Glue for amorphous C-cores

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
It is often recommended to glue the core halves of power transformers and chokes, where high magnetic forces forces due to AC flux causes the halves surfaces to pound against each other, resulting in 100Hz noise and harmonics.

Easier said than done though. It seems achievable on precisely cut cores that match well, but on amorphous ones with a lot of gap due to poor cutting, I need a strong, inflexible glue that will fill the space between and won't shrink. My later attempt to glue the halves with a MMA glue on a pair of amorphous core choke was a disaster of buzz noise, while the same kind of glue works well on a choke with a HiB well cut core.

Any recommendations?
 
Ideally, you would need a purely mineral cement. The problem is that they need a highish temperature to cure: I don't think you will find anything working below 250°C at the very least.
This could be tolerable for the cores alone, but typically the coil-former and winding are also going to be part of he curing process, and this could be the real obstacle.

There are probably room-temperature alternatives: for completely different reasons, I have recently experimented with mixes of sodium silicate as a "resin" and metal-oxides as hardeners and long-term stabilizers, and the result seemed to be hard, adherent and stable, but I didn't explore these properties in detail since my interests were elsewhere.
If you explore that route, you should document it exhaustively using the net, and be prepared to experiment by yourself.

A less promising but safer and proven option is to use a good quality resin (epoxy, if nothing better is available) mixed with a high proportion of a hard mineral filler: many are availaible, for example glass microbeads, ceramic powders, silica powdered, etc..
You should use as much filler as possible, but the mix still needs to remain "sticky" in contact with a surface.
To use it properly, you must not attempt to spread it: put a generous (meaning excessive) dab or nut in the middle of the surface to be glued, and apply a strong compression during the curing time.

This will ensure that there are no air pockets trapped inside.

Be aware that this is a one-way process: the excess compound will lock together the core halves and the coil-former, meaning any error will be definitive (for some reason, the resin/filler mix tends to be more adherent than the resin alone)
 
Ready-made epoxies with fillers are probably not going to fit your bill: they tend to be reasonably general-purpose, meaning the filler proportion is not extreme (and its nature may not be optimal for your purpose). Even putties are somewhat runny and self-levelling.

If you shop from a dedicated resin seller, you will be able to buy the ideal resin, and the ideal filler separately.
You will then be able to experiment by yourself on blanks what works best (it takes very little raw materials to do a meaningful test: also buy a cheap precision weight-scale).
 
There are probably room-temperature alternatives: for completely different reasons, I have recently experimented with mixes of sodium silicate as a "resin" and metal-oxides as hardeners ...

Sodium silicate is the keyword: Mix an aqueous solution of sodium silicate (yes, the agent that once has been used as a preservative for eggs) with talcum powder to a pasteous consistence. It cures within a few hours to something very hard.
But why would one use C-cores with other than high-gloss polished interfaces?

Best regards!
 
Of course it is important the filler is not a metal powder. What about the JB weld?

The advice with the sodium silicate is good, thanks gentlemen! A thing remains though - is it chemically neutral to the core surface?

Kay, I've yet to find well polished amorphous and nanocrystalline cores. Still not sure why though.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.