Potting materails

[kb1gtt]

I've been looking for some potting materials. I think it's a good idea to have a water resistance option. I think a low melting point thermoplastic potting / encapsulant would be good, AKA hot glue. However generic hot glue has unknown dielectric properties, unknown thermal expansion properties, and unknown long term shrinkage properties. I found this one from 3M which claims it's good for electrical product potting http://www.digikey.com/product-detail/en/3797TC/3M6414-ND/1306091

I just asked 3M about it's properties as I can't seem to find them in the datasheets they have online. Perhaps that's what I think we should use, but I suspect they don't offer that kind of data. We'll see how that goes.

I also found this one http://www.hot-melt-glue.com/industry_list02.asp?iid=34 which notes some of it's properties, although I can't seem to find a supplier and can't seem to figure out how much it would cost. As well I can't find it's thermal expansion properties or it's shrinkage properties.

I like the low temperature thermoplastic potting materials, as you can re-heat them and re-gain access to the PCB if you need to make a repair or something. I also like the clear stuff as could allow you to see if something has failed, or do some basic optical inspection.

Does anyone know of such a material? I fear the best I'll find is that clear silicon stuff, which is a pain to remove.

[kb1gtt]

This 3M 3779 shows some promise. I wonder how well it reacts to humidity absorption. It probably rejects the water quite well, as it's listed to be good for potting. It can be obtained from here http://www.digikey.com/product-detail/en/3779-Q/3M6421-ND/1306098

This notes most of the critical specs like dielectric strength (650V/mil), thermal conductivity (.114 @ 45ºC btu-ft./sq.ft.-hr.ºF) and thermal expansion (506 x 10-6 {-30ºC to 130ºC} in./in./ºC).
http://media.digikey.com/Resources/3M/3m-epoxy-hot-melt-adhesives.pdf

From the below its applied at 350–385°F (177–196°C) with a Heat Resistance of 300°F (149°C)
http://multimedia.3m.com/mws/media/372489O/adhesives-and-tapes-design-guide.pdf

The design specs noted below indicate the PCB is typically only rated for 105C and the components junctions are typically rated for 150C or less. So if the potting materiel ever melts, it's because the design specs were not upheld. As well the transition from solid to liquid at 150C offers a bunch of protection for the chips, as you'll hit a thermal wall at 149C until you complete the transition from solid to liquid, then you'll start to rise again after that. I'm liking this potting materiel. Only down side I see with it, is that it's not really clear, it's a kind of translucent amber.
http://rusefi.com/wiki/index.php?title=Manual:Hardware

[kb1gtt]

This 3779 is Polyamide and according to the below link it's young's modules is 2.5 (10^9 N/m2, GPa).

http://www.engineeringtoolbox.com/young-modulus-d_417.html

According to the below, FR4 has a young's modules of 3.5×106 psi (24 GPa) or 3.0×106 psi (21 GPa) depending on the glass fiber grain direction.

https://en.wikipedia.org/wiki/FR-4

If we assume that the two have interfaced and have not sheared, they will have the same physical dimensions. As well they will bond around 160C-ish, and for nice round numbers lets say we aim for -40C as the cold temperature, so a delta of 200C. In the attached spread sheet, I calculate that on a .3 meter length of a PCB, the difference in expansion rates would be 0.02952M, which would turn into a stress of 2.0664 GPa for the FR4 and 0.246GPa for the 3779. This is good as it means the FR4 stresses and potting materiel stresses would be in the elastic region of their physical properties. What I don't know is if that kind of shear would de-laminate the glue from the PCB materiel. It seems figuring out the shear stress is far harder to do.

[kb1gtt]

One fellow e-mailed me a comment about using Young's modulus that way I did. He suggested it was not good, and expressed concern that the tensile strength and compressive strengths should be used. The FR4 would be under compressive loads and it's compression strength is 415 MPa. While the 3779 would be under tensile loads. According to the below that's a number between 99 and 264 MPa.

https://en.wikipedia.org/wiki/Polyamide-imide

So this potting materiel may not survive such large temperature swings on a 0.3M length. Or the more likely is that my math's is mangled. In my prior post I predicted a stress of over 2.0664 GPa for FR4 while it can probably only sustain 415 MPa. As well the Polyamide which is likely the same properties for the 3779, is also outside of it's tensile specs. I wish I could find a good white paper or similar that would better explain the typical stresses seen in potting. Below are some misc links about tensile and compressive strengths.

https://en.wikipedia.org/wiki/Ultimate_tensile_strength
https://en.wikipedia.org/wiki/Compressive_strength

[kb1gtt]

3M 3748 has much better expansion rates, as well it has a much more detailed datasheet, attached to this post. However not translucent.

Can be obtained from here http://www.digikey.com/product-search/en?keywords=3M6415-ND