[Wirecutter]

Ok, maybe I'll kick off a controversey here, but maybe we can learn something.

I've built a couple electric gokarts now, and I've done a fair bit of research on motors. I wanna go fast. DC motors seem to dominate the DIY EV world, but the commercial attempts seem to favor 3 phase AC induction motors. From what I can tell, 3PAC motors (I made the designation up for the purposes of this discussion) are a far more robust and simple design. (Ever take one apart?) I understand how they work, more or less, and I have a handle on why the controllers for 3PAC motors are so complicated and expensive. 3PAC motors tend to also be more efficient, IIRC, and they can have a much broader torque range. In short, they're capable of being hands-down better than comparable DC motors. It's possible to design an EV with a 3PAC motor that requires no transmission, and I think that's cool.

What I don't understand is why a 3PAC motor is so much more expensive than a comparable DC motor. The 3PAC doesn't need brushes and generally has less to worry about wearing out. My favorite 3PAC motor is the Brusa ASM810, or whatever it's part number is now. It's rated 40 HP for one hour and 87 peak. It's about 13.5" in diameter, 10" long, and weighs around 150 lbs. It's end-stackable, so if you need more power, add motors. Very cool concept. But Geez Louise, it costs over $15,000, and the recommended controller is over $25,000. It's cool, alright, but not that cool.

Ok, I have a small machine shop in my basement. What I've been thinking is how hard can it be to build a comparable motor? I'd need a bigger lathe to turn the motor housing. I don't have a source for the silicon steel sheet for the stator core, and I don't have a way to punch the plates out. Well, I should be able to get a stator core out of an existing motor, right? I can ditch the heavy steel housing and replace it with aluminum to lighten things up and facilitate cooling. The housing could even be built with water passages for cooling. And although it's a slow, tedious process, it's possible to wind a motor by hand.

I know that some thought goes into making a motor, with things like current densities, core losses, saturation, and balancing of the moving bits. It's not exactly trivial, but that $15,000 number makes it sound a bit, well, "NASA". Am I really this ignorant?

So does anybody out there "hot rod" electric motors, or know someone who does? Gear heads everywhere have been building and modifying infernal (sic) combustion engines since forever. Is this another budding area of interest to anyone?

- Mark

 

[Adrian]

OK, let's say you manage to build a great 3PAC motor in your shop. Then what? It appears that a lot of the great performance from these motors comes from the highly optimised controllers - which is why no one will sell you just a motor, you have to buy a motor/controller package.

Not to bring up the never ending "why can't I build my own controller?" thread that bogs down the EVDL, but a 3PAC controller would be the equivalent of 6 DC controller power sections (!) all working together in a very complex manner.

I think Lee Hart wrote a bit about making/rewiring AC motors for use with electric vehicles... It can be done, but I haven't seen any examples. And what would you use for a controller?

 

[Wirecutter]

Oh, I certainly realize that the controller represents a considerable amount of complexity. I believe it would be more complicated than just getting 6 DC power sections to work in concert, too. One thing I've considered is similar to my idea about the motor - "hot rod" an existing 3PAC motor controller. Yeah, yeah - I know I'm making it sound a lot simpler than it really is, but for smaller scale experimentation and "proof of concept", it could work. Start with, say, a 3 phase VFD, like the thing I use to run my 3-phase milling machine on single-phase residential power. It wouldn't be so hard to run one of these on DC power - the first thing a VFD does is rectify the incoming AC line into filtered DC - albeit a somewhat higher DC voltage than most DIY EVs. They also operate at much lower power levels than a typical EV, and that would have to be addressed. (probably first by starting with a much smaller motor)

I wouldn't expect to instantly produce my own Tesla or Wrightspeed-style drive train. I'm considering this for my next long-term project. I'm probably following more in the footsteps of Dr Emmet Brown ("Back to the Future") than of say, Thomas Edison. If I pursue this, I'm sure I'll have to get more accustomed to the smell of burning semiconductors. Perhaps I (we?) can come up with an "open source" design for a 3PAC motor controller (platform?). We all know how closely guarded and proprietary motor control designs are - it sure would be cool to get a good one out in the open.

But back to the motor - I'll look for info on this Lee Hart guy, but if you have links to his work, I'm interested in seeing them. Thanks.

-Mark