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

[MPaulHolmes]

The FOC code is running through over 10,000RPM (2 pole pairs) in order to produce 5,000RPM of revolution. That may be too fast for the 10kHz control loop... I wonder if we should bump the A/D readings up to 20kHz, and do the IGBT switching still at 10kHz... That would mean needing to boost the microcontroller frequency from 14.7MHz up to 29MHz. That's no big deal. It's a 30MHz microcontroller. Fortunately you don't need to go over 10,000RPM mechanical! haha.

[MPaulHolmes]

usually DC motors are rated for quite a bit slower operation than 9000Hz. It's still coupled to the AC, right? Is there a chance it could throw a copper rod on the rotor while spinning it up past 5000RPM? The RPM graphs for netgain only show data up to 5000RPM.

[thingstodo]

Originally Posted by MPaulHolmes

This is awesome info. I just made a change where it will show every fault that happens rather than just one of them.

That sounds like a good improvement!

Also, I can make the rotor test probably 100 times faster now too.

It is a bit long to wait through. Faster is better if it's just as reliable. If it's even a bit less reliable, leave it long, in my opinion!

Although we know what your rotor time constant is, so you probably never need to run it again. haha. Once you know it, it's fixed for the life of the motor.

Is that one of those things that we need to run the autotune for when we change a motor at work?

In fact, anyone in the country that wants to know what their EVTV rotor time constant is, we can now tell them. It's a critical variable in the FOC code, and is NEVER on the name plate. And in fact even when I've called the motor manufacturers, they never even knew what a rotor time constant was, let alone had the value available.

So, just for posterity, that motor's is about 0.046 seconds.

Excellent!

How is your 24v supply hooked up?

Not sure I understand the question, but here goes:
- 12V battery through switch and fuse to dc/dc
- 24V dc/dc output to the terminals on the controller
- steal 24V for the encoder, since it won't work under 9V ... and it *REALLY* seems to prefer 24V. Use the ground on the encoder connection to the control board

[thingstodo]

Originally Posted by MPaulHolmes

The FOC code is running through over 10,000RPM (2 pole pairs) in order to produce 5,000RPM of revolution. That may be too fast for the 10kHz control loop... I wonder if we should bump the A/D readings up to 20kHz, and do the IGBT switching still at 10kHz...

Are you doing averaging on the A/D readings? Are there other signal conditioning delays? Not that it's bad to increase the speed, more speed is usually better .. but I'm curious how you deal with the noisy environment

That would mean needing to boost the microcontroller frequency from 14.7MHz up to 29MHz. That's no big deal. It's a 30MHz microcontroller. Fortunately you don't need to go over 10,000RPM mechanical! haha.

10,000 rpm mechanical would vibrate a bunch of stuff apart in my test rig. I don't think there is much torque in the band above 6000 or 7000 ... if I remember the Siemens curve. I can't lay hands on it right now.

[thingstodo]

Originally Posted by MPaulHolmes

usually DC motors are rated for quite a bit slower operation than 9000Hz. It's still coupled to the AC, right? Is there a chance it could throw a copper rod on the rotor while spinning it up past 5000RPM? The RPM graphs for netgain only show data up to 5000RPM.

Hmm. Good thought. I should check. I think I have a manual around here ... the redline was in the 5500 range but I'm sure I have something about Netgain testing the motor balance into the 8000 or 9000 range.

The AC motor is only rated for 9700 rpm.

If I can't locate anything that states 'the DC motor will do 9000 rpm' then I will decouple it before the next test.

[e*clipse]

Wow! and a big BANG!

Shoot - this sounds like too much fun - LOL!

So, I decided to join in. I've got this old abused 1st gen controller board powered up. When supplied 15V, I have a 25V and 5V supply output. I've checked a few other places on the control board and they seem ok, except for the 5V high supply to the driver boards.

Paul - sorry this took sooo long.
The LED is not on; not sure if the PIC is on; it's supplied with 5V.

Do you have any "wake up" or easy spin test code?

