Paul and Sabrina's Cheap 3 Phase Inverter (AC Controller) with Field Oriented Control

[jackbauer]

What sort of motors are we planning on running with this controller? From what i've read industrial motors don't work too well in an ev. I'd like to be wrong about that as i've a brand new 11kw abb motor still in its packing case looking for something to do Problem with motors over 4kw in europe is that they are 415v delta wound and would need a dc bus of about 700v input to the controller!

[MPaulHolmes]

I heard that industrial motors don't work very well in EVs because they are often being driven with a more simplistic V/Hz = constant approach by modified inverters that were for industrial purposes. With field oriented control, the motor has considerably less heat loss, and a greater amount of the DC voltage makes it to the motor, and has the same torque control as what you can have with a brushed DC motor, but with the added bonus of regenerative braking.

I think it's all in the control software. I believe the industrial motors can have significantly higher power than they are rated for. Let's hope so! There ain't nothing like just trying it and seeing what horrible thing happens!

Edit: Actually, I read up on it, and I'm not sure now. hehe. I hope there are cheap used motors that turn out to work OK.

[NiHaoMike]

Here's a link a friend of mine came across:
Very strong and very small electric motor - diyAudio

In the recent past, some US collages and a competition of racing electric open wheel cars (that raced in conjunction with CART races). The one I saw started with a stock 3 phase AC, 60Hz 30hp motor. They oil cooled it and increased the drive frequency to maybe 200 or 300HZ I think that this resulted in a final 300hp. A web search may find more details.

There was some discussion about using industrial electric motors for EVs. Industrial motors are built many times tougher than design specifications, which gives lots of room for "overclocking". Especially for peak power which is what defines the "feel" of the performance. If a 10x overclock is typical, that means a 10HP motor would be sufficient for a more "everyday" EV with a peak requirement of 100HP. Since hypermilers generally do not need much performance and aerodynamic modifications reduce continuous power requirements, even smaller motors can be used.

(Maybe she would be known as the "EV girl"?)

[mrbigh]

Originally Posted by jackbauer

What sort of motors are we planning on running with this controller? From what i've read industrial motors don't work too well in an ev. I'd like to be wrong about that as i've a brand new 11kw abb motor still in its packing case looking for something to do Problem with motors over 4kw in Europe is that they are 415v delta wound and would need a dc bus of about 700v input to the controller!

That's a lot of weight in batteries....
You better change continent suppliers

[mrbigh]

Paul, so far, what's the status of the AC motor controller, are you pending on something?
Do not forget about Eric Tischer's AC motor conversion, it was very doable...

[NiHaoMike]

Originally Posted by mrbigh

That's a lot of weight in batteries....
You better change continent suppliers

Use a boost converter. (Note that the 2010 Prius has a motor voltage of up to 650v but the battery is only 200v.) As for where you would get an inductor for that much current, EV girl to the rescue again!
high current inductors from copper tubing? - diyAudio

One problem is that most industrial motors operate at 208-230v minimum (for rated power), which translates into 300-325v plus overhead of DC. Not exactly practical for a battery pack. But to step up a 180v battery to 400v with a maximum load current of 200A (translates into about 100HP) at an operating frequency of 15kHz only requires about 16.5uH. That is practical for a copper tubing inductor cooled with oil. One major problem is that the metal of the car would act as a shorted turn. However, hypermilers often add "Kammbacks" to their cars. If the Kammback is made from a nonconductive material, the inductor can be installed inside it in such a way that the metal would be far away enough to not cause problems. I'm not sure about heat dissipation of 1/4" tubing when conducting 200A, but I would not expect it to be a problem as the load would usually be much less, maybe as low as 15A when cruising.
...
And I realized that since it would need to be a bidirectional converter in order to be able to regen, there would be no need for a separate charger buck converter. Simply rectify mains voltage (using a voltage doubler for 120v) to the boosted rail and run the converter as a buck converter to charge the batteries. In a DIY EV where cost is a major concern, it can be very helpful for saving cost.

[jackbauer]

Don't get me wrong i think its a great project i'm just a little concerned that if we start going down the road of bidirectional converters etc it becomes insanely complex and we loose the advantage of a diy controller. What about having an industrial motor rewound to a lower voltage like say 100vac which would work with a 144v dc pack?

[NiHaoMike]

Nonisolated DC/DC converters are among the simplest circuits. It is not very difficult to build one.

[dcb]

Well, to Jacks point, it still will not be as simple or efficient a circuit as rewinding a motor for the desired voltage. I would definitely consider a DIY or shop based rewind before converters. adjusting the cell count in the pack may be an option as well.

[MPaulHolmes]

You can also just run the motor at a lower voltage than it's rated for, without rewinding it. It can have the same torque at lower rpms. I was talking to Fran about that a couple days ago. You could run a motor rated for 220v at 10v if you wanted (OK OK performance would suffer some, but you can do it).

Hey mrbigh! I'm working on the software right now. There are quite a few formulas involved, and a little chip with only fixed point math is what has to do all the sin, cos, sqrt(3), and multiplication stuff. I'm also getting the A/D converter working.