Hummina Shadeeba
10 MW
Why can’t they make the laminations and insulation microscopically thin?
Why not thinner laminations
- Thread starter Hummina Shadeeba
- Start date
Hello,
I get the idea of having very thin laminations to reduce core loss on motors
the very thin laminations does exist, like MetGlas cores, but it will be more expensive, and harder to manufacture into stator laminations, as it is only .025mm thick, and a brittle material
here's a paper about making MetGlas motors for increased efficiency, showing why it's difficult
there are also thin ( .08mm ) classic silicon iron cores, with also some difficulties at making it
I get the idea of having very thin laminations to reduce core loss on motors
the very thin laminations does exist, like MetGlas cores, but it will be more expensive, and harder to manufacture into stator laminations, as it is only .025mm thick, and a brittle material
here's a paper about making MetGlas motors for increased efficiency, showing why it's difficult
there are also thin ( .08mm ) classic silicon iron cores, with also some difficulties at making it
I'm told a decade ago that the cost to change a MAC motor from 0.35mm laminations to 0.27mm laminations was $20 per unit and added 1-2% efficiency.
This efficiency adds to the thermal capacity of a hub motor when climbing. It's very important that we negate the effects of having a fixed gear ratio in that scenario.. removing just 50 watts of continuous heat production does matter.
I cannot convince any motor manufacturer to adopt these thinner laminations and make a higher than average efficiency hub. Truly unfortunate.
Admittedly the cost - benefit ratio on thinner laminations than 0.35mm for most motors is low. high powered geared motors and mid drives would benefit a lot more than DDs.
This efficiency adds to the thermal capacity of a hub motor when climbing. It's very important that we negate the effects of having a fixed gear ratio in that scenario.. removing just 50 watts of continuous heat production does matter.
I cannot convince any motor manufacturer to adopt these thinner laminations and make a higher than average efficiency hub. Truly unfortunate.
Admittedly the cost - benefit ratio on thinner laminations than 0.35mm for most motors is low. high powered geared motors and mid drives would benefit a lot more than DDs.
Hummina Shadeeba
10 MW
What u mean by adding thermal capacity to a hub motor?
I don’t think the grin simulator allows me to show how a super thin lam motor would compare. How would a magical .01mm lamination motor do? (Assuming the insulation between lams is so thin as to be insignificant and the same amount of steel in the stator)
Okay not thermal capacity, but thermal headroom.
ebikes.ca sim doesn't get that granular that you can set parameters like that, maybe reduce some factors that affect eddy current by 5 or 10%, and we might have a 'close enough for government work' approximation.
ebikes.ca sim doesn't get that granular that you can set parameters like that, maybe reduce some factors that affect eddy current by 5 or 10%, and we might have a 'close enough for government work' approximation.
Hummina Shadeeba
10 MW
I think the eddy currents would be reduced much more and half the lamination thickness results in half the eddies, no?
Hummina Shadeeba
10 MW
Oo yea, In my looking I can’t find a rough estimate of what percentage of the eddies are produced where. Stator, magnets, and the back iron too? And I guess the wire
why no one bothers to built poles with multiple magnets I also wonder. The increased cost of thinner lams n multiple magnets is surely worth it for some but unavailable.
Last edited:
Chalo
100 TW
Lams are bonded together with some kind of adhesive. The thinner the metal, the larger the bond lines' fraction of the total volume. At some point there's no additional efficiency benefit to be had that isn't more than offset by reduced stator density.
mxlemming
100 kW
- Joined
- Jul 17, 2020
- Messages
- 1,200
I haven't got much experience in motor design, i focus on the controller side, but i recently went to a coil winding and steel stamping expo (yes as exciting as it sounds...) and it was very clear that no one wanted to deal with thin laminations. Everyone would do 0.35mm (literally hundreds of factories) but 0.2 was like a dirty word for most of them.
I pressed and got a lot of mumbling and vague answers but it seems the material is more expensive since it takes much more rolling and it doesn't play so nicely with the stamping dies.
Practically, it can be seen on the loss vs frequency curves published for various steels that 0.35mm becomes prohibitively lossy around 600hz and 0.2mm around 1khz. 600hz is already very fast for most motors. Most controllers start to struggle at this speed, though it's perfectly possible to achieve speeds in the ~3khz region for silicon controllers.
Last edited:
Similar threads
