50Hz hum

[txstrat]

I'd have the bias tap return (the yellow wire in the diagram which I'm assuming is one of the two wires from the bias winding) go to the bias circuit board where the 100uF/100V + and 10k are.

That's where I had it in the first place, but changed to the way the other amp is wired (the one which is extremely quiet). Gonna put it back.

For all your grounding hints, would it be undue to ask for a sketch? I think I know what you mean but having a circuit described is different from looking at it. I could send (or post) a layout with no wiring. That way you could just draw the wires into it.

[dai h.]

how about if you draw (or just edit the "layout grounding") diagram and I'll tell you if it's what I described? (just easier for my zero schematic drawing skills)

[txstrat]

Thanks for your time. Here you go.
I've added numbers and letters for further reference.

[dai h.]

okay, here's my attempt at a crappy reworking of your diagram. Hopefully it's more clear.

[txstrat]

Simulpost. Look at #33

[dai h.]

lol. Yes. It's good that I put that up. Is it more clear now (what I'm saying)?

[defaced]

Also, (Mike) even if the pots are connected together (such as on a typical older style grounding scheme on Fenders, Marshalls, etc. which use the chassis) I don't think it's a loop. I see it more as one connection (the potentials are brought closer together by connecting the backs of the pots). I think it also makes sense in case a pot becomes loose (what would happen if there wasn't a wire connecting the backs?).

That's because of my background. I build high gain amps, so I'm a little neurotic about grounding. There's a really good ground scheme sketch (that's not mine) that is referenced alot on a different forum I frequent. I asked the owner for permission to repost it here.

[txstrat]

Is it more clear now (what I'm saying)?

Yes, thanks alot. I'll see what happens if I put the preamp back in circuit. Hopefully there's no additional noise (although I don't expect it to get noisier - since the hum I was tinkering about - with help of you all - was the only noise, up to now).
I'm going onto vacation in the next days and don't know if I can show up here on the MEF. But I'll let you know about the result as soon as I can.

[txstrat]

I asked the owner for permission to repost it here.

Wow, makes me curious.

[dai h.]

That's because of my background. I build high gain amps, so I'm a little neurotic about grounding. There's a really good ground scheme sketch (that's not mine) that is referenced alot on a different forum I frequent. I asked the owner for permission to repost it here.

I favor the more modern way (everything off of chassis except for one point), but since that older style can work I just had to try to think of a sensible explanation. I've actually seen that sort of thing in a Tascam 488mkII(old cassette multitrack I mess around with sometimes) mixer section. Each channel ground line is lengthwise from top to bottom, and the controls along that length (gain, cue, high, low, pan, etc.) have a sort of "boxed C" shape cover which a) is (obviously) a cover for the potentiometers, and b) since it's made of (tin plated brass?) and is connected one side towards the top and other side towards the bottom along the ground trace running in the same top to bottom direction (I think) helps to lower the ground impedance and "bring the potentials closer together" as with the pot covers, metal front of chassis plus tinned copper wire connecting the backs of the pots.

The ground scheme sounds like a good idea to share. Ideally, there should be an attempt to understand since the non-understanding is the source of the puzzlement.

[defaced]

This is courtesy of Joey over at the SLO Clone forum. It's not called out in the sketch, but pot casings are not used as ground tie points, all grounds go directly from the pot lug to the ground point of the stage (normally the cathode resistor ground point). If the pot casings are isolated from the chassis, a separate bus wire can be added and tied to the chassis for shield purposes. I have used this ground scheme and it works quite well.



As a general comment, this is how I think of grounding. Grounding should mirror the connections of the power supply/signal routs. If the PS/signal wires were drawn in the sketch above, they would follow the ground wires almost identically. The screens and plates would be the only divergent point because they share a ground point (the power tube cathodes). To choose a more stable grounding point, the screen PS cap is used as the cathode ground location.

[dai h.]

oh I've seen that one. He seems like a very smart guy. One issue I have is that is the input jack as the chassis connection. In principle, it's fine, but input jacks do sometimes come loose (i.e. doesn't seem reliable enough for the circuit ground to chassis ground to safety ground connection). I think it would be safer to make a redundant one to the chassis very close by (very close by = not much difference), say the side panel (I've seen that for instance in one of Paul Ruby's builds).

As a general comment, this is how I think of grounding. Grounding should mirror the connections of the power supply/signal routs. If the PS/signal wires were drawn in the sketch above, they would follow the ground wires almost identically. The screens and plates would be the only divergent point because they share a ground point (the power tube cathodes). To choose a more stable grounding point, the screen PS cap is used as the cathode ground location.

Totally makes sense to me. But sometimes though you might have to deal with something (a vintage amp, clone, DIY build, whatever) that doesn't have the more rationalized, modern (or however you want to call it) grounding scheme, so there might be a need to "make the best of a situation" instead of just re-doing everything with a super rationalized approach, so I think it's good to be able to try to figure out the least that can be done to make things more-or-less viable. These might be things such as making a "sort of star"(like with the power tube cathodes in joey's diagram), or being able to have a filter cap ground not immediately close to the stage it's decoupling, using the chassis for a part or portions of the ground line, etc.

[defaced]

I agree about the jack statement. I prefer to use isolation washers and make a dedicated wire chassis connection.

[dai h.]

oops. Missed a couple of things in the diagram (brain kept thinking ground line wasn't connected to the back of the pots since that's what the diagram appeared to show). Here is an updated diagram:

Also, the state of some of the other grounds (isolated or not?, etc.) are not very clear to me because the pictures are rather blurry (and all ground connections are *not* drawn in in the diagram).

[Joey Voltage]

oh I've seen that one. He seems like a very smart guy. One issue I have is that is the input jack as the chassis connection. In principle, it's fine, but input jacks do sometimes come loose (i.e. doesn't seem reliable enough for the circuit ground to chassis ground to safety ground connection). I think it would be safer to make a redundant one to the chassis very close by (very close by = not much difference), say the side panel (I've seen that for instance in one of Paul Ruby's builds).

Actually Dai, thats the Idea! the non-insulated Input jack accomplishes two things, it's big, and makes a large contact point, and is as close as you are going to get as a direct contact point to the chassis from the sleeve lug to zap all the crap riding on the cable to earth. I never addressed this on the scheme, (but I have in other forums) I do both, I use both the non insulated jack as the main connection, and I usually use, a safety net connection right directly underneath with as little lead length as possible in case the jack comes loose or corrodes.

Totally makes sense to me. But sometimes though you might have to deal with something (a vintage amp, clone, DIY build, whatever) that doesn't have the more rationalized, modern (or however you want to call it) grounding scheme, so there might be a need to "make the best of a situation" instead of just re-doing everything with a super rationalized approach, so I think it's good to be able to try to figure out the least that can be done to make things more-or-less viable. These might be things such as making a "sort of star"(like with the power tube cathodes in joey's diagram), or being able to have a filter cap ground not immediately close to the stage it's decoupling, using the chassis for a part or portions of the ground line, etc.

To be honest with you every old amp, and even new ones I work on (you would be surprised at how little grounding practices have changed!), I just put in the extra work...... You have to deal with the multican scenario often, but alot of those amps are lower in gain levels so some routing is not as critical.