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

[thingstodo]

Originally Posted by Astro

I think the 75kW may be peak not continuous.
On the Custom EV Performance web site it is called the "75kw AC induction EV motor".
75kw AC induction EV motor customevperformance.com

Thanks for the link. 10X power out of a motor ... perhaps it is wound for a much higher voltage than the original (my search showed 1800 rpm, 60 Hz, 230/460V) and perhaps they balance the rotor so that it will spin faster?

Lets see - I have seen curves for standard induction motors that show maybe 3X - 3.5X the torque at the top of the operating curve with over 4X the rated current. I have not been able to verify any of this with measurements. But lets say they can get 3.5X the torque for a short period of time.

And I have read about some of the boys in Australia and New Zealand spinning 50 Hz/1500 rpm motors at up to 5000 rpm. I'm not sure that I would go that far ...

At 3X the max rpm, you could get 10X the power ... 3.5X the torque and 3X the rpm .. if the torque held constant above rated speed. It does not. The torque drops relatively linearly with speed over rated .. so that does not work.

But if you raise the maximum voltage WHILE you overspeed the motor, keeping the current constant as you go above rated DOES work for keeping the torque constant as you overspeed the motor ... so a 230V motor running at .. 690V .. could have 3X the speed AND maintain the torque.

If they are re-winding the motor anyway, maybe they can re-arrange the coils, putting them in parallel instead of series? That could drop the voltage from 690VAC to a reasonable value ...

Can you share info on these custom wound motors? Voltage, speed and torque?

[MPaulHolmes]

I just want everyone here to know that "off topic" never ever bothers me. You can have every comment be about Danielson's different girl friends from the karate kid movies, and I'd be fine with that. I actually prefer to sprinkle a variety of topics with controllers.

[e*clipse]

Originally Posted by thingstodo

Sounds like a good application. I'll check around and see if one of the local (in Canada at least) Highlander Hybrids has had a tree jump out in front of it ... and become a source of parts (salvage wreck).

100 lb for the motor, gearbox and differential is AMAZING! I may even call Toyota and ask what a replacement costs. Likely $20K but I should ask

If you can find a junked Highlander Hybrid or Lexus equivalent, you can get these REALLY cheap! I'm talking a few hundred dollars, because no one knows what to do with them.

I must warn you, a big downside is the voltage they want to run at - think 2nd generation Prius. about 450VDC > 650VDC. I did find a thread where a guy tried to use one of them (sorry, I don't have the link) and did get it to work, at something like 200V or 300V. However, he was very dissappointed with the performance. (obviously)

[MPaulHolmes]

A boost stage would work. Another thing similar in size to the controller after the battery bank that feeds the controller. That could be a handy thing for people who want to run a 144v pack on a 230vac motor too. And I think it could create a very stiff constant voltage supply for the controller. I wonder what would happen w.r.t. regen. I would have to look at the schematic of a boost converter.

[thingstodo]

Originally Posted by thingstodo

Thanks for the link. 10X power out of a motor ... perhaps it is wound for a much higher voltage than the original (my search showed 1800 rpm, 60 Hz, 230/460V) and perhaps they balance the rotor so that it will spin faster?

Lets see - ...

Just to be clear .. none of this information was thought of my be. It's a repeat of things that I read from people in other groups. If I wasn't so lazy I would find the references and link to them.

[thingstodo]

Originally Posted by e*clipse

If you can find a junked Highlander Hybrid or Lexus equivalent, you can get these REALLY cheap! I'm talking a few hundred dollars, because no one knows what to do with them.

I must warn you, a big downside is the voltage they want to run at - think 2nd generation Prius. about 450VDC > 650VDC. I did find a thread where a guy tried to use one of them (sorry, I don't have the link) and did get it to work, at something like 200V or 300V. However, he was very dissappointed with the performance. (obviously)

This is not a big downside for me.

High voltage and low current means small cables .. that you can actually BEND ... along with *LOTS* of connections between batteries. I'm OK with that.

If I get the Warp-11 to fit into my Mazda MX6, the Mazda will be running a 380V pack (Renault/better place 60 A-h surplus pack) and a Netgain Industrial controller rated at 400V, 1200A. The Netgain can limit the voltage to the motor separately from the bus voltage (which is COOL). The warp 11 should stay in the 160V to 170V maximum. It''s a ridiculous amount of torque to put into an MX6, but POWER is gggGGGGOOOOOOOOOODDDDdddd!

