Guide to winding brushless motor

[carbon.nanotube.capacitor]

Hello hello!

I have a brushless motor from an electric bike gearless hub, and there is a lot of room to see where the windings go... but I'm having a terribly difficult time seeing how the windings are actually routed. My goal is to disconnect the windings from one another so that I can connect them in wye formation, as they are almost certainly in delta formation now. Actually, I want to connect the windings to relays that can switch the motor between wye and delta (my idea, but many others have thought of it too - and it's already patented - and there are motors and controllers being produced that probably do this already). So I was hoping someone might have a suggestion on where to find a guide to brushless motor winding. Right now I'm not sure if the original hall sensors will be okay in wye, or if a new set will be necessary at a different position. Also, on this motor there are 50 magnets with neighboring ones at opposite polarity, and 46-ish windings... so why the number of windings vs number of magnets? Are the windings spread out so the 1st, 2nd and 3rd are neighbors, repeated in sequence?

Yeesh - so simple; it's all right there for me to see - I'm so close but I can't figure the windings out. Help!

-Colin

 

[fechter]

There might be something posted in the Basics section. It can get complicated, especially with a large pole count. There are dozens of possible configurations.

First, try to determine if the windings are monofilar or multifilar. That is, is each winding made up of a single wire or several smaller parallel wires. It should be easy enough to find where the phase wires attach to the windings. If they are delta, then each phase wire will connect to two different winding ends. Take apart these connections and you can measure resistance to find out where the ends of each winding are.

The magnets have to be an even number of poles so they can alternate north/south. The stator poles are in 3 groups, one for each phase, usually spaced in a 1-2-3 pattern. BMC motors are different and have 3 adjacent poles would with the same winding.

If you re-combine the windings in an incorrect configuration, there will be a strong resistance to turning the motor. If the combination is valid, there will be no resistance when the phase wires are open, but strong resistance when they are shorted. Be sure to do this test before attaching a controller.

I have no clue if you need to relocate the hall sensors in order to switch from delta to wye. You may need to switch the controller from 120 to 60 degrees too, but I'm not sure.

Is that a Fisher Paykel motor?

 

[dirty_d]

you would use the same hall sensor locations for wye or delta. i really dont think it worthwhile to make it switchable between wye and delta, you can achieve the same thing by raising and lowering the voltage. as for winding it, how many magnets and how many poles are there?

 

[carbon.nanotube.capacitor]

Hey,
It's not a paykel motor, it's a hub motor for an electric scooter-style electric bike. the name stamped into the housing (the cover plates) is Bokang. I can't find anything online about bokang except a medical equipment company. The reason it's easy to see the windings is because the motor is a large diameter with a narrow profile. Ebike people like to call them high-torque, which really means large diameter, large pole count.
I don't have the motor in front of me, as I'm not at work at the ebike repair/sale shop, but I remember the windings had several wires grouped together and fed through as one cable, which went round each pole several times. The power wires attach to three points (one each) along the row of winding poles. There are long leads of cable running out of some windings, along to another further over.

I'm not interested in rewinding the motor, I just want to know how it's done so I can connect these poles correctly. And the reason I want to be able to switch to wye is so that at low speeds, such as on a hill, the motor can have more torque without overheating (my bike's controller and cables have been modified to allow much more torque through, and the motor is receiving 60 volts instead of 48). Maybe the increase in torque will be minimal, but I suspect it will be substantial, and having a low starting gear might prove good for improving efficiency in city driving. Stop and go traffic really eats battery power, especially with more current draw!

Thanks for the help thus far.

 

[carbon.nanotube.capacitor]

When I said there are cables running out of windings to another further along, I meant that there are cables running from windings to others further along - the cables don't all go to the same place, otherwise I'd suspect the motor is wye.

 

[johnrobholmes]

Just take a pic. It will help. If you see three phase bundles wired together, and three other ends exiting by their lonesome, it is Wye. If there is no internal connection and two phase bundles per exit wire, it is Delta. IF there is a ton of spliced ends, you need a pen a paper to figure it out.

You will need to select the proper order for re-termination. Otherwise the motor will want to run in reverse when you switch to Wye, but since it is sensored it just won't work. It won't give it more torque for hill climbing either, it will just make you have 2/3 the speed so the amp draw will be lower. Probably good to keep you from overheating though.

 

[carbon.nanotube.capacitor]

Confusion on one point, but first about the picture... I took pictures of the motor on the camera at work. I should be back there today sometime, just to tinker with my bike - I'll see if I can upload the pictures for you then.
But the confusing bit for me is that about the speed and torque characteristics of wye versus delta. Virtually everywhere I read people saying that delta is twice the speed of wye but at half the low rpm torque; wye is half the speed but twice the torque. Do I have this wrong? You mention no increase in torque in wye, and a reduction in speed of only a third. I'm not arguing by the way, I'm just trying to learn. And thanks for continuing to help me understand this puzzle.

 

[johnrobholmes]

The equation is Wye speed x 1.76 + delta speed. So a 1000kv motor terminated Wye will be 1760kv terminated Delta. Not quite half speed, not quite 2/3. 76% faster to be exact.

The torque of the motor doesn't change, but the torque constant does (Kt). It is directly relational to the Kv with the formula Kt = 1352.4 / Kv. The torque constant is the torque produced per amp. A slower motor will produce more torque per amp, but draw less amperage on a given voltage and geardown. Thus, if the setup is not changed except for the termination of the motor, the Wye termination will have less power. This could be good for going up hills if your hills are too much work (and thus heat) for the motor while terminated Delta. When you are on flats or downhills you could switch to Delta to get more speed.

About the only advantage to a motor that can switch from Y to D termination is that the motor controller can run cooler. When going up a hill that is too much for a D termination, the throttle must be scaled back and the duty cycle is less than 100%. This imparts switching losses into the controller. If you could go up the hill at full throttle (100% duty) with a Y termination you would at least avoid most switching losses in the controller.