How to Control AC 600V PM motor???

[DKirkham]

Hello everyone,

One of our customers has asked us to put an electric motor in one of our cars for him. You can see the cars we make here at out website:

http://www.kirkhammotorsports.com

so you can get an idea of what we are trying to do.

I have searched high and low and I have settled on using the rear electric motor out of a Toyota Highlander. It is cheap, and undoubtedly bullet proof. It comes complete with electric motor and differential as a unit.

I had all the industrial electronic control (CNC stuff) salesmen come by to look at the motor and no one seems able to make it work--let alone even make it move. I then went to the internet and there is precious little information on actually controlling an AC PM motor.

The motor is a 600 volt unit with 8 poles. It has a Hall Effect sensor on it which seems to be one of the sources of our trouble.

We are really fine when it comes to high performance automotive alloy selections, mechanical engineering, and design--but, frankly, electrons are quite mysterious to us, but I am determined to learn, if possible.

Does anyone have any idea on where to point me? Books? Websites? Any help would be greatly appreciated. We prefer to keep the work in house.

 

[itsmoked]

Dang! That sounds like a fun project!

Likely not going to do it with books.

An AC PM motor requires accurate commutation of the supply power.

Another words, the motor has to be 'rewired' hundreds of times a minute, electronically, at the exact correct moments. The Hall Effect sensor is used to locate the particular wiring scheme "of the moment".

How many wires are there?



Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[Skogsgurra]

Hi,

The Hall sensors shouldn't be a problem. They do help, actually.

The CNC guys are not used to that kind of sensors, though. The need encoders and sometimes resolvers. That's why it didn't work with their stuff.

There are quite a few suppliers of electronics for such motors. Googling things like "Hall commutated" produces a nice number of hits, 49 when I tried.

There's a book at

http://www.library.tuiasi.ro/ipm/vol15no14/electrical.html

, which looks interesting.

There is also a guy in Italy that has been doing this for "centuries". He is really good. Now doing work on plastic rotors with rare-earth powder for high-torque PM machines.



Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[DKirkham]

Keith,

I'm sorry, I should have said it was a 3 phase unit. It has 3 wires going to it. Yes, the CNC guys have no idea what to do with a Hall Effect sensor. I think they were out of their field of knowledge and shooting blanks.

I am a pretty determined guy and electrons are one of those last frontiers I have very little knowledge about--but want to know more. Books don't scare me (I have literally read a thousand pages on the subject of PM motors and controls--but not much sticks!

So, I have a pretty good basic (and I do mean basic) understanding of how PM motors work on paper, but I can't find any info on how to really make it work in real life.

Gunnar,

Thanks for the link on the book. I will check to see if I can find it. My wife thinks I am crazy to read such things at night. It is such a fascinating field I know so little about.

Do you know any more info on the "guy in Italy?" Maybe I will stop by on my next trip to visit my factory in Poland.

David

 

[itsmoked]

You kinda have two problems.. (probably more) One is how to control any motor like yours. And then how to step up the power dramatically while doing it.

Would you be interested in hands-on learning how to control these motors? Willing to maybe go to a one day course somewhere?

Not sure where you will find out how to raise the power. By the way, that is an extremely hazardous realm with lethality written all over it.

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[DKirkham]

Keith,

I am sure I have a million problems--most I don't even know about. But, you have to eat them one at a time until nothing else stands in your way--unless I die first. (That would be from old age, not a SHOCK!) I know there are extreme hazards with such voltages and I do appreciate the warning. I can use all the help I can get.

I thought I would first get the motor running from the wall, and then do the testing for regen, accelerator control, etc. Then I would figure out how to get batteries to power the thing. I think the battery power problem won't be as hard to overcome as I have several friends (professors) at the local university (BYU) who already have an electric car and know how to do that stuff. They just don't know how to do it with a PM motor. They use AC induction motors. We are trying to keep the weight to an absolute minimum and PM motors are 1/2 the weight from what I have found.

I would be happy to go to a course or to learn this stuff hands on...but where? I am open to any and all suggestions.

