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

[Stiive]

Hey guys,
first post on this forum, just registered to subscribe to this thread - seems a lot of people working on AC motor controller, this is good. Over the past days I read through this whole thread.

I am also working on a 3-phase DTC controller. Hope to finalise my PCB this week. I will prob make a video about my progress thus far, I have been working on it for about 2 months.
My build is not aimed to be the cheapest method, but hopefully an extremely reliable, safe top range build. Therefore I have decided to use Semikron products. I will probably release the hardware schematics once I'm done and its tested.


FYI nlc, you may consider the Semikron SkyperPro32, it is top of the line 15A max gate driver and includes its own isolation with short circuit/UV/OV protection. Also includes soft turn-off, adjustable deadtime control , HLS and much more. I also read you are concerned of auto-commutation, but these turn off at -7V so is not an issue. I am using these modules for my build, but they are expensive. They also include the open-source schematic if you wish to build one yourself.
I cant post the datasheet because I am a new member, but you can find it on the Semikron website or farnell/element14

Regards,
Stiive

[nlc]

Received my 6 pack IGBT module driver board pcb, my less than 3 year old son helps me to populate the board

http://www.youtube.com/watch?v=T7_gdgxylo0

[KUSAW]

Hi Paul,
I want to order the PCB and kit.....
Hopefully Will be ready soon..
Thank You..

[roguethunder]

Originally Posted by MPaulHolmes

Just tested the AC controller with a DC bus of 308v (288v nominal). It worked! Hurray!! It was only on the small AC motor, which is a few amps per phase. Here's a video of it running at 84v nominal. It's just V/Hz at the moment to check if it was working:
...

Awesome. Any ETA on the large motor test? This makes me have to be less lazy about my thoughts on a chassis ;P... what horrors you have unleashed.

Can't wait to see this thing purring. Hopefully not in a shower of sparks ;P

[e*clipse]

Wow. This is surreal. I'm also working on a motor controller.

Same microcontroller, same safety philosophy, same current sensing.

My main goal is a controller for a high voltage BLDC motor I have. I also saw the same controller could be used for 4 quadrant DC control.

That's cute. I guess you could say it's my progress... We need some "fast" smilies.

Regarding power switches, has anyone looked into the SOT-227 package?

1) Both IGBT's and Mosfets are available
2) The Mosfets can handle >200A @ 200V. IGBTs are available at >1200V. The limiting detail is the connector.
2) you get "bolt together" bus bars.
3) they are very compact
4) They are electrically isolated from the heat sink, like larger modules

Prices are ok; It may be worth paying a bit extra for the easy assembly and capability of replacing blown switches. The compact form factor and easy Isolation are also big advantages.

I would like to use your control board circuit - I think it's nicely worked out and the safety details are proven with the Cougar controller.

I am considering making a surface mount version of the control board. I don't consider surface mount soldering an problem - you just need to learn a few tricks like the ones Nic showed us. I have had great success with many boards in my toaster oven. The advantages I see are better shielding/less EMI problems and the ability to make a very compact controller. It may not be an issue with a car, but this also has potential for a good motorcycle controller.

I like the idea of a separate power/driver board.

Would anyone else be interested in a "compact" version of this design?

[flores]

Sorry.. but this already exists:

http://ecomodder.com/forum/showthrea...rds-19570.html

[e*clipse]

Cool!

That's one step closer to running car. Woo Woo!

Wow. I'm making so much progress just sitting in this chair.

[skeeterb]

Hey Paul, how goes your actual testing? It's been a while since I have posted in this thread. What are the actual prices you are going to have for the kit and just the board if you decide to sell the kits? Anything under one grand is going to be a big coup for us that want to convert to AC Electric. I've bounced around the 'net and the AC controllers are pretty expensive. I hope to raise enough money to convert a car to electric. The biggest issue probably will be the batteries though. I want to go straight to LiFePO4 batteries instead of fighting the limits of the lead acid batteries. I'm still trying to find a formula that I can use to find out what A/h batteries to use. I'm debating on using 120 A/h to help get a range of at least 100 miles and a comfortable cruising speed of at least 70Mph. But I may need higher to get those stats. Well, keep up the good work and keep us updated.

[Anaerin]

Originally Posted by flores

Sorry.. but this already exists:

http://ecomodder.com/forum/showthrea...rds-19570.html

Unless I've missed something, I thought the Uprising controller was for Brushed DC, not BLDC or ACIM (Which is what this is meant for).

[e*clipse]

Unless I've missed something, I thought the Uprising controller was for Brushed DC, not BLDC or ACIM (Which is what this is meant for).

Interesting. There were a number of things that weren't clear on the thread. I do like the connector, but it wasn't clear how the driver boards were connected - through the connector, with everything else??

Also, the project seems to have evolved into a much different direction. I do like the small surface mount construction. However, the controller is not a PIC, so code won't be portable. In other words, all Paul's work to develop code would need to be translated. If there were any improvements in code (from either side), they would also need to be translated. My guess is eventually, no one is going to bother translating.

I need to post a few questions about these details on the other thread.

My concept was to build a code transparent controller, just use surface mount versions of the IDENTICAL components. Where there is an advantage to using through hole, TH components would be used. If this hardware were allowed to evolve differently than the board Paul is working on, there will come a time when they will no longer be compatible. This has already happened with the other controller.

I really like the open source philosophy on the Revolt controllers, and I merely wanted to offer another hardware option. The idea was any bugs, fixes, etc that show up with the TH board should also work for for the SM board.