insulating jacks and pots from the chassis

[zhyla]

I've read in various places that jacks and/or pots should be insulated from the chassis for some reason. I think it had something to do with ground loops. Opinions? I'm not entirely sure how I would do this if it's necessary.

Thanks.

[Enzo]

Can't recall ever seeing a pot needing to be isolated, but maybe. Jacks all the time, but it is not a universal. It all depends upon your grounding scheme.

To insulate a jack with a metal bushing, you would make the panel hole oversize and then slip a "T-washer" or "shoulder washer" over the jack bushing before installation. On the outside add an insulating washer under the nut.

http://www.switchcraft.com/products/..._schematic.pdf

Figure 3 is the T washer and figure 2 the flat insulating washer.

The T washer keeps the jack unsulated from the chassis and centered in the hole, and with the dlat washer on teh other side, the jack is mounted securely but not touching the panel.


But if I really need to insulate my jacks, I just use ine that is insulated to start with. For example the Cliff style jacks that Marshall uses - the ones with the large plastic nut.

[mooreamps]

I've read in various places that jacks and/or pots should be insulated from the chassis for some reason. I think it had something to do with ground loops. Opinions? I'm not entirely sure how I would do this if it's necessary.

Thanks.

Input jacks : yes
Pots : no
Output jacks : yes


-g

[Enzo]

I say again, it all depends upon your grounding scheme. Plenty of amps where the preamp ground bonds to the chassis at the input jack, so that would be a place NOT to insulate the jack.

[mooreamps]

I've read in various places that jacks and/or pots should be insulated from the chassis for some reason. I think it had something to do with ground loops. Opinions? I'm not entirely sure how I would do this if it's necessary.

Thanks.

It's neccesary if you want to biuld a quiet amplifier.

-g

[MWJB]

"It's neccesary if you want to biuld a quiet amplifier." No it's not. There's not just one grounding scheme that works and if you want to go that route, then you may find a scheme that works. But there are plenty of nice, quiet running amps that have inputs, pots & outputs all grounded at the chassis.

[Shea]

It never improved anything when I tried replacing an input jack that shorted to the chassis with an insulated that didn't. It didn't reduce noise. But noise wasn't excessive in that amp anyway. The one time I can say that it made a difference to use insulated jacks was on a Twin Reverb, when I replaced the RCA jacks for the reverb tank, and maybe the footswitch too (not sure about that), with isolated ones. I ran one or two leads from the isolated jacks to appropriate ground points on the circuit card, and that elminated an annoying hum that I was getting in the reverb circuit.

I've read that replacing the insulated jacks in Marshalls with switchcraft-style jacks that ground to the chassis is one possible fix for Marshalls that pick up radio signals. I might have read it on the old Ampage.

There are some techs who say the ground for the preamp power nodes should connect to chassis at the input jack or as close to the input jack as possible, just as adamantly as other techs say that you should use an isolated input jack. It's supposed to reduce RF noise, I think. If the preamp's grounding point is right next to the input jacks, then there can't be any harm in using a non-insulated jack because the path from the jack to the grounding point will be very short.

FWIW, I never had good luck when I tried to follow all the star grounding rules. I'd get weird interations between the controls, or hum, or odd behavior by the LFO circuit if there was tremolo in the amp. When that would happen, the fix would always turn out to be going old school - soldering ground lugs to the backs of the pots, grounding cathode pins on the power tube to the socket's mounting ring, stuff like that. And I never found noise to be a problem when I did that.

Usually what works for me is kind of a hybrid ground scheme where I have a ground buss running down the ground side of the circuit board. Sometimes I'd put a break in the buss between the preamp and the phase-inverter / driver circuit, in which case I'd connect the phase-inverter / driver buss to the chassis near the power tubes and connect the preamp buss to the chassis near the input jacks -- or sometimes connect it directly to a jack, but I'd use a star washer and make sure to tighten the jack hard so it won't wiggle loose. For pots, I'd run leads from the ground lugs to the buss. First, I'd connect those leads to the buss at the same point as a cathode resistor from the same preamp stage. If that causes any weirdness, then I'd move that ground lead to the point on the buss closest to the pot, so the lead would be as short as possible. And if THAT causes weirdness, then I'd solder the lug to the back of the pot.

Shea

[Enzo]

AS I already said, and as MWJB also points out, you cannot look at the jack out of a circuit context. That is why you get these stupid arguments where one guy says always ground it and another guy says never do. It is not a jack alone doing the job. There is a plan to any proper grounding scheme, and it can take more than one form. A grounding scheme where the jack is to be grounded is not likely to be improved, and may well be degraded, by lifting that ground. On the other hand, an insulated jack plan may well be compromised when you add the unplanned for ground.

In high RF environments, if radio stations are coming in, one possible cause is a lack of a ground path for the RF. Grounding the normally ungrounded jack will perforce kill the RF most likely, but may degrade hum performance. The real way to do it would be to add a small cap from the jack ground to the chassis. SOmething like a 10pf disc right at the jack.

[Shea]

AS I already said, and as MWJB also points out, you cannot look at the jack out of a circuit context. That is why you get these stupid arguments where one guy says always ground it and another guy says never do. It is not a jack alone doing the job.

I hope I didn't sound like one of those guys. I'm just an amateur repeating some stuff I've read (usually during the course of one of these arguments) and saying what has worked in my limited experience.

In high RF environments, if radio stations are coming in, one possible cause is a lack of a ground path for the RF. Grounding the normally ungrounded jack will perforce kill the RF most likely, but may degrade hum performance. The real way to do it would be to add a small cap from the jack ground to the chassis. SOmething like a 10pf disc right at the jack.

Ooh, thanks for the suggestion - I should try that. It's even simpler than replacing the jack, and it would preserve the cosmetic appearance of the amp.

Shea

[rathmann]

To add a tiny bit of additional info - part of the issue pertains to whether you are using the chassis for shielding, as a ground bus, or both.

If the overall grounding scheme is isolated from the chassis it's probably a good idea to earth the chassis and go ahead ground the input to it.

When using the chassis as a ground bus more attention needs to be paid to where the power supply nodes are. Typically you want to keep big current bus returns far away from signal returns. If the layout dictates that the input is very close to the power supply (and an isolated "star ground" isn't utilized) lifting the input from the chassis could help to avoid hum (ground loops).

I've done a few scratch builds and even one channel switcher with a high gain channel. Good layout and a sensible "hybrid" star ground approach has resulted in never having to troubleshoot hum/grounding problems even though they all have grounded inputs. Certainly plenty of other troubleshooting challenges though!

[Enzo]

No SHea, I was not labelling you, you were reporting that attitude in others.

And thanks rathmann, perfect example of what I meant about context.