David

ps. anyone need any help in material selection for high performance suspensions on cars??? Now there is a place I can help.

 

[itsmoked]

DKirkham; I was going to recommend the Microchip Motor Control seminar. I took it and loved it. Came with a development board and software that you could run, and multiple algorithms for controlling BLDC motors. The people were very experienced and helped numerous people with specific problems on their projects. It was about $125 and included the board, motor, firmware, etc.

I'm not seeing that class as being available at the moment. You can always inquire and hire them to show up and give you the class. Sorry.

Moving on.. What HP is the Prion motor? Yes you may like the weight of it but jesh!!

If you went with a commercial 3ph 400Hz motor you could buy sophisticated VFDs that would handle all your needs at prices found in a highly competitive product field. I could easily see you spending 1/6th the money and 1/12th the time required for your present course of action.

In my opinion you are making a serious economic/production misstep sticking with the Prion motor.

Realize that a 400Hz motor is 6 times smaller than a 60Hz motor for the same HP.

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[Skogsgurra]

David,

Sorry, he lives in Switzerland. You will understand the mistaken nationlity when you see his address.

http://0-

http://www.uspto.gov.mill1.sjlibrar...eek40/OG/html/1311-1/US07116030-20061003.html

He was one of the guys behind ABB:s drive technology. But is now working as an independent consultant. Hungarian/Swedish/Swiss - a truly international character.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[DKirkham]

Keith,

The Highlander motor is 50KW. The whole unit only weighs 90 pounds. I will look into the Microchip Motor Control seminar to see if there is anything I can find. Thank you very much for the tip.

I am certainly not opposed to using a commercial unit and I will definitely look into it. My big hesitation is Toyota used what they did for a reason...one I most certainly don't understand all the ramifications of. I have enormous respect for Toyota's engineering. They regularly do things the Not-So-Big-Three don't (or won't) think is possible, but that is a discussion for another day.

Sorry for being naive, but I thought the speed of the motor was controlled by the frequency/hertz signal it was given??? If you up the frequency you up the speed of the motor--or so I understood. So, how does a 400 hz motor give any advantage over X hz motor when you are varying the frequency anyway to even drive the motor? I admit my knowledge fades quickly when speaking about electrons and I appreciate your willingness to even answer.

Gunnar,

Thank you for the article. I will look into it tomorrow.

David

 

[itsmoked]

Hi David.

The size of a motor is mostly controlled by its magnetic circuit size requirements. The size is a direct function of the frequency the motor is designed for.

A 400Hz motor will scale to approximately 60/400 of an equivalent 60Hz motor's size. This is why aircraft all standardized on 400Hz - small, very high power motors.

Its a shame the whole US isn't 400Hz so all the transformers and motors would be much, much smaller. The 400Hz has other serious negative ramifications in the area of power transmission over distances. Not a problem in aircraft or cars.

I went to an army base during an open house and looked at a Field Hospital. I asked what powered it and was lead around back to a box that was about 4 ft long, 4 feet tall, and 3 feet wide, with a small black hose running across the large lawn from a deuce and a half tanker. I asked how many kW it put out. Figuring that box.. Not more than 15kW. The operator said 75kW and opened the box. I was SHOCKED! The generator was maybe 16 inches long and 10 inches in diameter. It was being driven by a gas turbine that was about twice that size. I was incredulous! "How can that be 75kW!?", I shouted over the large turbine whine. The Corporal smiled and said, "400Hz, the whole hospital runs on 400Hz".

As for the Toyota I couldn't agree more about the success and creativity they have shown in their hybrids.

The problem you are going to run up against is that many things in the automotive realm are heavily designed for a very narrow specific task. The entire design; car/brakes/controller/motor/batteries are totally optimized to suck every last penny out of the design and provide precisely what is required to get the job done.

The electrical stuff is driven in this direction with huge pressure on the vendors and the production for penny shaving to the max.

