can't quite cure a V1a ground loop

[Luthierwnc]

Hello All,

Troublesome 102 hybrid build nearing completion. The schematic and a closeup of the layout are included with this post. Since this picture I've shielded the small wires going from the bright switch to the relay.

I've thrown a few extra gadgets into the amp and I've got a buzz. I've isolated it (somewhat) to the tone-stack caps. Touching a grounded .1 cap to the treble, mid or bass cap does knock the buzz down noticeably. Touching it to the V1b .047 cap kills it completely.

What's different than the stock build is that there is a series cap circuit that either runs at .047uf on the Mid or as .014uf. Those both have leads that run to a mini switch on the panel. The treble cap goes to a mid-boost relay on the relay board. Both of these need extra wire.

All of the ground sources for these circuits are at the same point -- close to the bolt holding the bus wire -- a couple inches from the input jack. The jack is a Switchcraft and it is not isolated. I did detach it from the chassis and ground it on the bus and it didn't make any difference. Disabling the relay circuit didn't make any difference either.

The grounds for the main board are shown with green circles. The grounds that go to either the panel switches or the relays have green squares. Both wires are grounded on the bus at the same solder joint. They are wired based on the 102 ground schematic here at TAG. The red oval is the ground for the loop circuit which is near the power tubes.

I've tried three different 12AX7's (all ANOS) with the same result.

Tonight I disconnected all of the V1a cap wires and shorted the treble cap to the PI. It worked fine and sounded clean -- except for a little hash from using a non-shielded jumper. The filaments are behaving. I also jumped it to the input of the loop and it was fine as well. That's as sure as I am that the issue is between the V1a plate and the Volume input.

So now I guess my question is if there is a better way to run the wires from the caps to the pots, relays or panel switches to keep them from picking up trash? Or, is there something about the ground arrangement that seems out of sorts?

I'll fiddle with it until I get it right but if anyone has an "A ha" moment, I'm all ears.

Thanks, to all, Skip

[bcd123]

Hi there,

First thing to check is are you sure its a ground loop? If you have a access to a DC supply, disconnect the heater wires to V1 A & B and feed it off 6.3VDC - should pull about 300mA. If your hum is gone its not a ground loop its the heater getting into V1A so then you have a few options. If your hum reduces but doesn't completely go away try feeding more heaters off DC. I've recently had this problem with a Fender 75 that I converted into a Two rock hybrid. Possible remedies below;-

1) better shield your heater wires by running them close to the chassis or wind them together more tightly to achieve more self cancellation (common mode rejection).

2) use a trick I've seen on Mesa Boogie stuff where they rectify and filter the heater supply to the first valve only.

3) get a different mains transformer with a tap that can be rectified to supply DC to preamp tubes. I have a US Voltage transformer out of a Rivera Bonehead 100w which would do this which you can have for the cost of freight..........unfortunately I live in Australia



Best of Luck,
B

[67plexi]

Your heater wires look real low. it may be AC induction to DC


Steve.

[erwin_ve]

Your input grid resistor looks like a wirewound?
If so induction from closeby heaterwire is very likely, like 67plexi suggested.

[rogb]

Your heater wires look real low. it may be AC induction to DC


Steve.

This ^^^^^ I thought it was a trick of the light at first, they need to sit high over the sockets.

[Colossal]

Your input grid resistor looks like a wirewound?
If so induction from closeby heaterwire is very likely, like 67plexi suggested.

That grid resistor is a Kiwame 2W carbon film. But agree about possible induction from proximity. Also that orange drop cap just down the line, to the right of V3, along the axis of the heater winding. If that cap is physically close to the heater winding as it looks in the photo, move it.

[Structo]

If this amp has the Dumbleator built in, it could be in that.

When I first built my stand alone D'lator, I had a bad hum, it was a ground loop associated with the shielded cable.

I remember clipping a ground on one of the cables, but right now I can't remember which one it was.

[Luthierwnc]

Good eye Colossal. The carbon films are Kiwame and Takman. Takman metal films except for RN plates and a few CF65s because they were in the drawer. Altogether a festive color palate.

I took a selfie of the heater wires. They are a little squashed around from working on it but they aren't as low as they look in the overhead shot in the first post. I may redo them anyway because they are the worse for wear after two builds.

More importantly, when I jump from the .002 treble cap to either the input of the loop or the PI, I get no buzzing other than some hash from using an unshielded jumper wire. I haven't done the mid or bass caps but it seems like a lead-dress issue coming off the tone caps to their respective connections. It also suggests to me (which is by no means gospel) that the ground location for the V1a cathode is correct.

Tom; all of the loop grounds on the board go to the power tube cathode lug on one of the socket bolts. The the two 33uf filter caps go there too along with B+2 and B+3. 4 and 5 go to the bus. The loop recovery cathode is the red oval next to the V2 cathode cap/resistor networks in the first post shot.

I think I'll get a chance to put it back together tonight with more attention to keeping the wires better isolated. Will report back when I can air it out.

Thanks all, sh

PS Welcome to TAG, bcd123

[MKB]

It might be worth to try pulling the high impedance (grid) wires away from the chassis and see if that makes a difference. Marshalls are notorious for induced hum to wires near the chassis.

Also double check the coax shield connections, it may be possible that one of those isn't making good connection to either the circuit ground point or the cable shield.

Finally, verify the back of the volume pot is well grounded.

[Luthierwnc]

It still has a little noise but moving the .02 mid cap from the board to the switch knocked it down quite a bit. I also redid the heater wires so they are 8mm higher than they were. The attached shot is straight on so you can't see the difference but it is there. 18AWG on the power tubes. 20AWG on the 12AX7's.

I still have to to some fine tuning. The PAB is a bit hot. That is controlled by the 10-turn trimmer on the relay board. I'm also getting more volts on V1 than V2 so I'm playing with the rail resistance. The amp sounds really good. With the cab shield and away from all the basement fluorescents I'll put it through its paces tomorrow.

Thanks to all, sh