Recommendation for very thin dialectric coating on metal

[ZeroCubic]

Hi guys,

I need recommendations for thin insulating/dialectric coatings I can put onto metal with reasonably cheap equipment. I'm willing to spend a few hundred on the solution, but beyond that, its easier to just reorder the magnets with a coating from the factory (which is what I would have done originally if it was required at the time)

A small assembly I'm working on needs to have two permanent magnets glued together but be electrically isolated. The magnets are very small, less than 2mm wide at the gluing point and the thickness of the glue interface must be sub 100 microns. So the dialectric coating has to be similarly thin. The voltages being dealt with are from 10 to 100v. I'm using epoxy to bond the magnets, which is an insulator but not a reliable one at these thicknesses.

The requirements:
- coating thickness ideally less than 10 micron, no more than 100 micron
- voltages in the range of 10v to 100v
- coating resistance minimum >10kOhm, ideally >1MOhm
- coating should have reasonable mechanical stiffness
- chemical compatibility with common epoxies and CA glue.
- the whole magnet can be coated if needed (this is ideal actually)
- scalable: I have about a hundred of these assemblies to make so the solution has to be quick at scale

My ideas so far:
- Liquid electrical tape: appears to be too viscous for my use and not stiff enough when set
- parylene: as far as I'm aware, is essentially impossible to apply without specialised equipment
- Thin CA glue: Ive tried this and it works ok-ish for one at a time. but isnt really scalable to have a bucket of CA glue and requires several coats
- thinned out lacquer: I'm leaning toward this, perhaps I can just dip all the magnets in a bucket and dry on a fine wire rack, rinse and repeat a few times?

Any ideas?

[Ian.M]

Why not just mix 80 micron solid glass beads in the epoxy to ensure a guaranteed minimum bond line thickness adequate to provide electrical insulation?  Somewhere between 3% and 5% by volume should be sufficient to space flat surfaces.

[jpanhalt]

Ordinary (not heavy build) insulation on magnet wire is on the order of 25 microns or less (thin wire can be as low as 5 microns). 

I would not rule out shellac, enamel (a catalyzed version will cure quickly), or polyurethane varnishes.  Lacquer is a great finish but is not very durable and tends to crack.  If you want lacquer, an acrylic spray conformal coating might work.  Also finishes for fabric covered aircraft are designed to be flexible.  The earliest  dope and later polyurethane finishes might work.  Butyrate dope is similar to lacquers but is more durable.

[richard.cs]

Why not just mix 80 micron solid glass beads in the epoxy to ensure a guaranteed minimum bond line thickness adequate to provide electrical insulation?  Somewhere between 3% and 5% by volume should be sufficient to space flat surfaces.

Another vote for this suggestion.

[ZeroCubic]

Why not just mix 80 micron solid glass beads in the epoxy to ensure a guaranteed minimum bond line thickness adequate to provide electrical insulation?  Somewhere between 3% and 5% by volume should be sufficient to space flat surfaces.

That is a fantastic idea!

[Benta]

What exactly is a "micron" here?
A micrometer (0.001 mm) or some obscure Imperial unit? Please be precise.
Thank You.

[Ian.M]

The term micron and the symbol μ were officially accepted for use in isolation to denote the micrometre in 1879, but officially revoked by the International System of Units (SI) in 1967.

[Benta]

The term micron and the symbol μ were officially accepted for use in isolation to denote the micrometre in 1879, but officially revoked by the International System of Units (SI) in 1967.

Thank You. Clarity is fundamental.

It still doesn't answer my question to the PO. This was on my mind:

1 micro inch (µ”)   = 0.000001 x 25.4 x 1000 µm (micron)

Again, please do not use ancient units. It wastes time for everyone. "micrometers" from the start would have saved time and effort.

Thank You.

[Neomys Sapiens]

KAGER does supply ceramic particle filled epoxies and castable compounds.

https://www.kager.de/en/hochtemperaturprodukte/keramik-kleber-giessmassen-coatings-hochtemperatur-epoxies-reparaturmassen/

[ejeffrey]

A common technique in cryogenics is to use thin wire varnish along with cigarette rolling paper.  This lets you put some compression on the varnish while it cures to get a thin bond line without risking short circuits.  The same would probably work with different adhesives and coffee filter paper or any other thin paper product that won't just fall apart when it gets wet.

[thm_w]

Thank You. Clarity is fundamental.

It still doesn't answer my question to the PO. This was on my mind:
1 micro inch (µ”)   = 0.000001 x 25.4 x 1000 µm (micron)

Again, please do not use ancient units. It wastes time for everyone. "micrometers" from the start would have saved time and effort.

Thank You.

No one uses "micron" to mean micro inch, its 0.001mm. Very commonly used term in industry.

https://www.mcmaster.com/about-filter-cartridge-micron-ratings/
https://www.google.com/search?client=firefox-b-d&q=1+micron+in+mm

[duak]

Polyimide (Kapton) tape is available in 1 mil (25.4 um) thickness and can have silicone adhesive on none, one or both surfaces.

I understand that there might be the occasional pinhole in the film so two layers may be needed to ensure at least one layer everywhere.

[richard.cs]

"mil" of course is a more ambiguous unit. English-speaking North Americans use it for 0.001 inches, in the UK (and possibly other Commonwealth countries?) 0.001 inches is 1 "thou" and "mil" is usually shorthand for "millimetre".

As a UK-English speaker I recognise "micron" to be equivalent to "micrometer", but it feels dated like something my parents would say. I'd agree with Benta and try to avoid it on an international forum.