So along comes you.. You want to move this motor over for your application. Are you willing to spend a small fortune to create a different controller to run that motor effectively the way you need it to run? What happens when you discover that the motor chokes when you use it exclusively for many miles? Especially after sinking a fortune into it's controller.

I don't mean to throw cold water(saltwater) on your plan but I really think you're biting off more than you can possibly realize in the development realm here. I'd be one thing if you were going to build hundreds of thousands of these things but a one off? I bet Toyota spent far more than a million dollars US designing their controller.

If you really, really must use that motor. Then use everything else too. Controller, engine, battery, and motor. Otherwise go for off-the-shelf with industrial proven stuff to be assuredly successful - in your customer's lifetime.

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[HCBFlash]

Certainly you've looked into the Tesla Roadster!?!?

Please let us all know how this works out for you! and don't forget faq731-376

 

[DKirkham]

Keith,

First I want to sincerely thank you for taking the time to write to me. I must sound like I am crazy--which may be true! Please throw all the cold water my way you can find. It is much less painful now than after I find myself on the side of the road wondering where all the smoke is coming from.

I certainly realize attempting to redesign something that took Toyota millions (I actually hear the number is closer to a billion from my friends at Ford) to develop is quite daunting if not impossible. I probably didn't explain myself adequately. I was actually hoping more toward getting a controller off the shelf that would work, or be modified (without too much difficulty). That is why I called in the CNC guys. (We have several CNC machines and get along well with the service guys). But, they couldn't make the motor so much as move.

That said, I will be the first one the throw the Toyota motor out the window if there is something else out there that might work. It is widely believed in the electric car world there is NOTHING off the shelf that would remotely work for an electric car traction motor at a reasonable weight. I have found a few that are extremely heavy.

Now, that doesn't mean there actually IS nothing that will work. It is just BELIEVED there is nothing that will work. I am happy to try 400 Hz motor or ANYTHING else that might work. My goal is to drive down the road with electrons--NOT to figure out how to do it with a Toyota motor.

My first choice (over a year ago now) was to actually use the controller, motor, and batteries from the Toyota. But, we found out that would not work. Last year for SEMA we built a prototype car for Ford using their new 3.5 liter Duratec motor. It was a nightmare to get it done. The CAN will not allow the motor to run when things are disconnected--such as the ABS, seats (believe it or not), fan, airbags, etc. Ford eventually had to assign an electrical engineer specifically to us to get the car running. We had to go back and forth with 5 flashes of the computer before we got everything running. Of course, in this case, that would be hopeless with Toyota (unless I can convince them to let us do the car for SEMA!!!)

So, any advise on where to get a 400 Hz 50-75 KW motor? I would hope to use a PM motor as all of them I have found are 1/2 the weight of an induction motor. But I am open to ANY suggestions you may have.

HCBFlash,

Yes, I have spoken at length with the guys at Tesla motors. They could not have been nicer. However, they use AC induction motors and very far behind on their own projects and can not help with our project. Their motor is twice the weight of the Toyota motor. Now, in the end, I will certainly use an induction motor if I have to. It is better than pushing--or dreaming which is what I am doing now.

David

 

[ssn596]

Am in pipe dream stage save money for retrofit of 92 IZUZU pickup. That vehicle is chosen cause I have it. My dream would be anything from '30 to 55 pickup, van - as long as it is uncommon and funky I guess is the best term.

I will use 3 phase drive -- at present will probably use Azure (formerly Solectria).
Do NOT let 3 phase drive throw you off or in a big tizzy. It is nothing new. We have been doing it since Tesla and Westinghouse in industry. We have been doing varible freq drive (VFD) for about the last 30. Railroads have been doing it for 20 years, nearly all electric forklifts produced for about the last 5 years are 3 phase drive.

Now for your dilemna.
All the above are generally induction motors ie squirrel cage. You are using permanent magnet 3 phase also known as brushless DC - who cooked up that title?? The nearest comparison to the PM 3 phase is a synchronous. Now in an induction the rotor does NOT have to rotate at same RPM as stator (this is "slip") in fact it cannot because with no slip to make a long story short there is no torque. NOT true with synchronous the rotor MUST be at same RPM as stator.

