Electrically Grounding Powder-Coated Parts

[EngineerErrant]

I'm trying to find an elegant way of reliably grounding a sheet metal housing that's been powder-coated, without masking off the hole where the grounding lug attaches. If I do it that way, I'll have to make the part out of aluminum instead of steel to prevent corrosion. That's doable, but I feel as though there must be some standard hardware that does this.

I've been thinking of something like a pronged washer under a nut (e.g. square-neck bolt washers) to bite into the material, but they seem relatively uncommon, and only really exist in larger sizes. Kepps nuts don't have enough bite to reliably get through the powdercoat.

Any ideas?

"Engineers like to solve problems. If there are no problems handily available, they will create their own problems." -Scott Adams

 

[MikeHalloran]

Use a Nelson stud, welded before powdercoating.
Mask it during powdercoating.

Or use a PEM stud, pressed in after powdercoating.
( Proper hole prep involves a milling machine to cut away the powdercoat. )

PEM also makes blind studs.
They install from/on the hidden side, and have good retention, but they may not fit your application. Check the catalog.




Mike Halloran
Pembroke Pines, FL, USA

 

[MintJulep]

Braze or solder on a copper pad, then mask that prior to powder coat.

 

[TheTick]

Attach ground wire using thread-cutting screw.

 

[CorBlimeyLimey]

I've used both weld studs (suitably masked) and self-tapping screws with great success.

Magnetic pads on steel parts have also worked well, provided the connection is subsequently sealed (painted)

 

[JohnRBaker]

The actual reason for grounding the sheet metal housing may help to determine the most appropriate and thus economical solution. For example, if it's being grounded simply to provide RFI shielding, than the self-threading screw will probably do the trick. However, if the ground is to close the loop on an electrical circuit then more variables will probably need to be considered, such as the voltage across the circuit as well as the potential current that may have to pass to the 'grounding' point, before you can decide on how one configures it to meet these requirements.

John R. Baker, P.E.
Product 'Evangelist'
Product Engineering Software
Siemens PLM Software Inc.
Industry Sector
Cypress, CA
Siemens PLM:
UG/NX Museum:

To an Engineer, the glass is twice as big as it needs to be.

 

[MikeHalloran]

There are basically two reasons to ground a removable sheet metal panel:
1. For RFI/EMI control.
2. Because UL (or similar) requires it.

The ground stud and braid strap that is sufficient for electrical safety as in (2) is not sufficient for (1) if the frequencies of interest include MHz and above. For MHz suppression, the ground connection has to be essentially continuous around any and all seams.

This appears to be an instance of (2).




Mike Halloran
Pembroke Pines, FL, USA

 

[EngineerErrant]

As suggested, it's for UL compliance. (As such, I don't think that thread-cutting screws are allowed on an electrical housing.) Using a weld/PEM stud might work, although the cover is bent into a relatively small "C" shape, which might make actually attaching the thing difficult once bent. I'll have to check if it's too close to the nearest bend that it can't be attached before bending.

Thanks for the suggestions, guys. I'm all ears if there are any more ideas out there.

"Engineers like to solve problems. If there are no problems handily available, they will create their own problems." -Scott Adams

 

[HDS]

You can insert the PEM before coating and just mask it. Use a stainless PEM and corrosion will not be an issue.

I also do not believe the UL or CE allow self taping fasteners for the main ground.

Some people use star washers but they would still create a corrosion problem IF they actually cut deep enough through the powder coat but not ground.

 

[DRWeig]

Which UL standard applies to your product? Conduit locknuts are OK in powder-coated enclosures for UL508A. It's really performance they're after, not so much the method (in my experience). The test involves 20 or more amps (AC or DC) through the connection, measure the voltage drop, and compute resistance. If it's less than 0.0005Ω, you're good. You could try your Kepps nut and test it.

Best to you,

Goober Dave

Haven't see the forum policies? Do so now: Forum Policies

 

[EngineerErrant]

It's a light fixture; UL 1598.

"Engineers like to solve problems. If there are no problems handily available, they will create their own problems." -Scott Adams