[Bowser330]

Thanks for the reply...so i guess i was wrong about the RC motors being weak sauce, looks like they can and are being built pretty powerful.. man i wish you weren't literally across the continent from me, It would be cool to have your help on building a large burshless motor...

I guess we will need to keep hitting the boards about the controllers..there is bound to be someone who knows how to take this... (see link)

http://www.robotpower.com/products/osmc_info.html

and make it handle the parameters we need....for our motor...

[Madmac]

The link is to a PWM output stage to drive a DC brushed motor. It could be used in other applications but better options are out there.

I think several members who post are working on their own AC induction controllers. The permanent magnet brushless motors are effectively AC synchronous motors and the same control hardware can be used.

If you do a search for BLDC motor control there is plenty of technical information and application notes. Most is related to small devices though.

Madmac

[Gravitic Anomaly]

The model airplane world of motors is an odd mix. In some ways it is on the cutting edge and in others it is in the stone ages. However almost any hobbist can and do make their own motors. Me being one of them.
You will find some good basic recipies for the motors and and equal amount of dis-information.
A motor large enough to power an automobile could be no more difficult to build.
Rather than trying to upscale an existing design you are better off designing from scratch. I think the simplest motor would be a PM brushless motor built onto a manual transmision/transaxle. The stator assembly would be bolted to the transmission housing and the rotor would be attached to the transmission input spline. This way the builder does not need to make housings, shafts, or add bearings. All the pecision machining is already done. The motors airgap would be relativly large for this type of motor. This is to make the assembly more tolerant of misalignment, more resistent to demagnetization, and run cooler.
Builders will have to be warned to closely follow the assembly instructions. If handled carelessly the rotor will quite happily destroy bearings or chop off fingers when placed over the stator.
I first used this type of motor in the early '90's on a turbine powered motor/generator project that was jointly funded by the DOE and a major automaker. It was a test bed for a bi-directional inverter concept.
A controller for a PM brushless motor is easier to build than an AC vector drive. A industrial vector drive can be used to run a PM brushless motor if it has the right software. Many of the higher end ones have this built into them.
I can design a suitable drive for the motor. We need an experianced board designer and electronics prototype guy to do the hardware.
I have too little time and too many projects to build prototypes myself. I will work with someone willing to put the time and money into building a prototype motor. Including sources for all of the parts. For strictly selfish reasons someone converting a Fiero would be preferred.
The motor cross section I previously posted is of a radial flux, outer rotor hub motor I did a few years ago for an unmanned vehicle. It was roughly two foot in diameter.

[judebert]

Originally Posted by 3dplane

Bowser!
I know your question is not directed at me but the way I see it is the easy and cheap part of this is to actually build the motor. My question is : what are we going to use for controller and how much does it cost? I know I can't build one!

Originally Posted by Gravitic Anomaly

A controller for a PM brushless motor is easier to build than an AC vector drive. A industrial vector drive can be used to run a PM brushless motor if it has the right software. Many of the higher end ones have this built into them.

I would like to make my part in this project clear. I'm a programmer, by trade and inclination. I have experience in many computer languages, almost entirely C or higher-level, but a very little embedded and real-time experience.

I am formally offering to donate my programming time for any controller designed for use in a highway-capable electric vehicle, under one condition: anyone must be allowed to use and modify the resulting code. (Like the GPL.)

Originally Posted by Gravitic Anomaly

I can design a suitable drive for the motor. We need an experianced board designer and electronics prototype guy to do the hardware.

You build the motor, someone else builds the controller, and I'll program it. Where are our electronics guys?

Originally Posted by Gravitic Anomaly

Rather than trying to upscale an existing design you are better off designing from scratch.

The other controller designers on the forum (or is it the EVDL?) seem to think differently. I don't know enough about it to say.

Originally Posted by Gravitic Anomaly

I think the simplest motor would be a PM brushless motor built onto a manual transmision/transaxle. The stator assembly would be bolted to the transmission housing and the rotor would be attached to the transmission input spline. This way the builder does not need to make housings, shafts, or add bearings. All the pecision machining is already done.

Interesting idea. I kinda like it. It doesn't take advantage of the previously stated advantage of brushless DC (good torque at low RPM), but it's definitely the elegant solution.

I'll restate my ACIM preference, while I'm here. For my own selfish reasons, I would prefer a motor as maintenance free as possible. That means no brushes. I understand ACIM better than brushless DC, but only barely. I'll still be happy to program a brushless DC controller, too.

I'd also like regen, and a smart charger built into the inverter/controller.