With the PM 3 phase the controller does a bunch of calculations based on magnetizing current (ie motor not mechanically coupled) compared to total current when loaded. The torque is also calculated from these. The controller must know the mechanical position of the shaft and must have some kind of encoder to sense that. From these two (uhhh three) basic values the relative positions of the stator field and the rotor are derived to give the angle between the two and then control voltage and current to make sure the two are "in phase" so to speak.

We are now rapidly reaching the end of my knowledge.
So lets try to get pragmatic.

You are using motor and axle from Toyota. Are you using the associated controller ie VFD.
Before they were taken out of previous vehicle were they working OK AND did you see that??

Did you take them out?

Did you take good notes on how stuff was conntected and MARK your wires before disconnect?

Do you have the software to reprogram the controller?

Do you have the manuals?
Do you have the prints and drawings?

Hopefully you can say YES to all above. If so you are in good shape.
NEXT
Read the manual. RTFM again. RTFM - AGAIN.

Now if all still does not come together get real friendly with your Toyota guy - Especially the MECHANIC who works on these. He has the experience, the manuals, the tools, the software, etc etc.

Let me know how you make out. I thought about doing what you are doing -- salvaging guts out of a Toyota and sticking them in something else. Not a bad idea hot rodders been doing it for how long?

 

[itsmoked]

I think David mentioned that they tried to run the Toyota stuff but unless everything down to the seat belts, door locks, and anything else, is there, it's not going to work. The problem is that you can't just jump this, and that, signal because most things are now talking CAN Bus and need a bunch of hand shaking. Not easy. (At all!)

Like the LiIon battery in my Sony laptop. The battery choked and I tore it apart and replaced the cells but the communications where damaged somehow. Since the battery no longer talked to the computer Nah DOH! was going to happen.

David I will ask about 400Hz motor suppliers. Off hand I don't know who they are.

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[ssn596]

QUOTE
I think David mentioned that they tried to run the Toyota stuff but unless everything down to the seat belts, door locks, and anything else, is there, it's not going to work. The problem is that you can't just jump this, and that, signal because most things are now talking CAN Bus and need a bunch of hand shaking. Not easy. (At all!)
UNQUOTE

Yeah I saw where they tried to do what you said BUT that was Ford equipment. Hopefully (I doubt it now that you mention it) but maybe Toyota does not do all those dang interlocks.

OK so it is CANBUS. First my ignorance will become most blatant with a few questions.
CANBUS is communications protocol hardware and software.
AND one must wire components to it
AND program the communication addresses and all that.

THEREFORE what was put together can be disassembled
IF you have the software (this may be the brick wall on which to beat brains out)

HMMM this is food for thoght for when I do my 3phase VFD auto project. Azure does have a few interlocks. I will try to make the connections very accessible for troubleshooting ease.

Dan Bentler

 

[DKirkham]

There has to be an answer out there to make an electric vehicle. I just have to find it. I appreciate all the help you guys are giving!

Dan,

I just bought the motor straight from the dealer so I didn't get all the goodies with it. I figured I'd just plug it in...ha, ha, ha. 1000 pages of research later I am beginning to realize how complicated these babies are. I do have the manuals, etc for the Toyota, but they are liberally sprinkled with "Proprietary to Toyota" and I can't say I blame them. As for using their computer, I see that as no hope. My experience with Ford (and Toyota is WAY ahead of them) tells me it won't be possible. There is too much hand shaking communication that goes on. Literally, we had to turn off the seat belts and the seat heater in our prototype car for Ford. The days of taking parts and "Hot Rodding" them are rapidly coming to a close.

As for the mechanic who works on these for Toyota...well, he told me they were under strict orders to not touch anything that goes wrong. They just send codes back to Japan and then remove and replace as directed with the bad parts going back to Japan. Toyota is very protective of their family jewels and I can't say I blame them.

