Sadly the premise that it operates on a thermal principle has been pretty much debunked - it operates on electromagnetism, due to the induced EMF when the ball bearing rotates through a magnetic field caused by the current through it. Here’s a paper describing the effect (which has been experimentally verified)... http://www.physics.princeton.edu/~mcdonald/examples/motor.pd...

I built one of these at university as a research project and it works well (albeit drawing a huge amount of current!). Interestingly enough it works just as well in either direction, depending which way you give it an initial push - the polarity of the DC terminals doesn’t matter at all!

Sadly the premise that it operates on a thermal principle has been pretty much debunked - it operates instead on electromagnetism, due to the induced EMF when the ball bearing rotates through a magnetic field caused by the current through it. Here’s a paper describing the effect (which has been experimentally verified)... https://news.ycombinator.com/item?id=20415755

I built one of these at university as a research project and it works well (albeit drawing a huge amount of current!). Interestingly enough it works just as well in either direction, depending which way you give it an initial push - the polarity of the DC terminals doesn’t matter at all!

Roobert33 on YouTube demonstrates this: https://www.youtube.com/watch?v=f1xnQ9gWy1o

I sort-of feel that anything which uses electricity purely for a thermal effect, is actually in the carnot engine space and not actually "electric" in any meaningful sense of the word. They could have modified a stanley steamer to boil water electrically, and made similar claims (ball bearings aside)

But undenyably very .. cool?

I wonder if you could put a current regulator on it that would provide the minimum current needed to achieve a certain RPM. Seems like the heat issue could just be that after it reaches its maximum speed, the excess current is just wastedin ohmic heating.