Rewiring a fried Crystalyte Hub Motor?

Hello Again,

If you don't want to read my history, then just read the last paragraph to see what I am contemplating doing.

First History:
I have not been on the forum for a bit. I finally had put my bike back in operation towards the end of April after multiple issues including a blown controller. Was able to get another 600 miles on it, by the middle of June until driving home late one night I hit the brakes while coming to a stop light and I heard and felt the same noise and vibration I had when the controller blew. I immediately unplugged the controller from the motor but found this did not make a difference, the motor would not roll easily, had the feeling of a stationary bike on the hardest setting. This was about 16 miles into my 20 mile ride home from work at around 11:30 PM and made an already long day much longer. I parked the bike and did not look at it until last week due to other priorities and just overall tired of working on the bike and money put into it.

I have now had enough time to cool off and am looking at my options for getting this going again, but stronger than it was before. I had purchased from Lyen what he had called a Crystalyte 5303 v2 /Phoenix Racer II motor. At the time he said it was the equivalent of an H3540 Motor. This is a rear sensored motor that could go about 36 MPH at 48V. I have a headway headquarters pack that I had put together that was a 2P16S setup 48V Nominal 30 Ah capacity. This was able to get me the 20 miles to work, charge it at work and then back in the evening, averaging high 20 MPH. I would have an elevation change of almost 1000 ft with a 3% grade maximum, but this would get the motor fairly warm by the time I made it to work. I ended up changing my block time from .2 to 2 sec in the controller settings when I got it working again in April since it would give me additional hill climbing power that really seemed to make a big difference in the overall bike operation. I know what many of you will say, "you put too much power into the motor" I agree and take full blame for that, but now I want to get it operational again and would like to have it handle these wattages 1800 to 2200W for the hour long ride into and back from work reliably.

Possible Repair Plans:
I looked into replacement motors from HPC and they recommended a H4080 or H4065 motor with the side cover drilled open, but were talking over $600.00. I contacted Lyen regarding this to see what he would recommend, and I never heard back from him, probably because he is tired of dealing with me.

I am now debating about trying to rewire the motor I have myself using an upgraded magnet wire from a company called Superior Essex. They have a product called Alex that can reliably run up to 240 deg C. http://www.eis-inc.com/Files/pdf/su...loads/superioressex/Allex_240_Magnet_Wire.pdf I would imagine this is much better than the wire that comes in the stock Crystalyte motors. I think the wire gauge that is currently in the motor is either 23 or 24 AWG, but I am not sure about this, need to more carefully measure the wire that is currently in the motor. So my possible plan would be to rewire with the higher temp magnet wire, drill large holes in the side covers that line up with the coils and install some type of blower motor that will force air across the coils to keep them as cool as possible. Has anyone here rewired one of the crystallite hub motors? IF so, what wire did you use to do this? Any recommendations or things to watch out for? I have searched for links for someone that has done this on here but could not find much. Placed a couple pictures below to show you what happens when you continously place 1500 to 2000W into a motor of this size for an hour at a time.You can even see how burnt the normally white fiberglass insulation had gotten, it is now black

Any ideas or recommendations other than, "less power" would be greatly appreciated.

Thanks again,

Ed

If you look around for "rewound" or "rewind" in posts by people like Methods, Liveforphysics, etc., you should find a few complete rewinds of these hubs, discussing winding methods, wire, etc., and some results.

I have a feeling that unless you do some pretty good cooling of that motor, and make sure to wind it for the speed you want to run it at on those hills, it's not going to fare too much better the second time.

A few interrelated things with what you want to do are the winding scheme of the motor to give it a kV which lets it spin at the speed you need it to with the lowest current at the highest load, the wheel size (which changes what kV you need the motor to be), the phase current the controller puts into the motor, etc. On the latter, a typical ebike controller does not even know what the phase current is. Some do, most don't. If you use one that does, you can limit the current there instead of batteyr current, and it will much more effectively / efficiently run the motor at higher loads, by limiting the phase current to the max that the motor cna handle without saturation. (which you would have to experiment with, probably).

There are some technical discussion threads about these things, partly in the technical reference section and partly in build threads and idea discussion threads, mostly in the motor technology section.

amberwolf said:

If you look around for "rewound" or "rewind" in posts by people like Methods, Liveforphysics, etc., you should find a few complete rewinds of these hubs, discussing winding methods, wire, etc., and some results.

I have a feeling that unless you do some pretty good cooling of that motor, and make sure to wind it for the speed you want to run it at on those hills, it's not going to fare too much better the second time.