We actually never tried to run the Toyota stuff. We tried to get the Toyota motor running with industrial stuff first and failed. Then we did the Ford SEMA car and realized it was hopeless to cannibalize a Toyota.

Keith,

Thanks for doing a little rooting around for me. I truly appreciate it.

Dan,

I think keeping connections accessible is always a good idea!

I will do more digging tomorrow. Thank you all again.

David

 

[ssn596]

Dave

I do not think I would throw in towel on Toyota yet.

HOWEVER
In your shoes I think I would do some comparison shopping.
Contact Azure and enquire into AC55 motor and 445 controller. They sent me the manuals and all and I have not spent a dime yet. While Azure will not sell to me (an individual) they will sell to you (a company).

The most specific page is

http://www.azuredynamics.com/FlexibleComponent.htm

look at controllers and motors.


Dan Bentler

 

[sreid]

PM Motors for electric vehicles make a lot of sense in that no current is required to "flux the rotor." The downside is the cost of the permanent rare earth magnets.

To drive one of the motors, a three phase current drive is required. This is no different than what a VFD Vector Drive uses.

AC PM motors are usually made in smaller sizes for servo appliciations and I don't know where one would find a 50kW amplifier, there is simply no demand.

The motor can be driven using the Hall sensors in six step mode but I would think that this would generate too much acoustical noise. It is more common to use the Halls for initial start up and switching to an AC current drive using an incremental encoder for position information. If an absolute encoder or resolver is used, the Hall sensors are not required.

The amplifier and motor are used in torque mode. The accelerator asks for torque (current). Braking requests negative torque which causes regenerative braking and battery charging.

 

[itsmoked]

sreid; Are you saying he can just use a VFD vector drive?

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[DKirkham]

Dan,

The Azure motor is 233 pounds! The Toyota motor AND differential unit is only 90. There is no comparison. My ultimate goal was/is to have two of the Toyota units on board. One in the front and one in the rear. As you can see, weight rapidly becomes a heavy problem. Considering I can make a gasoline 500-800 horsepower motor that only weighs 350 pounds you can see where the problem is. (500 hp for the street and 800 if you want to see parts coming out of the side of the block as you do the 1/4 mile). I need to keep the motor as light as possible because there are still a ton (well 500 pounds) of batteries that have to be added.

sreid,

Said 500-800 horsepower motor is $25,000. The Toyota motor AND differential unit was $1200 from the Toyota dealer. I almost passed out. I can't buy a set of head for our cars for $1200. The cost of permanent magnets is high, but not nearly as high as exhaust systems, water systems, CNC porting of heads, etc.

You have hit the nail right on the head! I can't find the amplifier! I have looked and looked. Now you all know why I am posting here! From all I have read a VFD SHOULD drive the motor. All the salesmen that came by thought so too...

As for acoustical noise, I am at the point I don't care. I just want to motor to move. I will worry about noise later. In an open car like this, noise is really not an issue.

I would be interested in talking to you further. You can look me up on our website.

http://www.kirkhammotorsports.com

if you are so inclined.

Now, as a true Hot Rodder and CNC whacko--if I can't find a 50 KW amplifier, can't I make one? Hence, the original reason for my post. It is hard to explain all of this stuff writing posts so lots of stuff falls through the cracks. The road eventually lead me to using a traditional CNC industrial unit to send signals to an amplifier that would in turn drive the motor. But, as I said, the Hall Effect sensor seemed to throw sand in the works and the salesmen gave up. Any suggestions?

Keith,

I am still confused about the 400 Hz motors. From what I have read (and I readily admit I may understand everything wrong) you vary the frequency (hence variable frequency drive) to accelerate the motor--so I am not sure why a 400 Hz motor would be of any advantage in an electric car. In all my reading I have never seen a traction motor referred to by X Hz. Now, for a turbine generator that is only spinning one speed I can see the advantages of 400 Hz over 60 Hz.

From all I have read and heard, the Toyota motor should be able to be run by a VFD--the problem is getting one that will talk to a Hall Effect sensor and one that can take 50KW.

David