Nickel plating of aluminum for good EMC still ok?

We yellow-alodine every aluminum chassis we've ever made in the last 36 years. I've got a prototype audio selector panel from 1973 in the "museum" box that looks pretty much like it did when I first put it together. It spent its first few years in San Diego a few miles from the coast and the rest in an unheated drafty barn.

I've got some prototype stuff flying around in the 182 that I put together somewhere around '77 that has been sitting in an unheated drafty hangar for twenty years and it, too, is still in excellent shape.

That chromate Alodine dip is amazing. BTW, you can get small quantities (gallons) of both Alumiprep (the acid etch first dip) and Alodine in most well-stocked professional automotive stores. You can also get it from Chief Aircraft up in Grant's Pass Oregon. Chief also has Boeing's T9 corrosion protectant that it uses on its own aircraft, but it doesn't look anywhere nearly as nice as Alodine.

My tinbender sends my stuff to a plating shop in Fresno for Alodine; he says they do the best work he's ever seen. If you like I'll get their name.

Jim

performance

aluminum.

process.

You can nickel plate aluminum - it looks *very* nice - but it can peel off if stressed, and nickel on steel is cheaper.

You can anodize aluminum and then push in stainless PEMs and screw the board to them.

John

performance

aluminum.

any

or

very

process.

For hardcore situations that is often not good enough unless you place a screw every 3/4" or so. My case is roughly 100 watts worth of combined switchers with very sensitive analog stuff moving in about two inches down the road.

Will the PEMs actually cut the anodize layer or do you have to grind (or mask) the anodize layer first? I'd worry a bit in a salt environment about the effects of stainless on aluminum; PEM makes some very nice aluminum (CLA) fasteners. WIth those and a nylon standoff and nylon hardware, I'd not worry about corrosion for a long, long time.

Jim

The pems that bite well, like the ones with exterior sawtooth knurls, seem to ground fine. They dig 5 mils or so into the hole wall, and the anodize is half a mil if that.

John

performance

aluminum.

any

or

very

process.

I don't remember ever using aluminum, though I haven't been involved in too many mechanical assemblies. Everthing mainframe was steel and built like a tank.

Are the PEMs then conductive to the aluminum? Does the connection remain good (corrosion?)?

:Hello Folks, : :Got to thoroughly connect the perimeter of a circuit board with an :aluminum panel, one screw every inch and so on, for good EMI performance :mostly of the internal kind. In the good old days that was simple, :nickel plating of the PCB perimeter and nickel plating of the aluminum. :It even looked really posh. The FR-4 is a no-brainer but are there any :road blocks with aluminum nickel plating such as stricter EPA rules or :other new regs? : :Reason I ask is that I had run into this in the past, where suddenly a :process became hard or prohibitively expensive to contract out inside :the US because of some new rule.

I suppose the board is already finalized and can't be changed to include custom emi/rfi shielding cans? This method would make the connection to the aluminum casing quite simple using the normal metal stand-off pillars and braided straps as required. eg.

A good alternative to beryllium-copper is tellurium-copper, which has largely replaced beryllium-copper in spot welding electrodes where there tends to be some particulate shedding. Properties are almost identical, but it might not be available in EMI gasket finger strips (which I have only seen on panels or doors intended to be opened frequently, so that cheaper solutions like plating and bolting are inconvienient).

Custom = expensive, plus too much heat. There will be provisions for a local shield can though, an off-the-shelf version. But no matter what, the board must be thoroughly connected to the aluminum. Thoroughly in terms of great RF performance in the connection. Instead of the can we could, of course, also use metal strips. However, then the problem is the same: How to plate them?

Tellurium is pretty evil stuff too, though.

Cheers

Phil Hobbs

the=20

=20

disease=20

areas=20

=20

hands.

Went looking, the MSDS does not look that nasty to me. Gallium looks worse. And about 0.5 % in a Cu alloy scares me a lot less than any Beryllium Copper alloys.

You got me curious, so I compared the MSDS for Tellurium with the MSDS for Lead. Both rated health risk 1 but the permissible exposure level for Tellurium is twice that for Lead. If not for all those "not available" entries on the Tellurium MSDS, one might conclude that it is about half as dangerous as Lead, and that it's use requires similar precautions. Like lead, evil when ingested, but useful enough to outweigh the risk in some applications.

Funny thing about tellurium, it's more acute, particularly the part about the insane garlic breath.

The -ides are particularly bad with acid (e.g., copper telluride, which should be present in tellurium bronze). H2Se and H2Te are even more posionous than H2S.

Tim

As far as galvanic corrosion goes, aluminum and gold don't mix:

Active SS 430 looks like a "maybe" compatibility-wise w/ aluminum 6061, but I don't get why it appears twice, widely separated, in lines 28 _and_ 41...that's a bit mysterious.

Al oxidizes super quickly, which is a pain.

Nickel plating the Al would prevent oxidation (and it looks great) but it peels off, IM limited E. I don't know if that's inherent or just bad workmanship, but that's what I've seen.

Al and Sn are sympatico--that's handy. Maybe anodize the Al, but mask off and tin-plate the part that needs to contact the PCB? (You can solder Al, so maybe you can tin plate it.)

This Al-solder patent Al might qualify for "the least RoHS-friendly substance on the planet" award:

Must watch out for corrosion failure at the solder/Al interface. Special solder alloys fix that.

Hmmm. For best corrosion performance it looks like you should make your fasteners from indium, tin or lead. :-)

YMMV.

Cheers, James Arthur

Yep :-(

Actually the nickel plating we employed on my very first ultrasound design held up very well. This was a surface that pressed against large Cannon connector that was mated and unmated quite frequently by (non-technical) customers. By gynecologists, for example. I never saw any peeling or wear and our sale guys really loved the shiny "mag wheel" looks.

Well, looks like the good old nickel on aluminum plus nickel (no gold) on the circuit board is the ticket again here.

Sounds reasonable. Nickel being the most cathodic on that list, its surface area should be minimized.

Throwing tin on top of both surfaces might be better yet corrosion-wise. Then your exposed galvanic cell is Al-Sn everywhere, the most benign natural combo possible, AFAICT.

That would be easy to test--tin up some Ni-plated Al and Ni-plated Cu & dunk 'em--plus a few other misc. experiments--in saltwater. If they're awful you'd know in a week.

Or you could post on a metallurgy forum.

I think I'll let the client do that since they must deal with some other parts and surfaces as well. Would be good to have one common process. I am not too familiar with tin yet, always used nickel or nickel-gold.

Oh, about tin plate- beware of whiskers! I've seen a few fairly old items literally crawling with whiskers, they look like extremely fine carpet, rough to the touch. TO-3's, various hardware. I noticed some parts inside my Tektronix 475's attenuators are tin plated -- 30 years later, they're going intermittent, in need of cleaning (which I just did last week).

The only tin plating that's any good is when it's married with lead. ;-)

Tim

Thanks for the hint, Tim. I certainly will stay away from tin then since this gear can reach such lifetimes.