Input shield connected to Plate??

[bnwitt]

I'm wiring up this new turret board in the JMP50 and I've noticed that the shield on the existing input wires is connected to the V1 a&B plates and on the V2a plate instead of to ground.  I've never seen a shielded input wire's shield connected to high voltage like that.  Obviously it worked in the amp with the old pcb but it just seems like a safety issue to me.  Maybe I should just redo the jacks and put the 68k resistors on them (abandoning the board turrets I have for them) then run a shielded wire with it's ground on the pot bus wire ground assembly like a Hoffman build.  Anyone have any thoughts?

[phsyconoodler]

It's the best place to connect the shield for maximum shielding.There are quite a few amps that have done that.It requires very careful manipulation of the shield wire so there are no stray strands poking out anywhere.
  But for the most part there is no need to do that unless you don't want to drill more holes in the chassis to mount ground lugs.
It might help on some amps and others it makes no difference.

[HotBluePlates]

I don't agree with that.

A hot shield was mentioned by Gerald Weber in his second book. I don't know that it was done a whole lot prior to that, because all the vintage equipment I've seen runs the shield to ground. I'd bet money it was someone's bright idea after reading about it.

The idea is that the plate is 180 degrees out of phase with the grid, so if a noise signal was induced on both the shield and the center conductor, they'd cancel at the tube. You might get some local feedback at a very high frequency to cancel trash. It's dangerous if not executed perfectly, and not a good plan even then. It is a good way to burn up parts or shock yourself.

The big objection I have is few people who would try it would bother to double-check that the insulation in the cable can even withstand the d.c. present. And shields are better run short and straight to the a.c. ground of that stage they feed (or from the a.c. ground of the source).

[Fresh_Start]

+1 on safety concerns regardless of possible noise benefits.

There's a PPIMV circuit which ties the shields to B- instead of a DC ground.  I assume it works fine since B- is an A.C. ground but I'd prefer not to surprise some future tech with any D.C. voltage on a shield.

I was under the impression that you generally connect the shield to the closest ground to the "quiet" or source end of the lead.  IOW ground volume pot to grid shield at the volume pot, or input jack shield at input jack.

Is it important the the shield be tied to a signal ground at all?  Can't you just ground to the chassis (directly or indirectly)?

Related Question:  do you all think it's "bad practice" to use a shield as a ground conduit?  I was able to quiet down a bit of noise on a Blackface amp by grounding the Reverb pot with the reverb return stage by using the shield.  However, some future tech might miss the connection on both ends and wonder why the Reverb pot didn't work at all...

Cheers,

Chip

[bnwitt]

Ok well a few different opinions here.  This setup sure seems original Marshall and not some tech alteration but who knows.  But, it scares me so I'm going to use the back of pot bus wire for the shield ground connection.  I just don't like the idea of a short between shield and conductor putting 200 volts on the input jack 68k resistors.

Attached are two photos.  One is a JMP50 I found online with no shields and just straight wire.  The other is the one I'm working on with shielded wire.  Maybe I'll try just straight wire and see how noisy it is.  I sure wish I had some Mullard mustards for this amp instead of Orange Drops.

[jjasilli]

Freshstart:  There's a PPIMV circuit which ties the shields to B- instead of a DC ground. Isn't B- ground? Or do you mean something lifted from ground, like a bias supply?
Is it important the the shield be tied to a signal ground at all?  Can't you just ground to the chassis (directly or indirectly)?
But the chassis is signal ground.  

The point seems to be that the plate of the 1st stage is reverse signal in relation to the signal at the input jack; and a reverse charge on the shielding braid purportedly reduces noise or hum.  Assuming this works well enough to bother, the problem is the inherent danger of the design.  Cutting the shield a tad short at the input jack side, and sealing any exposed braid with heat shrink might work. If only the AC and not the DC is needed on the braid, then maybe a cap between plate and braid would make this safer.  Pure speculation though.

[RicharD]

>I've noticed that the shield on the existing input wires is connected to the V1 a&B plates and on the V2a plate instead of to ground.

What a horribly unsafe idea, especially on the input lead of all places.  Seems like a potentially harmful "bandage" fix to either a flaw in the grounding scheme and/or a feedback/oscillation issue brought on by excessive gain and layout problem.  I'd change it too if I were working on it.

>do you all think it's "bad practice" to use a shield as a ground conduit?

In most cases no.  Often times it's the tidiest and quietest method especially with controls.  The one bad instance that comes to mind was something that Silvertone did in a split chassis amp.  IIRC, the harness had a 6 pin connector with filaments, B+, and some switch legs but NO GROUND.  The sole means of grounding for the top chassis was the RCA cable with the signal wire.  When I was servicing this amp, I had forgotten to plug the RCA back in.  I went to plug it back in with the amp turned on and just about got knocked out of my boots.  That amp got a chassis to chassis ground strap that did not cause a loop, in fact it got quieter.
 

[jojokeo]

I've noticed that the shield on the existing input wires is connected to the V1 a&B plates and on the V2a plate instead of to ground.
Wasn't this Weber's "tricks of the Ninja" for finding those nasty parasitics and/or oscillations that hide in those nooks and crannies like a theif in the night!?  I must admit that I have tried this but it made no difference from my standard shielding method. Connecting the shield to the chassis or any other ground should be fine when it's only connected at one end. But others think it's best to not use circuit/signal ground either way and I've read where Merlin likes to shield his circuit ground along w/ his signal wires.

bnwitt - nice board work, looks clean & neat.

[bnwitt]

bnwitt - nice board work, looks clean & neat.

Joe thanks, but neither one of those boards is mine.  They are both original Marshall boards.  My board looks like this

[plexi50]

Having a high voltage sheild scares me. If it works then col. But i would be warry of another deadly issue to deal with. It's dangerous enough as it is

[bnwitt]

I'm with you Plexi50.  I took it out.  In fact, I used no shielding on the input wires at all and this amp is very quiet.

[plexi50]

I suppose in theory it works but i pass

[jojokeo]

Joe thanks, but neither one of those boards is mine.  They are both original Marshall boards.  My board looks like this

Still looks good to me either way. I like different colored board material too. I wish it wasn't so expensive though.

[bnwitt]

I like different colored board material too. I wish it wasn't so expensive though.

I get my red GPO3 in 24"x36"x1/8" sheets from mcmaster for about $5 per square foot:

http://www.mcmaster.com/#electrical-grade-fiberglass-(gpo3)-sheet-stock/=7shsap