I don't have any first hand experience with demagnetizing magnets but I seem to remember that the field strength to demagnetize an Nd magnet decreases more or less linearly until it's 0 at like 300C or something.

I agree though that you could turn the current up quite a bit above what I'm running. I think you would find a real limit somewhere around 120-150A, but how do you know until something goes wrong?

As for overall power, the voltage pretty much determines it. It's 50Kw @650V, so at 390V (about what my pack sags to under heavy load), that's 30Kw. I've seen 20-25 but as the car speeds up I have to back off the accelerator to avoid control issues. I found it easier to just limit the power at all speeds.

Overall, I totally understand where you're coming from. Two motors and transmissions for $1000 sounds great, but they're just not easy to control. The Curtis won't do it, you'd want Tritium or Rinehart, or one of the really expensive ones. And then, it's not cheap anymore and you will want more power for your money. To do what you want, you need to be able to hack into the power modules and use their control code. Obviously I don't have the knowledge to do this or I wouldn't have bought a controller. I do plan to find some sort of project for the extra IGBTs I now have lying around though...

 

I don't have any first hand experience with demagnetizing magnets but I seem to remember that the field strength to demagnetize an Nd magnet decreases more or less linearly until it's 0 at like 300C or something.

I agree though that you could turn the current up quite a bit above what I'm running. I think you would find a real limit somewhere around 120-150A, but how do you know until something goes wrong?

As for overall power, the voltage pretty much determines it. It's 50Kw @650V, so at 390V (about what my pack sags to under heavy load), that's 30Kw. I've seen 20-25 but as the car speeds up I have to back off the accelerator to avoid control issues. I found it easier to just limit the power at all speeds.

Overall, I totally understand where you're coming from. Two motors and transmissions for $1000 sounds great, but they're just not easy to control. The Curtis won't do it, you'd want Tritium or Rinehart, or one of the really expensive ones. And then, it's not cheap anymore and you will want more power for your money. To do what you want, you need to be able to hack into the power modules and use their control code. Obviously I don't have the knowledge to do this or I wouldn't have bought a controller. I do plan to find some sort of project for the extra IGBTs I now have lying around though...

If the KV is that low I agree with you. If I was you I would scope it while spinging it by hand and work the numbers to find the actual KV out. As for controllers well I have built one for 3 phase BLDC and Im using microchip right now but I sugest anyone looking at a controller project look to Texas Instraments they have a few kits with partial (and maybe compleat) code to get you going. They show videos of Instaspin which they can spin almost any bldc in 30 seconds or less!!!
As for this motor I think I want to find two and start with one get it working then add the other to the front wheels. I will look at rewinding it. But for now I don't have the cash or time.
As for demagnatizing heat alone will kill a magnet. We uasaly try to find the temp rating of the mags and usualy try to use mags with 180 deg C temp rating. Which meens they can be heated anything below that without damage. Here is a thread I started asking about this on endless sphere http://endless-sphere.com/forums/viewtopic.php?f=30&t=44831
Cheers.
Arlin.

 

Here's one demag event...

http://www.diyelectriccar.com/forums/showthread.php/ripperton-electric-track-bike-41173p21.html

He says they overheated, but he also said he put like 500A into the motor to create the heat, I don't see how you determine which one killed the magnets. I bet the magnets would have been fine if they were subjected to only the temperature and not the field (and vice versa).

 

I don't think the peak power comes at 650V. The Toyota highbrids have a pretty low amp capable pack and only 210V or so. They jack the voltage up occasionally but that would drop the amps proportionally so no power gain would come from it. I'm guessing the volt boost is used to keep the motor from becoming a generator(drag) at highway speeds and to allow regen at lower speeds. I still think the real deal to be had is using the factory inverter. One of the plug in conversion companies has Highlanders staying in pure EV mode up to 70mph so it can be done.

 

I know heat will kill a magnet, but heat and field will kill it at a lower temperature. You could demagnetize them at room temp with a strong enough coil. http://www.mcmaster.com/#demagnetizers/=jzpd5y (I know those aren't actually strong enough for a Nd magnet, but same principle)

I have logs of output V and rpm, so I can get you the kV if you want. I think I've seen up to about 900V though, certainly well over the 410 of my pack. Toyota achieves peak power by using MG2 as a generator to supplement the battery power. I think peak battery power is like 20Kw or something much less than the motor rating. Maybe that's for the Prius and it's a little higher in the Highlander, but not much.

