What to do with this motor

[mistercrash]

I had trouble with the motor on my scooter recently and after looking at different options, I decided to just go the easy route for now and replace it with the same thing from the dealer. So I have this extra motor lying around in pieces with two magnets that are damaged on it. I'm thinking that it's a good opportunity to learn about this motor and play with it a bit. So what can I do to make this 500 watt motor a little more than what it is right now. Any ideas other then replacing the broken magnets?
I have a few questions here, there's 46 magnets on the rotor/rim and 17 poles per phase on the stator. Isn't the amount of magnets suppose to be the same as the number of poles? Or are they just there to create some magnetic field and it doesn't really matter how many there are? I also noticed that the poles of the magnets are alternating, how is that important? The magnets are 45 mm long by 13.8 mm wide by 3 mm thick. If I was to replace all the magnets with 50 mm long ones, would that make a stronger magnetic field and help the motor? John in CR suggested ventilating like he did in here http://endless-sphere.com/forums/viewtopic.php?f=2&t=39773. This would be a good start but is there anything else that can be done to make this motor more powerfull or efficient or both?

Thanks and here's a few pics and a drawing with dimensions of this project motor.

The one way sprocket side

The brake disk side

Without the covers

The laminations, the caliper is spread 1/4''. The stator is 45 mm wide

The hall sensors

And a drawing with dimensions

 

[mistercrash]

I started doing some stuff to the end caps. 12 holes on the brake disk side around the middle as air intakes and 20 smaller holes on the wire side close to the outer edge for air exhausts. To help the air go inside the intake holes, I thought of fabricating a fan out of a piece of tin that would be behind the brake disk. Anyone here that know something about aerodynamics can tell me if this would work well? I'm thinking it might help a little but the wheel doesn't turn that fast so I'm not sure if it's worth the trouble.

EDIT: That fan won't work at all, I didn't think of the brake caliper. I'll have to check if there's a way to make it smaller to clear the caliper but still big enough to move some air.

 

[mistercrash]

I finished making a small fan to put behind the brake disk. It moves air Not a lot of it but when I make the end cap turn about as fast as it would turn with the scooter going at 45 km/h, I can feel a flow of air twirling around and going on the stator. Some air is better than no air. I received the magnets I ordered to replace the two that were damaged. I couldn't find magnets that were of the exact dimension of 1.75'' X 0.5'' X 0.125''. What I got is 2'' X 0.5'' X 0.125''. Other than being longer they are also a little thinner and a tiny bit thicker but they fit. I'm just wondering if having two magnets that are longer than the others is going to mess up the motor. Also, the magnetism of those new ones seems to be stronger than the old ones.

Should I just order more and replace all the old magnets? Am I wasting my time working on this motor? :lol:

 

[dnmun]

thinner but thicker, but they fit? if you already bot the replacement it seems it is wasted effort but you may make it work.

you need to worry about the lengths because the ends may hit a rib on the cover. each magnet is reversed in polarity from its immediate neighbor so you have to account for that too. how will you glue it in place and how long with the other magnets remain intact if these others broke already?

if you still have the old magnets then you could glue the pieces back into place where they were originally located, and the motor force would be about the same.

the difference in magnet strength should be a non issue since it only affects one phase current magnetic field at a time. the others are interacting with the normal magnets so the unbalanced forces should be unseen. jmho.

 

[mistercrash]

thinner but thicker, but they fit?

Sorry, wrong choice of words. English is not my first language. I meant they are longer, a little less wide and a tiny bit thicker. The old magnets are 45 mm long by 13.8 mm wide by 3 mm thick. The new ones are 50 mm long by 12.7 mm wide by 3.175 mm thick. Despite the fact that they are 5 mm longer, the new magnets fit well and the end caps don't touch them. Some high temp JB Weld will be used to glue the magnets. I will not put epoxy underneath the magnets, I'm thinking that magnetic and centrifugal force keeps them in place on the rotor, I will smear epoxy at the ends on either side so they don't move side to side. I'm afraid that by putting epoxy underneath, there will be too much glue which would raise the magnets and hit the stator. I don't want to push too hard on the magnets to squeeze the excess epoxy out and risk cracking the magnets so I won't put glue underneath.

Years ago I used to race RC cars and it was common for racers to zap the magnets of their motors to increase their magnetic field which made the motor more powerful.
Since I noticed that these new magnets are more powerful then the old ones, I thought that putting more powerful magnets all around would increase the magnetic field thus making the motor more powerful. Also there would be more magnets because of the different dimension of the new ones. The rotor would end up with 50 magnets instead of 46.