If we were doing requirements analysis like we do on software projects, I'd want:

1. The drive system shall be maintenance free after installation.
2. The drive system shall operate the vehicle from stationary to top speed without requiring the driver to shift.
3. The drive system shall recharge the energy storage system using energy recaptured during braking.
4. The drive system shall recharge the energy storage system using energy provided from an electrical outlet.

That's what I'm willing to work for. Design me a system that fulfills those requirements, and I'll be staying up late to program it for you.

[Gravitic Anomaly]

If the battery pack and controller can provide the current the motor will be able to provide peak torque at 0 rpm. Smoke 'em if you've got 'em.
I would add to the list of desires:
The controller/charging system shall be capable of accepting on-board generators for hybrid operation.

My work will be open source, free to use for NON-Commercial purposes. Including the design, instructions, drawings, & schematics. Offered as-is not warrented. Suggested sources of components are not affiliated with me in anyway financial or otherwise. Commercial sale of components, kits, drawings, circuit boards, ect. can only be arranged through the principles involved with the development. The selling of the open source information on Ebay® is strictly prohibited.

- GA

[Madmac]

I am in the early development stage of an expanded version of the Circuit Cellar AC motor driver. I have added a two phase boost circuit to allow control of regeneration voltage to battery charging, PWM driven load resistors to loose regeneration voltage when batteries are fully charged and a 3 phase bridge driver rated at 600 volts 400 Amps.
My interest is in driving a Ford/Siemens AC motor that I bought surplus. I will be doing a 4 layer PCB with the drive circuits and an interface to a BMS, I am also working on, and a USB interface to a car PC for data logging and monitoring.

If some one wants to do the firm ware to support a BLDC motor I am willing to supply either the gerbers or boards (at cost). This would be under the same conditions as the post above....no commercial use. Supervisor is a Silicon Labs 8051 derivative and DSP is a Microchip dsPIC30F6010. PM me if you are interested.

The current design, not yet finalised, may need a few additions to support BLDC motors depending on position and current sensor arrangements.

Madmac

[CNCRouterman]

I have a CNC router, which means I can machine most materials, but not rock, steel or glass. I have a customer with a very high end CNC punch, and that might be even better for some components. I would like to get on this wagon and contribute to the project. Personally, I am still a bit skittish about hanging the motor in the wheel, but an aero engineer friend of mine (who also has an automotive engineering background) explained a lot of the advantages as well as addressing my concern about unsprung weight. That said, I would still probably lean towards mounting the inboard and using CVJs and half shafts.

[manic_monkey]

Hi all. I found this site a couple of weeks ago and am really fired by the idea of open source motor building. Seems like a great way to share knowledge and work for a common goal

Ive still got a lot to learn about motors, so please forgive me if some of this is a bit simple.

theres a lot of talk about going for a bldc design, but most of the focus seems to be on outrunners. would an inrunner not be a simpler design? i understand they provide less torque, but at this sort of power and size, it would surely still be far more than a car engine. also, wouldnt an inrunner require less magnets? one thing which concerns me is the magnets that will be require to make an effective bldc. neodymium magnets of the required size would surely be very expensive, but more importantly, they would be very very difficult (and possibly dangerous) to work with. ive seen 50x50x25 (mm) magnets that produce 90+kg of force, so i imagine the magnets for a 40kw plus bldc would be similar.

I understand that going BLDC means a less complex controller is required, but it seem that by doing so, the complexity is being passed onto the manufacturing side. (making it very difficult for a homebrew enthusiast to make it)

[3dplane]

Hey manic!
One of the things makes outrunners easy to homebrew is that the magnets are not going anywhere in the bell.Centrifugal force just helps them stay put.In an inrunner one would have to use maybe kevlar wrap or some kind of pretty sturdy method to make sure the mags not going to fling off.Rewinding an outrunner is also a breeze compared to an inrunner. The stator arms on the outrunner are like open fingers(on a hand) making the task of wrapping wire on it easy compared to imagine a ring with sharp fingers pointing to the inside of the circle and wind wire on them(neatly) for the inrunner.
So no an outrunner is a heck of a lot easier to construct on the diy level then an inrunner. Btw welcome to the forum. Barna

[manic_monkey]

ahh yes, i see what you mean. even at relatively low rpm, the weight of the magnets would put an inrunner under a lot of strain.

while your here, could i pick your brain for a second on outrunnners / BLDC?

does the thickness of the bell material have much of an effect on motor efficiency? (larger iron mass, magnetic fields, et al)

should the core material be made from soft iron, or is mild steel enough?

back to the point of hub motors. if two wheels are driven seperately, the electronics will have to function as an active differential (speed one up, slow the other down) to ensure you go round the corner. has anyone given much thought to how the controller could sense the amount of difference needed?