Ideal Grounding...Question for MerlinB

[Wilder Amplification]

Hey Merlin...I read your article on grounding. Great article btw.

Just had a few questions if you don't mind...I know some on here are thinking "It doesn't matter in guitar amps", "You're overthinking it", etc etc...and yes I've built several amps using "not-so-textbook-ideal" ground schemes that were dead quiet, but I'm doing this "textbook ideal" grounding scheme on my next build just for the "fun" of it (if anyone can nail "fun" down to an absolute please let me know ) so...here goes -

1) You stated that ideally in a star grounding scheme that there should be only two connections to the chassis itself -

a) Mains safety ground

b) Input jack ground

You also stated that you want these two connections to be as close together as possible. Now since most amps have the input located clear at the other end of the chassis opposite the mains/power end, this chassis ground wire that runs from the input jack to the mains safety ground area of the chassis would have to run the length of the chassis. My questions are -

Would running a length of wire that long create any noise issues?

Or is it a non-issue as this wire would have no current flowing through it and as such that wire only provides a chassis reference?

2) On a typical Marshall style preamp that uses dual can filter caps with a common negative...you have two stages on one positive terminal of the filter cap while you have two other stages on the other. I know that the common negative design of these caps isn't really in favor of "textbook ideal" but for the sake of conversation -

For the most ideal grounding using these caps, would you just run 3 wires off the cap's negative terminal -

Wire 1 to the Rk/Ck of one pair

Wire 2 to the Rk/Ck of the other two stages

Wire being the return back to the supply?

3) Is the whole idea behind proper textbook grounding to "theoretically" seperate the audio current from the power supply current and not have them flowing through the same wires?

Would greatly appreciate your feedback on this...thanks for your time.

[Merlinb]

Or is it a non-issue as this wire would have no current flowing through it and as such that wire only provides a chassis reference?

Yes, you shouldn't get problems since no current flows in it, except when a mains ground loop has been created, and even then the noise voltage does not appear between grid and cathode so should no be amplified. That's the whole point of grounding the input jack rather than any point further down the line. In practice you can just ground the input jack right to the chassis without a long wire all the way to the earthbond- it's highly unlikely that any problems would arise.

For the most ideal grounding using these caps, would you just run 3 wires off the cap's negative terminal -
Wire 1 to the Rk/Ck of one pair
Wire 2 to the Rk/Ck of the other two stages
Wire 3 being the return back to the supply?

Yes, you'd be best to use the negative terminal as a sort of mini 'global star'.

3) Is the whole idea behind proper textbook grounding to "theoretically" seperate the audio current from the power supply current and not have them flowing through the same wires?

Yes exactly, and also to avoid heavy PSU ground currents from running in the same wires as 'light' PSU ground currents. The ground scheme should ideally be a mirror image of the way the positive side of the power supply is arranged / fed.

[Joey Voltage]

On a similar note Jon, I would avoid crappy multi cans like the plague. There is only one place I would ever consider using one and thats for the reservoir cap, in parallel or series parallel,

[Wilder Amplification]

On a similar note Jon, I would avoid crappy multi cans like the plague. There is only one place I would ever consider using one and thats for the reservoir cap, in parallel or series parallel,

And the beauty of that is that I'd only have to get rid of the preamp filter cap on my purple plexi since the rest of them are all paralleled caps.

Thanks much Merlin. Much appreciated!

[Joey Voltage]

And the beauty of that is that I'd only have to get rid of the preamp filter cap on my purple plexi since the rest of them are all paralleled caps.

Thanks much Merlin. Much appreciated!

yeah get rid of one, and add a few more for the preamp for what you have been trying to do. same ol' crap I have been advocating to you for the last 8 months! LOL

[Wilder Amplification]

yeah get rid of one, and add a few more for the preamp for what you have been trying to do.

Well actually I was just thinking about getting rid of the preamp filter cap altogether. You mean I shouldn't do that?

[cminor9]

On a similar note Jon, I would avoid crappy multi cans like the plague. There is only one place I would ever consider using one and thats for the reservoir cap, in parallel or series parallel,

just wondering why the hatred for can caps? I don't have feelings on them either way, but I have used them before and was planning on using one on an upcoming build. On the spitfire I built, the can seemed to work well.

[Wilder Amplification]

just wondering why the hatred for can caps? I don't have feelings on them either way, but I have used them before and was planning on using one on an upcoming build. On the spitfire I built, the can seemed to work well.