Thanks Dnmun

 

[dnmun]

miles knows this stuff. but i would think that if you changed the number of magnets, even though you do use an even number, that the physical distance in radians around the hub would be different and the full 360 degree rotation of the magnetic field for each two magnets at each hall sensor would have to be the same distance where they are located in the stator. so the total distance in radians between the two end hall sensors and the two magnets would have to be the same, imo. 120 degrees for each phase times 3 phases = 2 magnets with opposing fields so the surrounding magnetic field goes through the full 360 degree cycle.

if you wanna use these magnets then maybe consider how you could interpolate spacers and then see if it works to switch the hall sensors at the right time if you stick with the original 46 pieces. jmho.

 

[mistercrash]

You answered a question I had before about the number of magnets so thank you for that. So you made me understand that there are rules to follow that determine the number of magnets to be used. I'm a bit lost on the rest of what you said. Do you mean that the two end hall sensors have to be over two magnets with opposing fields at all times?

 

[mistercrash]

Work continues to bring this motor back to life and learn a couple things. I took the bearings out, they are actually pretty good bearings but they felt a little dry so I took out the seals, cleaned them with WD40 and repacked them with bearing grease and popped the seals back on. The magnets are positionned on the rotor, I still have to mix some JB Weld and glue them. Here's a pic of the two new magnets.

I put the stator and end caps back on to see how it goes. Smooth as when it was new, the thicker magnets don't touch the poles at all. I drilled the hole in the shaft a little bigger and now I can push three 10 gauge Turnigy wires and the hall sensor wires which are wrapped in a thick heat shrink through the axle. I also Dremeled the holes in the end caps to help scoop the air in and out.

The small fan I made for the intake holes, just three blades to move some air in.

 

[dnmun]

3 hall sensors are included in the two magnet's fields. that is where the 120 degrees comes from 3 sensors over a 360 degree change in the magnetic field means 120 degrees/sensor

how did you verify that the magnetic field of the magnets you inserted were reversed from the adjacent magnets?

 

[mistercrash]

I used another earth magnet to feel the polarity on each magnet. One magnet pushes, the next one pulls, the next one pushes.....

 

[dnmun]

trick question. if the poles were aligned side by side, the magnet would flip out when you put it in between the other two. i think you could use the JB WELD under the magnet too, but it will never come out again. it should squish out very thin. you will need to make a wooden stick or brace to hold it in place pushed out against the hub while it sets up or it will slide around.

BOL, maybe you will be riding on this one yet.

 

[mistercrash]

I must be thick because I'm not getting what your saying again the magnets are polarized on their thickness, not their length or width. So when I place a magnet carefully in its slot, it is attracted to the metal rim of the rotor and stays there quite firmly. It is not pushed away by the adjacent magnets. Or if it is, the force exerted by the magnet towards the metal rim far exceeds the pushing force of the adjacent magnets. That is why I thought of just putting epoxy on the edges of the magnets to keep them from sliding side to side. Some epoxy will also find its way in the crevasses between each magnet so it should hold them firmly in place.

I'm also planning on water proofing the interior of the motor since I put holes all over the end caps. I know someone in here has done it, I have to look it up. So the product I'll use to do that is also going to help hold the magnets in place.

 

[Kirk]

For coating the inside any dialectric Urathane Varnish will do the trick. In my motor the rotor and stator have two coats of CRC Red Electric Urathane Varnish.

I'll get the part number off of the spray can after the wife moves her truck out of the garage this morning and post it.

 

[mistercrash]

Thanks Kirk, right now my problem lies with finding a place to get this stuff or an online store that will accept to ship me a can here in Canada. I went shopping today for Varsol to clean every part, some flat black paint for the end caps, new high quality SS M5 screws for the end caps, some heavy duty double wall heat shrink with the glue in it and 5 small spools of 26 gauge wire to fix the frayed hall sensor wires. I'll just replace them all with new ones. I already have enough 10 gauge Turnigy wire for the phases so all I need is to find this darn dielectric urethane insulating varnish.

 

[dnmun]

you don't have to go to such lengths as using insulating varnish if you just wanna keep it dry. i think almost any spray varnish would accomplish what you need to do. the windings are already insulated, it is only the legs of the hall sensors that need to be waterproofed to keep them from shorting when the water puddles on them. i used silicone on the hall sensor leads. i think the spray paint would be adequate if it covered the leads all around. jmho, but you can expect the experts to trash me and my lack of knowledge here.!!!!!!!!!!!!!!!

