically not the thing to do - but it's just about dead quiet as far as hum so I'm leaving that alone. When I saw loudthud's comment I thought this might be the same, but no.

[Fletcher Munson]

That was it. Don't know what I was thinking putting the virtual CT there. I moved it to the ground buss and the amp is dead quiet now. Many thanks.

[g1]

Has that file been edited so we can no longer see the error?
Please post the original file so the thread makes sense to anyone with a similar problem in the future.

[Leo_Gnardo]

Has that file been edited so we can no longer see the error?
Please post the original file so the thread makes sense to anyone with a similar problem in the future.

Just had another look, looks like no edit, the junction of 2 100R resistor pair off the filaments connected to the junction of 6K8 resistor and bias rectifier, on the AC side. I'm glad to hear the fix worked well as it should.

All it takes is one wire in the wrong place...

[g1]

looks like no edit, the junction of 2 100R resistor pair off the filaments connected to the junction of 6K8 resistor and bias rectifier, on the AC side.

Thanks Leo, I see it coming off the pilot light now. Apologies to Fletcher Munson.
So the problem was raw AC going there. I was confused and thinking loudthud meant the DC bias voltage, which can be done, though uncommon.