Hybrid Motor Controller Developer Kit?

[EngMark]

I'm not sure where to ask this question, here in the hybrid section or in the electrical motor control section. I'll try here first because I think the expertise needed is more from the hybrid side.

I am beginning development of a system which would consist of a battery array, a motor controller (the subject of this post), an electric motor (BLDC , approximately 30 kw), and a single wheel which would be coupled directly to the motor. The system would be mounted to existing vehicles (Truck, car, boat) although its only interaction with the vehicle would be to draw makeup power from the vehicle electrical system (limited to 12v 100A, hopefully much less). A typical operation cycle would have the wheel driven by a constant torque (external to the vehicle) for approximately ten minutes (backdriving the motor and putting electrical energy into the battery array in the device), followed by the motor driving the wheel for a similar length of time against a resistive torque of similar magnitude (energy out of battery array). A key function of the motor controller would be to maintain a constant motor torque throughout the cycle despite the change in direction of power, and despite the resulting speed and direction changes (including brief periods while stationary). The motor and controller will need to have cutting edge levels of efficiency so that the net energy expenditure for a cycle is small enough to be replenished by the vehicle electrical system. The motor will be of the "LRK Torquemax" aka "brushless outrunner" type and will be fabricated for this application. This type of motor has been developed in the last few years primarily by the model airplane community where it is valued for its extremely high torque output at low speeds and its high efficiency. Given the available information and work that I've previously done in solenoid design I think that I will be able to develop a suitable motor. The controller however is a mystery to me. Is there such a thing as a motor driver developers kit out there which I could buy and program to develop the concept and to learn from? Or maybe a few key components that could be pieced together to make a controller suitable for this use? Whatever I work with needs to have the capability of operating bidirectionally & regeneratively. My initial development could be done on a smaller scale using off the shelf model motors, if that works out as the best way to go. The market for this device is relatively small (dozens per year at best), so it would be nice if the "developer's kit" electronics package could also be related to the production electronics package. I would appreciate any help in locating and / or developing the required motor controller, and / or pointing me toward applicable texts. I may also consider paying for external development, although that would be from a modest personal budget.

Thanks,

Mark

 

[GregLocock]

A 30 kW, high efficiency, motor controller is going to be a pretty expensive thing. (Depending on what you mean by high efficiency).

Anyway, is this motor a chopped DC motor (ie an electrically commutated 3 phase DC motor)? If so that's what we use on solar cars, and these guys will know what to do.

http://www.tritium.com.au/legacy/legacy.html

We used the "Gold" in a couple of events recently and it is quite programmable.

This looks more like what you want

http://www.tritium.com.au/project/project.html

They are a small and helpful company.

An even smaller, and longer established company is AERL, the brains there is called Stuart Watkinson and he may have a motor controller available. Last address I have for him is

AUSTRALIAN ENERGY RESEARCH LABS P/L
2 ROSLYN COURT
NORTH TAMBORINE 4272
QUEENSLAND, AUSTRALIA

Phone: Int.+ 61 7 5545 0177
Fax: Int.+ 61 7 5545 0866


You might want to look at wind turbines.




Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[EngMark]

Thanks Greg. You weren't kidding, I saw the price of the Tritium controller. Wow! I've learned something already, although not what I wanted there. Yes, my motor is an electrically commutated 3 phase DC motor in most incarnations, although for smooth running at low (near stall) speeds I may find that it becomes necessary to build the motor with additional phases. The Tritium controllers look like the slick solution to getting an EV project going, but I am concerned that they might not have the flexibility to do what I need. For instance, they are set up to conveniently integrate a throttle pedal input, which is great for a car, but my primary input would be a torque request signal. Not sure if it would be able to hold a constant torque (or would I need to have torque sensors and then external hardware and software to generate an appropriate "throttle" signal?) Not asking you to answer that, just throwing it out as an example of where my application might have a difficult time with the Tritium controller. Also, I'm wondering if you have any knowledge / experience in using it for regenerative braking? That probably depends on whether your car is purely solar or has storage batteries. I think you've helped point me toward some good info, but I'm hoping someone else might be able to show me something else closer to my ideal.

Mark

 

[GregLocock]

Torque = motor current, no problem. I'm pretty sure it has regen, but we don't bother.

Here's another contact who is a fount of all wisdom: Paul Lillington, Lillington Motors, Sydney. He built a custom motor controller for us once. Haven't spoken to him for 10 years tho. This may be his address

http://oceania.manta.com/coms2/dnbcompany_5kjgkx

Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[EdDanzer]

You are taking on a very difficult task. Making a controller to do acceleration and regeneration will be difficult as the voltage levels change as speeds change. Regeneration at low speeds will output low voltage at high amperage. It may be difficult to send the regeneration voltage back to your battery pack. You will need some expensive simulation software if you plan on being successful. Something like Matlab

http://www.mathworks.com/

will help with the control development.

Ed Danzer

http://www.danzcoinc.com

http://www.dehyds.com

 

[EngMark]

Ed-

Wouldn't it work to have a DC-DC voltage converter between the mosfets (or whatever the drivers are) and the batteries? Thanks for the Matlab suggestion, I'll look at it.

Mark

 

[EdDanzer]

DC to DC voltage convertors usually are for set voltage out. Step up units are not very efficient.
Remember the battery voltage will be increasing as the regeneration voltage is dropping.
I’m not an electronics/electrical guy but have studied the concepts and how they compare to fluid power. Either system has problems with storing braking energy when charging to a single storage vessel.

Ed Danzer

http://www.danzcoinc.com

http://www.dehyds.com

 

[sreid]

First I'd like to point out that 12 V at 100 A = 1200 Watts which is 1.6 Horse Power (1 HP = 746 Watts). This is why a 30 KW system is more like 300 V x 100 A = 30,000 Watts.

 

[EngMark]

Sreid, you're misunderstanding. The 12v x 100 a limit (1.2kw) would apply to the vehicle power supply. If a cycle were to take say, 20 minutes then the vehicle would be providing up to 0.4kw*hr. That would be fed into the battery bank onboard the device, which as you point out would need to supply up to 100 A at 300 V or some combination of voltage and current to support the occasional peak load of 30kw. That battery bank would be expending energy for approximately 10 minutes per cycle at a very high rate - much higher than the 1.2kw available from the vehicle porer supply as you point out, but only a temporary situation. During the other part of the cycle the motor would be acting as a generator returning energy to the battery bank in the device (up to 30kt minus inefficiencies into the battery bank). If the energy expended per cycle minus the energy recovered during the cycle amounts to less than the energy available from the vehicle power supply (0.4 kwh) then it all works out and it is possible to start another cycle in the same place you started. I admit this would require a very efficient system and that given any reasonable efficiencies it wouldn't be possible to have every consecutive cycle be a 30 kw cycle (if the system operated at say, 85% efficiency the net loss for a 30 kw cycle would be greater than those for a 8 kw cycle). It'll be a game of efficiencies.

I did find BLDC motor developers kits on digikey.com and I've bought two. I'm struggling about with one of them (Silicon labs) and if I can't make reasonable headway in the next few days I think I'll offer a few $$ to a college kid to work with me on this.

Keep the thoughts and suggestions flowing!

Thanks,

Mark

 

[EngMark]

Correction to my post immediately above:

The "kt" in "device (up to 30kt minus inefficiencies into the battery bank). " was a typo. I was trying for "kw" (or should I say "kW). I was not referring to knots ot kilotons.