If I can get this working, I have a 3phase power stage that can be used for testing. It uses Powerex modules for driving some smaller IGBT's.

Also - thingstodo - you can get a very nice OE throttle pedal for really cheap! It has 2 hall-effect sensors and is already connected to a pedal! The Prius throttle pedal does all this, and it's been built to Toyota quality specs.
04 05 06 07 08 09 Toyota Prius Gas Pedal Throttle Accelerator 78120 47050 | eBay

I've got the service data for it if you're interested.

[MPaulHolmes]

And here's just a tiny radio shack replacement:
HRS100SSAB180 Honeywell | Mouser

Easy to test forward and regen. I guess you have to get 10 of them for it to be $25. dang it.

[MPaulHolmes]

No averaging on the current readings at all. There is an RC low pass filter for the current inputs, but it is very small, like 1microSec time constant.

sorry I just read the message about the first gen control board. I'll have to look around for the code. It's possible that it had debug code on it rather than regular code, but it was also a "pure sine wave" inverter code. I was messing around with that. I'll see what I can find.

I'm adding a new feature until you get a better throttle:

2<cr> increases throttle a bit
1<cr> decreases throttle a bit.

so, you could spin the motor up and backwards too.

Also, another new command is:

off<cr>

That's for if you want to trigger a fault and just turn it off. it will stop running the motor and then you can turn it off without wear on the contactor.

By the way, one danger of the bad throttle under very high RPM is, if it gets disconnected, while falling to zero, it may land on MAX_REGEN (which is near zero. remember that zero throttle is 512, or 2.5v), which will then try to command 400 amps into your battery pack, using the energy stored in the spinning motor. That could cause a catastrophically high voltage pushed back into the pack that could break something.

Another new command is:
run-inductance-test

Basically, you use your artificial throttle to rev the motor up pretty good (3600rpm?? whatever!). Then type run-inductance-test. it will dump some data. The first list of data is the voltage on phase 1 to neutral (if Y connected motor) and it is phase1 to phase 2 (or phase 3 I don't know haha) voltage if it's a delta connected motor.

The next list of data is the current on phase 1 (or line 1, or whatever the crap those words mean). Then, you see what the biggest voltage is, and the biggest current, and then convert them to real voltage and currents (not the lala land of the controller):
real voltage = your fake volt reading * 124 / 1500. 124v since that's your battery pack.
real current = your fake current reading * 600 / 4096
RPM will just be printed out at the end of the test.

L = VoltsYouFound/(2*pi*currentYouFound*RPM)

At least I think that's how it works. haha. I'm pretty sure it's at least close.

Another way is L = 1/(2*pi*rpm) * sqrt(voltsYouFound^2 / currentYouFound^2 - RotorResistance^2)

They agree pretty well. My motor's stator inductance was about 0.055 vs 0.062Henry for those 2 methods.

EDIT: That may only work where the mechanical speed is approximately the electrical speed. My inductance maybe 0.030 Henry instead of 0.060 Henry. I'm not sure.

[MPaulHolmes]

e*clipse!! I found the old PCB schematic!! I bet the software could very quickly be adapted to it. Just rename a few pins. I did have the V/Hz code working someplace. I don't know where, but it was on an even older board. I think it would be easier to start where I am now and change it to fit that board instead. but this only has FOC code on it at the moment. but the V/Hz was just a big lookup table, and then racing through the table at a given frequency.

I think Q3 needs to be rotated 180 degrees and resoldered.

[thingstodo]

Originally Posted by e*clipse

Also - thingstodo - you can get a very nice OE throttle pedal for really cheap! It has 2 hall-effect sensors and is already connected to a pedal! The Prius throttle pedal does all this, and it's been built to Toyota quality specs.
04 05 06 07 08 09 Toyota Prius Gas Pedal Throttle Accelerator 78120 47050 | eBay

I've got the service data for it if you're interested.

Thanks for the link.

For testing the volume control looking knob is easier to set than putting a consistent pressure on the throttle. Throttles have strong springs (they kinda have to)

EDIT - just like Paul said .. and I just read it.