Phase Inverter Hum in 6G3 Build

[pjhalliday]

I have a 6G3-based amplifier that I recently built that functions well, but has a pretty nasty 120 Hz hum that seems to be localized to the phase inverter section. Simply pulling tubes, neither the first nor second preamp tubes have an effect on the hum, but pulling the PI tube eliminates the hum and the output section is quite quiet. The hum is independent of gain in the two channels and I have not really found anything to help by moving wires. Things that I am aware of that differ from the original 6G3 are that inputs 1 & 2 do not share a cathode bypass but are set up with different cap values, the heater wiring is elevated to 40 V DC (after starting with a ground reference--no difference in hum), and the filter caps are a JJ 40-20-20-20 can rather than single electrolytics. I have also gone through and jumpered a 20 uF cap across each of the filter sections to make sure that there was not a bad section and this was also unsuccessful in changing the hum. Resistances and voltages while running read within spitting distance of the Fender layout diagram values. If anyone has a suggestion for where to look next I would greatly appreciate the help. Thanks a lot.

[Jana]

Change your speaker jacks to the isolated type and run a ground wire from the speaker jacks to the same ground point of the 1500 ohm resistor on the PI. Your hum will go away.

[pjhalliday]

I will give that a shot (hopefully tonight) and report back.

[pjhalliday]

Unfortunately that did not make a difference.

[Jana]

What's your grounding scheme?

Post gut shots. Clear ones.

It really does sound like a grounding issue.

[gingertube]

http://www.turretboards.com/layouts_sch ... xe_6g3.gif

It looks like the phase splitter is driven from a split load resistor on the final tube stage 100K between output and anode and only 15K up to B+.

The input impedance to the phase splitter is that 1M resistor bootstrapped up to about 2M since about 1/2 of the input signal also appers at the other end of that resistor.

2 things leap out at me from that:
1st that is fairly high impedance - it will be susceptible to hum pickup, Is that feed very long, does it pass close by anything like a tranny etc., You may need to use shielded cable for that audio feed.

2nd there is very little isolation from any residual power supply ripple on the B+. make sure that the bypass cap fro that supply node is good. I would prefer to see just thae 100K up to B+ and teh signal take from teh anode 100K juction. If the signal is too high then use an interstage divider. This gives 3 benefits, first there is less residual B+ nois appering at teh anode as there is then a 100K up to B+ and 62K looking down into the anode of teh tube divide. This lower noise is then divided again by the interstage attenuator. The bottom resitor of teh interstage attenuator then acts to shunt teh 2M input impedance of the PI thus making it less susceptible to noise pickup.

Cheers,
Ian

[pjhalliday]

Sorry it took so long to get back. Here are pictures of the interior.

Gingertube, I don't think that I'm quite understanding your suggestion--just me being dense.

The grounding scheme is grounds associated with V1 & V2 are tied tgether including from volum/tone/tremolo controls. The V3 ground has the speaker output jack tied to it. Output tube grounds are tied together. All of these groups go to one of the PT bolts and tie to the plug ground.

[pjhalliday]

For what it's worth, disconnecting the wire from pin 1 of the PI tube to the 82k resistor eliminates a significant amount of hum.