I don't see a problem with them either...in practice an amp that uses them can be just as dead quiet as an amp that doesn't. They just don't allow you to follow the exact "textbook/on paper" grounding scheme since the negative terminal is common to both internal caps and as such doesn't allow you to separate the ground/grid leak circuits of the stages that are powered from that cap.

[Joey Voltage]

just wondering why the hatred for can caps? I don't have feelings on them either way, but I have used them before and was planning on using one on an upcoming build. On the spitfire I built, the can seemed to work well.

They don't allow you to control ground currents as Ideally as you can with single units. in most builds people also have a tendency to fit them where ever they have room for them, which often puts them further from the stages they are decoupling, which can also lead to problems. sometimes it's not an issue with the typical dual cans in a preamp because these grounds currents are pretty low anyway, but why take the risk if you can avoid altogether?

[Joey Voltage]

I don't see a problem with them either...in practice an amp that uses them can be just as dead quiet as an amp that doesn't. They just don't allow you to follow the exact "textbook/on paper" grounding scheme since the negative terminal is common to both internal caps and as such doesn't allow you to separate the ground/grid leak circuits of the stages that are powered from that cap.

So what? LOL people use and advocate chassis ground schemes too, and get lucky and don't have issues, even heard people advocate all sorts of (to me) dumb and silly things just to be period correct. but again why take the risk when you don't have to? how about the time you don't get lucky, and you have to chase that demon through the amp, sounds like a pain to me, and I have been there. It's one of those things where it becomes a principal of: why not just do it as ideally as possible, If you know it's a bigger issue you wont have to worry about later. Just seems like an easy decision to make considering other aspects that can go wrong in any given build.

[Wilder Amplification]

why not just do it as ideally as possible,

Because you're already doing that...wouldn't wanna plagiarize your practices. Just tryin' to do my own thing and be different bro.

[clyde1]

I use the F&T cans in all my builds, different models and schematics, no problems at all. I'd be a little leery of the JJ 40-20-20-20 unless I were paralleling the different sections.

[soundmasterg]

Once I use up the can caps that I have, I won't be buying any more unless I need them for a repair job or if I am space limited in a build. The only way they are ok with an optimal ground scheme is if they are wired in parallel, and they are expensive when compared to discrete caps. The custom ones that Weber does can be made to have seperate grounds, and I did use some of these in a custom application where I was space limited.

On my recent build, I used a ground buss mounted on the preamp board, with every stage decoupled and the caps locally located. Each stage is grounded to the buss in order, then at the end of the preamp board, a wire goes to the chassis ground, which is located down by the power end of the amp. The power board has another ground buss and everything grounds in order and is decoupled, then grounds to this same point on the chassis with a seperate wire. The safety ground has it's own ground point that is about 4 inches away from the signal ground. I used two boards instead of one because of the output transformer's Z mount btw. I did use DC filament on the EF86 input tube on the high gain channel and AC filaments for the rest of the tubes, and they are not elevated either. I couldn't elevate due to my switching method. This amp is extremely quiet as far as it's noise floor, so I guess its ok sometimes to ground somewhere other than the input jack?

greg

[Wilder Amplification]

The only other question I have is...

Will there be a difference in tone depending on if I use cloth covered solid wire vs using 10 guage rebar for my ground buss?

[Steve Conner]

Yeah, rebar is made of steel, so it'll give more of a steely tone. Ideal for lap steel players.

I would tend to agree with the issue about multi-section can caps. The problem with them, as was mentioned, is that all the sections share a negative terminal. And you don't want your first filter cap sharing a negative terminal with anything else, because of the high pulsating current that flows in it.

You can still have a good grounding system if you bear this in mind, and use at least two multi-section cans (a noisy one and a quiet one) or one can plus a few individual caps.

I used the F&T two-section can caps on my Crown restoration project, but I connected the sections in parallel and used them as single capacitors, and used other small electrolytics to filter the earlier stages. I chose the F&Ts purely because they fitted the holes in the chassis and came in a 500V rating.

This was my quietest amp yet, the hum and noise level is 80dB below rated output, and that's mostly hum pickup in the input and output transformers. It'll go down to 90dB with the volume control turned down, which takes the input transformers out of the circuit.

I never thought of grounding in terms of mirroring the signal path, as Merlin does, but I suppose it makes sense.