Best EMI shielding solution for injection molded enclosure

I have a requirement for an EMI shield (to particularly attenuate the

300 mhz range) inside of an injection molded enclosure (resin being PC). I have looked at metal "boxes" to place all of the electronics inside of(which includes single board computer, card reader, etc) as well as some new GEP resins (EMI-X in particular), and then finally have tried a foil baked inside of the part which I believe is a CuMg mix. All of these have their ads and disads but I would prefer to go with the baked coating. I have heard that this has its share of QC problems (uniform thickness is hard to maintain) as well as some disposal issues if/when the part is scrapped? Can anyone help me with some ideas or past experience?
Reply to
Drew
Loading thread data ...

I wish I could remember the name of a company that dealt in resins that incorporate metal fibers. The data sheet is probably in one of three big boxes of electronics catalogs that I have archived in the attic from way back when I was doing PWB layout. Sorry, but looking through those boxes is more than I can devote time to right now, but if you can't find an equivalent by doing a Net search I'll dig the boxes out and look thru. The result of using the resin is a part that looks a bit like it's made out of Brillo pad material if you melt the plastic away. Of course, it's terribly hard on the mold. If this isn't a high volume application it might be practical, but you'd still have to pay for a high volume mold just to be able to get a medium volume from it. Perhaps needless to say, aluminum molds won't likely work well even for small volumes.

Mark 'Sporky' Staplet>

Reply to
Sporkman

Sporky, I have worked with some folks over at LNP Engineered Plastics which is a subsidiary to GEP. They can bond stainless, carbon fiber, and a few other materials into some base resins. It is basically like making a cake. You start with a polypro/PC/PBT base, add in the metal that you desire based on what needs to be attenuated, and then add in some fillers to provide other charatceristics (strength, chemical resistance, etc)...the only drawback I had was that the expected attenuation wasn't exactly what I needed. So now i exploring other possibilities. In either case, I highly recommend the LNP technology and others like it...it is just that for the frequency range that my ASIC is chattering at, I am needing something more?

Reply to
Drew

Have you considered sprayed conductive EMI coatings for the interior of the parts?

Art W.

...snip

Reply to
Art Woodbury

I hear ya. Well, you know you can have adhesive-backed foil (aluminum or copper) stamped out (custom) in such a configuration as it will fold up nicely into an enclosure. Even pretty complex enclosures can be handled that way, although of course the more complex the enclosure the more difficult assembly becomes. If this isn't a low-end (read "cheap") consumer product then the assembly time can be worthwhile. If it IS a low-end consumer product and if you DO come up with a viable solution, please post it here for us all to see. I've got some stuff in my head that could also use a little attenuation (not intended to be a double entendre, but it's probably appropriate anyway).

Maybe you should be posting in "sci.electronics.design" and/or "sci.electronics.equipment", etc..

Good luck, 'Sporky'

Drew wrote:

Reply to
Sporkman

I've been down that road many times, and I don't recommend it.

Conductive coatings all have the same intractable core problem as metal- loaded resins: You just cannot establish and maintain a continuous electrical contact to and among the actual metallic components.

Effective shielding requires that all faces of a box be in contact with each other along every edge, and that all metallic components of the shield be in electrical contact with each other. Almost by definition, metallic particles in paint can't possibly be in contact with each other, because each is surrounded by a layer of the resin that carries them to their eventual destination and bonds them together mechanically. Fiber loaded plastic has a similar problem; the fibers are in contact only in a statistical sense, and are not bonded in any way that could resist corrosion or mechanical strain.

The typical solution that people who haven't blown all their capital on magic bullets use is a plastic box with a thin, tin or nickel plated sheet steel liner. The liner is typically equipped with lanced and/or formed tabs along every edge that establish contact with adjacent faces, at intervals governed by the wavelengths to be shielded. At today's frequencies, the typical gap between such teeth, tabs, whatever, is typically not more than ten millimeters. For an example, just open the case of a recent Mac or PC. Open an older, slower one, and the tabs will be farther apart. Pretty soon, you'll have to solder all the edges.

Ever try to solder paint? Ever try to establish a continuous, reliable electrical contact to a painted surface? Not possible; trust me.

Another point: The 'conductive coatings' salesmen will _never_ mention the price of the stuff, until they've got you completely committed to using it. It costs more than twice as much as the paint that the salesmen use on their yachts.

-Mike-

Reply to
Mike Halloran

There IS such a thing as sputtering pure metal as a coating. No resins. It ain't easy, and QC is a bitch.

'Sporky'

Mike Halloran wrote:

Reply to
Sporkman

Mike, I appreciate yours and all the other feedback. I can attest to your experience as well and agree that a conventional "metal box" is the only statistically significant method of providing a true earth (or data/DC in this case) ground. I enjoy trying these new technologies that exist though and as I said in the previous email, the steel-based resin works at attenuating and shielding but it does not meet my immediate needs and does create havoc on the molds (the molds are steel but the gates and shut-offs were designed for a PBT. Well, this all helps me make some decisions. Thanks

PS...I have resisted metal boxes for EMI shields in this application due to the assembly issues, impact/vibration resistance and then of course weight - especially in this particular application these all make a difference!

Tajmccall

Reply to
Drew

I mentioned sprayed EMI coatings. 'cos you can buy small spray cans to evaluate.

You might have a look at electroless nickel coatings as well.

Art W

Reply to
Art Woodbury

PolyTech Forum website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.