This isn't a BLDC motor, which would be easier to control. If it were, there wouldn't be any power available over about 20 mph. IPMs rely heavily on reluctance torque and have a similar torque curve to an induction motor (and can actually have a larger constant power region if designed for such). It is, however, synchronous, so up to 20mph, the TI controller would probably work fine, the problem is when you have to advance the drive signal - it takes more knowledge than I have to get max torque per amp above base speed without resorting to trial and error. Also, they're more susceptible to noisy position signals since losing track of motor position by even a few degrees makes for a jerky ride. I think this is the main problem I'm having.

 

Ok, I just read the ES thread. Apologies for dismissing your ideas so quickly, it now seems likely that while demagnetization is possible with field alone, I probably won't be able to create such a strong field in this type of motor. I still have concern about 1) winding and feed-thru resistive heating 2) half-shaft mechanical strength and 3) control issues becoming more severe due to high power levels lower in the rpm range.

I'll see if I can get the car back together to work out my controller kinks on the MGR then dump a few more amps in. As I said though, this wasn't the only issue that prompted me to start the motor/battery upgrade.

 

Ryan

Do you have any pictures of your motor's wiring, ideally showing the coil linkage joints if possible. Thinking if you can reconnect them in parallel you will lower the KV of the motor and make life easier for your Tritium.

Take a look in the 'rewinding an AC motor' thread.

I will check out your thread, thanks for linking that.

 

Unbelievable! Forum sent me to some stupid survey!

I was saying...

Ryan

Have you got any pictures of your motor please, ideally showing the wiring inside, and if possible the coil connections. I'd like ot see fi the coils can be reconnected in parallel to reduce the KV. This would help out your Tritium by the way, and if enough connections can be changed it may even allow use of a Curtis controller...

See the 'rewind an ac motor' thread for my inspiration.

Thanks for posting your build thread link by the way, I will check it out!

 

Unbelievable! Forum sent me to some stupid survey!

I was saying...

Ryan

Have you got any pictures of your motor please, ideally showing the wiring inside, and if possible the coil connections. I'd like ot see fi the coils can be reconnected in parallel to reduce the KV. This would help out your Tritium by the way, and if enough connections can be changed it may even allow use of a Curtis controller...

How would this make it easier for the controller???
If you double the rpm per volt you will have 4x less inductance. Ryan you don't know the inductance by chance? I would bet its a few hundred uH so you can get away with it.

 

After all that it did post my first post... Figures!

 

Apologies, I meant a lower kV would give a wider choice of controller. Please forgive me, my knowledge of controller parameters is limited.

If the wiring is unchanged and we've simply connected the coils differently wouldn't it have the same net effect on the motor's performance as long as the controller can handle the higher inductance?

 

Higher inductance is what a controller likes. The inductance is what controls how long the amps take to build during fet or IGBT on time higher inductance meens less torque ripple!

If you want to spin the motor with less voltage to the same rpm you will have a higher rpm/volt (kv) so the inductance will be lower but from what I have seen most controllers are fine with motor inductance ~100uH or more. My controller is working with 8uH but I learnt to build my own just for the fact non of the cheep controllers could do it.

It is possible to build up a cheep controller to run the highlander diff if it was rewound but it will depend on the inductance or L/R but resistance is not a big deal and you simply want the resistance as low as possible to make the motor more efficent but the inductance is another story. I will see how efficient my system is with the 8uH motor I have but there is a chance I need to rewind for more inductance.

 

here is a nugget I realized while modding my PHEV Prius... the HV system has the needed voltage to run a rear motor.
whats even more interesting, the non-PHEV Prius has the AWD system with the tiny motor but they didnt make an AWD PHEV version. I still haven't figured out why!

that to say, yes, I brought this back from the dead but only because now technology has advanced and no one is talking about this powering electric motor jazz hands.

 

There's many other threads discussing the Highlander and the MGR. It was generally found to be unsuitable and efforts for an affordable diff-based motor have gone to the Outlander instead.

Voltage might have been the only issue though. Not sure, I revive my interest in it a couple times a year.

 

ohhh... Ive yet t find anything on this outlander motor... have a link to point me in the right direction?