Best Grounding Scheme for a Princeton Reverb

[kresejg]

Hello all,

I know this is a subject run into the ground, (ha) but I haven't found an explanation as it pertains to my situation. I've built a Marshamps Princeton Reverb clone, and there is a manageable but definite 60Hz hum. (also the hum is significantly increased when the reverb is turned above 3 or 4, even with no guitar plugged in, but that's a discussion for later) I wired it up with a ground bus, and it sounds great except for that hum.

My question is: I'm using a cap can, and all of the filter caps are tied to one ground that comes out of the can; therefore, I can't separate out the grounds of the filter caps to different parts of the circuit as in the Hoffman ground buss method. Does anyone think that could be contributing to this? I've looked at star grounding, but it seems like it would be a lot of work to rip my bus out and rewire everything.

I've been playing with it for a while, and the way I have it grounded now is in the picture below. This gives me the least amount of hum for my bus configuration.

Any suggestions would be much appreciated. I usually try to get what I need from what is already posted, but I haven't found anything addressing this particular issue.

Also, I know ground hum can be attributed to any number of issues... I'm just wondering about this particular issue with the caps grounded together.

Thanks!

Joe

[Gtr_tech]

If the gnd bus is attached to the backs of the pots on the front panel, remove that. Just run gnds from each pot lug and leave the gnd bus wire hanging in the air. Make sure the plate and bias supply gnds go to the same place.....preferably the gnd at the first filter cap. For even quiter operation use isolated bushing input jacks......get *everything* off the chassis except the "star" gnd point.

I just finished up on an old 70 or 71 Superlead and the gnd scheme was terrible.....it hummed like a bastard. Correcting the grounding quieted it right down.

[kresejg]

So it sounds like you like the star method better. If I wire it up that way, do I also need to run the grounds from the output and reverb tranny back to the filter cap ground?

[Gtr_tech]

Usually you can get away with the output gnd on the chassis. Its a good idea to keep everything else off it except for a single gnd point. The rev drive xfmr can ge grounded locally (where the drive tube cathode grounds).

[Jazz P Bass]

Measure the voltages coming off that "cap can". Look for abnormal Vac ripple.
I have purchased new cans that had a bad filter in it.

[Mandopicker]

I would also suggest checking the lead dress, specifically of wires in the reverb circuit. I say this becasue you mention that the hum increases when turning up the reverb control. It would seem to me that the problem could be in that particular area. It is a very sensative part ot the amp.

P.S. Best grounding scheme for a Princeton Reverb...just look at a Black Face layout...Fender figured it out a long time ago.

Good luck.

[kresejg]

Thanks for the tips.

Jazz P--
I went back and scoped the filters caps, and it turns out that there is significant AC signal at the first cap, which connects directly to the rectifier. About 10Vp-p. Would you expect some ripple since it's connected directly to the rectifier? Also, it's a distorted sine wave, almost like a sawtooth. There was about a 1Vp-p AC signal on the second filter cap, and the others had no appreciable signal.

Also, after idling the amp for about an hour tonight, my PT was too hot to keep my fingers on it for more than 5 seconds. I've never had a PT get quite that hot. Is that normal for a Princeton?

Mandopicker--
I have the wires at the D,B, and X nodes kinda twisted together just to keep them out of the way. Would that cause any problems?

BTW, I had wanted to get a brass plate to follow the old school grounding, but I haven't been able to find one easily.

[Michael Allen]

Since this is a push pull output section, the 10VAC should be canceled out and wouldn't add much noise. That sawtooth you're seeing is correct, it's 120hz ripple. If you're using the 5U4 then you'll have 5VAC on that filter cap anyway so there will be some AC ripple there no matter.

You said initially it's 60hz hum. That comes from your heaters. Can you show how the heaters are wired, or take a picture? Do you have a center tap or artificial center tap? If yes, try elevating it to the powertube cathodes.

[kresejg]

Wel

Since this is a push pull output section, the 10VAC should be canceled out and wouldn't add much noise. That sawtooth you're seeing is correct, it's 120hz ripple. If you're using the 5U4 then you'll have 5VAC on that filter cap anyway so there will be some AC ripple there no matter.

You said initially it's 60hz hum. That comes from your heaters. Can you show how the heaters are wired, or take a picture? Do you have a center tap or artificial center tap? If yes, try elevating it to the powertube cathodes.

Michael,

The third pic is my wired up chasis. I don't think the hum must be 60 Hz, because I think my heaters are wire up pretty well. (tightly wound and well above the other circuitry) Additionally, I do not have a center tap, and I do have the 100 ohm resistors wired up.

I scoped the output again, and it turns out that the wave I was looking at was actually around 18-20kHz, not 60 Hz. There is a little bit of 'buzz' associated with the hum when I listen, so I think that's what I was originally looking at. (this is in the first pic)

When I zoom out on the scope on the output in the 50-100Hz range, the signal is too noisy to really tell what the period is. (last pic)

I looked at the signal at the first filter cap again, and it really looks like it is 60Hz, not 120Hz. (second pic) The period in that pick is around 17-18msec. The tube I am using is a GZ34S.

All this was taken without anything plugged into the input jack, so it's definitely coming from the amp.

The brown wire with the rubber shrink wrap on it is there because the Mojo schematic and original Fender schematic hook up the driver tube to different filter caps, so I tried both of them and needed a little longer wire to do it. (I thought it might have something to do with Mojo using a 40mF cap on the signal from the rectifier)

I've been putty off rewiring all the grounding to see if someone comes up with something else first. It seems to me that this should work, because a lot of people use the ground bus method.

Thanks again.

[Michael Allen]

try moving the PT secondary center tap to the filter cap negative, keeping the wire from the cap to the star ground. This allows the CT to be directly connected to the filter, and then run to everything else. Do a search for RG Keen grounding schemes, he has a couple illustrations of how to separate different grounds. It should work using the grounding bus if you get the filter cap and center taps as well as OT ground in the right places. Try moving the OT ground to the star as well, if you can isolate the jacks from the chassis.