 

[mistercrash]

I ride this scooter everyday rain or shine. Heavy rain, searing heat, bitter cold, snow and slush. So I really want something to keep the inside from rusting. But what you say is true though, I know of a very good clear varnish that does a real good job at keeping things from rusting. I used it many times on other stuff and it's kept the metal which sits outside free of rust for a few years now. I might just grab a can of that and spray the insides of the motor.
Some more work done today, the stator has brand new wires all around with 5.5 mm bullet connectors I like to use.

The end caps are painted flat black. I read on the internet that it's the best color for heat dissipation so it must be true.

The magnets are glued with JB Weld

More on this JB Weld thing. I don't know if anyone has ever glued magnets with JB Weld. I can only imagine the mess one can make trying to put JB Weld first and the magnets second. Remember I wanted to just put some epoxy on the end of the magnets? Well it looks like there's JB Weld in every nook and cranny between and underneath every magnet. The magnets just suck in the epoxy spreading it everywhere. Proof that JB Weld is metal. I started with a metal rod to try to spread the JB Weld. I switched to a plastic stick. Then it took a lot of elbow grease to wipe the excess off with paper towels. But the only way these magnets are going to move now is to chisel them off.

 

[mistercrash]

First problem to arise, I put the motor back together this morning just to see if it works. After installing it on the scooter, the only thing it does it lock up when I apply the throttle. It seems to lock in specific areas. I move the wheel an inch and the motor will back up half an inch and lock up. I move the wheel an inch in the opposite direction and the motor will move forward half an inch and lock up in the same spot. I also noticed that the phase wires become warm to the touch. I hope it's a simple problem with a simple solution. Help please and thank you.

 

[mistercrash]

I inspected every millimeter of every wire to make sure there was no short anywhere and checked that every wire hadn't been severed somewhere. Everything checks out ok but the motor still locks up. I'm afraid I failed in repairing this motor. Could the metal in the JB Weld mess up the magnetic field? :|

 

[Punx0r]

Is the motor locked when you try to rotate it by hand?

 

[hydro-one]

Sounds to me like you messed up a hall sensor, can you measure the hall sensor signals, with respect to ground , with a multimeter on dc volts. as you spin the motor should toggle between 0 and 5v or so. . hint. the halls harness has to be powered by 5v from the controller.

 

[mistercrash]

The motor spins free when I turn it by hand, it locks when I apply throttle.

 

[dnmun]

you need to plug in the hall sensors and power up the controller.

disconnect the phase leads, measure all three of the hall sensor signal leads, BGY at the same time before rotating the hub.

rotate the hub by 1/3 of the phase which is 1/3 of 1/23 or about 6 degrees. the hall sensors should change state, move another 6 degrees, the hall sensor should change state, do that for the full hub if nothing shows up right away. the new magnets may not be able to switch the hall sensors. unlikely but maybe.

you have to have the controller powered up and the halls sensor plug, the 5 pin plug, connected. measure the voltages by putting the voltmeter probe into the back end of the plug so it contacts the pin inside the plug while still active.

 

[mistercrash]

Please be patient because I'm not getting it.

ok so I have the motor propped up in a large vise and the hall sensor plug is connected to the controller on the scooter. The phase wires are disconnected. I turn the power on the controller. I have my multimeter in sight, so what am I looking for? I have a red and a black probe from the multimeter, how do I check the BGY hall sensor wires at the same time? What state change from the hall sensors am I looking for? A change in voltage? How can I tell if what I see is ok or normal? What should I see if everything is working the way it should?

 

[mistercrash]

I think I might have figured it out. I connected the hall sensor plug to my scooter, turn the power to the controller on, phase wires disconnected. Then I put the negative probe of the multimeter to the black hall sensor wire and the positive probe of the multimeter to the yellow hall sensor wire and turned the hub and watched what the multimeter read. Then I did the same to the green and blue hall sensor wires. If it's the way to do it then all my hall sensors are toast because their state is not changing. Do I have to replace to hall sensors with the same ones I have now or can I get better ones?

 

[dnmun]

nope, i doubt if they are dead unless you accidentally shorted them to the phase wires.

verify the 5V on the red wire from the black wire, ground. on both sides of the plug.

if the hall sensor IC doesn't have power from the 5V then it won't work. that has to reach the inside of the hub. so you may have to open the hub again.

but first go back and double check that all the pins in the 5 pin plug are seated, make sure you have good contact inside the plug with your meter probe, maybe even then check continuity from one side of the plug to the other, anything to make them work without opening the hub again.

edit: power off the controller while checking continuity.