I was worried about that, then saw this (https://electronics.stackexchange.com/questions/210640/adding-a-gap-to-a-core-for-an-inductor ) on StackExchange:
Check it again just in case :>
I hope it's not too bleedingly obvious not to put a shorted turn into a transformer? But, point taken, sometimes doesn't hurt to be specific. Plastic and composite insulating materials are fine here. Metal is not.
Even more importantly for production purposes, they need to be part of the insulation system used in the part, which will further limit selection. That is, a range of materials that have been UL-approved (or other) to work together in a component up to the rated temperature.
As Andy said in the other answer, even paper will do, if it's just a few-off sort of thing. Mind that's class A insulation, not rated for very much in particular, and won't exactly have you passing UL, but it's unlikely you'll make a few and seek such approval.
As for grinding ferrite, I do not quite agree. It can be easily ground and lapped for a close fit if done with water -- just like ordinary glass can be ground and lapped. Maybe the core configuration matters?
I mean, dry is fine... it's not a particularly strong or hard material as ceramics go. It's comparable to glass or porcelain. Actually let me see here... NiZn can definitely scratch some, I think this is soda-lime glass; and maybe with difficulty, glass can just barely scratch ferrite, but it might also just be scuffing it and nothing more. MnZn ferrite seems comparable or maybe a bit harder. Can't seem to get NiZn and MnZn to scratch each other very much, any edge just kind of crumbles and it's hard to see scratches on the textured surface*. Anyway, the only thing you need water for, is clearing dust, generally lubricating the cut, and only at high speeds, cooling; and any other liquid will do just as well. You'll be working very hard indeed to need cooling by hand.
*Even a lapped (but not polished) surface has a granular appearance due to the polycrystalline makeup, probably thanks to cleavage planes giving randomly smoother or rougher micro-surfaces on individual grains.
Personally, I've used SiC sandpaper, diamond-embedded stones, and tile saws, to work with ferrite. Of those, the tile saw is obviously the most aggressive, and tends to chip out material in a saw cut, with finish depending upon how smooth-running the saw is, and how well you're able to guide the material (so, by hand, not very clean, lol).
Other comments for readers' benefit...
Gapping the whole core, will probably make spring clamps unusable. Or forcing them on will crush the ferrite (particularly if you've only gapped the center or something..!!).
The EMI comment is simply about fringing fields around the airgap, which does have some impact, albeit small. Obviously it's in proportion to distance from the component; a several-mm gap could well cook an adjacent (touching, or nearly) clip, heatsink or capacitor body, for example (induction heating), in a very tight (compact) power supply design. That's bad news for efficiency, and another good reason to keep the gap in the center. Unfortunately, the same is true of the wire, and litz is often required...
Thanks, it would be good if the following Mylar Film told us the tolerancee on the thickness.
https://uk.rs-online.com/web/p/plastic-film/7850815
That link is to mica film, not mylar film. Mica is a mineral, and it is going to have a much rougher texture than a plastic material.
Cleaved mica surface is atomically pure, at least until you reach the edge of the next layer(s). Note that mica is almost exclusively synthetic, so can be far more consistent than naturally mined minerals, if that's where you were going with that.
Using rocks to support rocks is rather appealing, if I do say so myself. The challenge is really just getting it flaked accurately enough. If you need it more accurate than, say, a skiving setup with polished carbide blade can afford, then gluing and lapping is probably your alternative.
Which...
I wonder if you can lap thin sheets with a flat vacuum holder
Vacuum pits will have to be very small in relation to the thickness of the material, otherwise the material flexes into them, and now what used to be a flat surface during grinding, bows back to a polkadot-bumped surface sticking out. It doesn't seem like much for bulk materials, and, well, bulky low-aspect things being what they are, sure. What comes to my mind is, I've got a chip of ferrite here that I've lapped to pretty close to parallel, around 80 mils thick; if I take it out and put it on the diamond stone, only the places where my fingers support and push on it (on the back side) take any cut (on the front side). It's not very flexible, but on the scale of 10s-100s µin where grinding cutting takes place, yeah, it's a floppy piece of paper.
Assuming you have some way to release it -- I would recommend super-glue: it can lay down very thin indeed, and then you have a solid backing plate to reference against. Heat and acetone will be fine here, but just to say, you might not always have that option, and if it comes down to mechanical means, peeling things apart is likely to damage them.
Tim
Home
Help
Search
Login
Register
