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

[nlc]

Paul, what is the max current you will do with your controller ? Surely above 200A ? Do you think 600uF on the DC bus will be sufficient ?

I will maybe developp a 700V / 200A AC controller in next month, and DC bus capacitor really becomes the most expensive part of the controller

[MPaulHolmes]

Hi nlc! I'm hoping for 200amp. My friend's 3 phase controller uses 760uF, and is rated for 400ampRMS and uses a DC bus of 330v. It does cause ripple from the batteries, but works very well. He also did one that is rated for around 200ampRMS and uses 380uF. So, rather than basing anything on math, I'm just basing it on him doing it and it working. haha.

It's all assembled, and now I'm testing all the control board's features.

700v 200amp would be awesome!!

[nlc]

Originally Posted by MPaulHolmes

Hi nlc! I'm hoping for 200amp. My friend's 3 phase controller uses 760uF, and is rated for 400ampRMS and uses a DC bus of 330v.

It's a commercial controller or he build it himself ?

It does cause ripple from the batteries, but works very well.

I believe the ripple is less important when using FOC, because of the simultanous PWM on 3 phase with PWM configured in centered mode.

He also did one that is rated for around 200ampRMS and uses 380uF.

Ok thus the 400A controller is diy too !!

So, rather than basing anything on math, I'm just basing it on him doing it and it working. haha.

No really other solution, I never find magical formula to choose the capacitor value. I believe the most important thing is the ESR, not the total capacitor value.

[DigiSoft]

I have 600V 5600uF capacitor. Can I use that? Is it too big?

[MPaulHolmes]

No such thing as too much!! But the ESR IS THE problem

[nlc]

Originally Posted by DigiSoft

I have 600V 5600uF capacitor. Can I use that? Is it too big?

Don't forget the precharge

[DigiSoft]

Originally Posted by nlc

Don't forget the precharge

yes you are right, it will need lot of precharge time.

Can you give me schematic of this project? Thank you.

[MPaulHolmes]

Absolutely no stinking mistakes on the AC control board or driver boards or power board. This has never happened. haha. All the safety features work perfectly. I finally assembled the whole thing, and tested the drive section without anything hooked up to the 3 phases. The -8v to 15v is working very well. the poor igbts are being pile-drived into submission on turnoff. haha. I'm going to do a video showing what the pwm signals of each of the 3 phases look like for V/Hz control. It's pretty interesting.

I have one major problem. The AC motor that actually has an encoder is only good up to like 1 amp per phase. haha. I feel fairly sure that it won't be sufficient resolution to test the FOC code on that motor. The big fat motor can handle higher amps, but the motor post is too big for the encoder from the company that I've ordered from before. So, if anyone knows of a good "encoder" company I'd be happy to hear about it, yo!

Yes, I'll post all the schematics once I see that it doesn't blow up under load. haha.

And yes, that other controller was a commercial design, but was very clean, using an intelligent power module from Powerex I think.

[nlc]

Originally Posted by MPaulHolmes

I have one major problem. The AC motor that actually has an encoder is only good up to like 1 amp per phase. haha. I feel fairly sure that it won't be sufficient resolution to test the FOC code on that motor.

You are right, 1A is very low current, resolution will be very low. What chip are you using for hall effect current measurement ?

Originally Posted by MPaulHolmes

The big fat motor can handle higher amps, but the motor post is too big for the encoder from the company that I've ordered from before. So, if anyone knows of a good "encoder" company I'd be happy to hear about it, yo!

Yes, I'll post all the schematics once I see that it doesn't blow up under load. haha.

If your first goal is to test the controller hardware with high output current, think you can easy test it on your big motor without encoder, just using your FOC algorithm in open loop mode ? In my case, I do this by simulating a rotating rotor position at the speed I want instead of real motor position. The result is a rotating field on stator with the speed I want and with the current value I want depending on my current consign. It permits to send a lot of current without load on the motor, thus testing the full current capacity of the controller without a brake on the motor.

I think you can do the same by simulating the output speed value of a virtual encoder ?

Originally Posted by MPaulHolmes

And yes, that other controller was a commercial design, but was very clean, using an intelligent power module from Powerex I think.

For the PMAC 700V/200A controller I will maybe have to develop in next month, I will probably use powerex products. They have very good thing at very competitive price !! As an example, a dual IGBT module in 1200V/300A configuration costs 135$ : Darrah Electric
They also propose the corresponding isolated driver board for 14$ : VLA503-01 Powerex Inc | 835-1065-ND | DigiKey
and the isolated supply for 14.7$ : http://www.digikey.com/scripts/DkSea...-15242&cur=USD

Thus total cost become 3x150$ + 6x14$ + 6x14.7$ = 622$.
"Just need" to develop the 3 small board to connect 2 drivers module and 2 corresponding isolated supplies, the control board, and the DC bus capacitors filtering board.

Really interesting I think to quickly develop a 700V/200A AC controller when very low price is not the first consideration !!

[jackbauer]

Paul , what about replacing one of the bearings in the big motor with an encoder bearing?