For the Highlander drivetrain, 180 smaller cells, like perhaps the CALB CA-40s would do about 608V fully charged, but the charger would go up to maybe 640V. The pack would be about 23 KW-h and weigh .. 600 lbs. But where would you buy a charger?

[thingstodo]

Originally Posted by MPaulHolmes

A boost stage would work. Another thing similar in size to the controller after the battery bank that feeds the controller. That could be a handy thing for people who want to run a 144v pack on a 230vac motor too. And I think it could create a very stiff constant voltage supply for the controller. I wonder what would happen w.r.t. regen. I would have to look at the schematic of a boost converter.

I remember reading about this somewhere ... AEVA? Yep. Here's the link big batteries and step up DC converters - The Australian Electric Vehicle Asn - Page 1

Its a few pages long, and I asked some dumb questions as I usually do. Their conclusion appears to be that the lower battery count and cheaper BMS are not enough of an advantage to offset the added complexity and cost of the boost device if there is regen to deal with.

[e*clipse]

I am really happy to see you are using the ring capacitors for this control.

I've been working with several ways to use the ring capacitor and the SOT-227 package IGBT's. I like them 'cause they're sooo compact and still use a bolt-on connection.

I would like to work out a system that could use 6 or 12 (2 paralleled) IGBT's and the ring capacitors. Working with the physical layout in order to reduce the system inductance is an important issue when working with those ring capacitors. I think I've worked out some possibilities that are significantly better than the suggestions in the ring capacitor application notes.

BTW, the IGBT that I'm considering is the Microsemi APT150GN120JDQ4. Yes, it does include a diode, and it's rated at 150A @ 75C. It's also rated for 1200VDC, so it will work with my motor's high voltage requirements. It's supposed to be easy to parallel, and is also supposed to be very robust, including a built in 5 Ohm gate resistor. All this for about $52 in qty's of 1 or $49 in qty's of 10 from Digikey.

150A would be just fine for my application, and if I want to push things (of course ) I could add a second one fairly easily. The small SOT-227 package is pretty easy to work with and fit in various configurations with the ring capacitor.

Would you have any physical dimensions for the driver boards? I would like to see how these integrate with the systems I'm designing. ( I'm doing this work in SolidWorks ). If it's not too much to ask, the interface pinout hole position would be nice...

Also, from the schematics, it looks like the driver IC is directly soldered to the control board, is this true? Would it be best for the system for the driver board to be right next to the control board, or wherever the IGBT it's controlling is?

When I come up with some useful pics, I'll post them for everyone...

[e*clipse]

Originally Posted by thingstodo

This is not a big downside for me.

High voltage and low current means small cables .

POWER is gggGGGGOOOOOOOOOODDDDdddd!

For the Highlander drivetrain, 180 smaller cells, like perhaps the CALB CA-40s would do about 608V fully charged, but the charger would go up to maybe 640V. The pack would be about 23 KW-h and weigh .. 600 lbs. But where would you buy a charger?

I haven't completely answered this issues for my project, but here's what I've got so far:

1) Nissan Leaf battery pack (actually two of the large modules that make up a standard leaf pack. I may need about 24 more modules to completely match my voltage requirements - we'll see....

2) My solar system puts out about 300V per rack. If I'm crazy enough, I could hook those two racks together for 600V. On the other hand, a boost converter would prolly be a lot smarter solution.

[MPaulHolmes]

The 6 drivers are on the same board, but have built in very high isolation and common mode noise immunity. It's the FOD8316. FOD8316 Fairchild Semiconductor | Mouser
it also has a desaturation isolated fault signal that goes right to the micro and a hardware fault circuit. Being a 4 layer board with no traces on the internal +5v and ground planes, all signal paths have a very small loop area. Also all signals are separated sufficiently so as not to cause cross talk. There are no cables that could pick up noise. Also, all signals that are square waves don't change sides of the board since they radiate at sharp points. I'll post the hole coordinates tomorrow. Typing on phone. Hate tiny buttons!!! Haha