A few interrelated things with what you want to do are the winding scheme of the motor to give it a kV which lets it spin at the speed you need it to with the lowest current at the highest load, the wheel size (which changes what kV you need the motor to be), the phase current the controller puts into the motor, etc. On the latter, a typical ebike controller does not even know what the phase current is. Some do, most don't. If you use one that does, you can limit the current there instead of batteyr current, and it will much more effectively / efficiently run the motor at higher loads, by limiting the phase current to the max that the motor cna handle without saturation. (which you would have to experiment with, probably).

There are some technical discussion threads about these things, partly in the technical reference section and partly in build threads and idea discussion threads, mostly in the motor technology section.

Click to expand...

Hi Amberwolf,

Thanks for the quick reply. I searched rewind instead of rewire and it seems like more posts show up. Thanks for that info. I was thinking if a can create a fan shroud that blows air evenly across the motor windings this would increase the cooling I had, "which was almost non-exisitant" and keep the ultimate temperature of the motor cooler throughout the hour ride. I can also turn the block timing down a bit so that I am not pushing this high wattages for as long of a time. The problem with the block timing is it works when you first accelerate, but if you are doing 25 mph and then let off the throttle and it WOT again the block time starts all over again so you can have very high wattage often especially when stop and go. Additionally if I use the the higher temp Magnet wire it should be able to withstand more of a beating than the standard wire that Crystalyte puts in their motors. There insulation is probably the lower temp class around 105C whereas the Alex wire is supposedly capable of 240C. This has to help a little. I have not done a lot of searching but I was thinking of using a blower motor like the this one http://www.grainger.com/product/DAYTON-Axial-Blower-5JME6 that is rated at 48V and only uses 5.8W of power with would be negligible on compared to the overall pack capacity. I could even put a handle bar switch so if I am not climbing hills just leave it off. Of course the final question is once I purchase a blower motor, the wire and all the little small miscellaneous pieces that you always need, will it be much of a savings in the end? Not sure, but for the wire I have estimated about $150 and this blower is $84, but I am sure I can find less expensive ones with the same specs. So possibly for $250 and a lot of time, which is sort of free I could get it working again and hopefully more robust than before.

Not sure what you meant by the winding scheme, giving it a kv, I assume you meant a way to switch the way the windings are configured?

I will keep this post updated as I make my final decision. My other issue is I don't want to let my batteries sit all winter without any use otherwise I will end up wasting a lot more than just $250. I need to continue reading other post to see if I can come up with a better plan, that doesn't cost $800by purchasing a Crystalyte Crown or H4080 motor. I would love to do this, but my wife would not understand.

Thanks Again,

Ed

I'm not sure if any of the videos were ever put back up, but there was a German guy with a trike / velo he'd custom built, that included ducted forced-air thru his rear motors. It seemed a clever way of doing it, but all I can remembe ris an impression, not details. I cna't recall his ES username, though, but it was a danged impressive build if you can find the threads (about 3-4 years ago, I think).


Then there is a recent thread that Cowardlyduck has been posting to about little tiny fans inside the hubmotor itself. cheap and relatively easy to do since you ahve it open anyway.


As far as how much of a savings, well, if it keeps it from burning up and failing on you on a ride to or from work, it is worth it.

There are other ways to make things reliable, too, and that's where winding it for how you wnat it to behave, and oeprate with hte controller you have (or get a controller that operates it the wya you need it to) comes in. If the motor and/or controller aren't setup and designed to do what you want, and do it efficiently, then all the wasted power from the inefficiencies ends up as heat (and lessened range and/or greater weight on the bike for battery you wouldnt' otherwise need, *and* more mass for it to have to acclerate and push up the hills!).

So if you can read around on all those threads about motor building and redesign/rewind/etc., then you will have a lot more information to help you make this thing do exactly what you need.

(I don't know enough to directly help, unfortunatelY).

My rebuild is on hold for at least a couple weeks, but I will post back here once I order the wire and start the process. I am attempting to go in on a group purchase of a 3kW motor for around $200/ea in addition to this rebuild, but we need at least another 13 people as of this posting. I posted the link to that thread if anyone is interested that has not read the thread yet.

http://endless-sphere.com/forums/viewtopic.php?f=31&t=63394

http://endless-sphere.com/forums/viewtopic.php?f=2&t=63142

Aren't the magnets and halls trashed as well in your motor?

I think you'd be wildly better off simply getting a new motor(preferably one that has a continual rating that matches up with your usage) and venting it yourself.

It might be a good idea to use a temperature sensor with a cycle analyst version 3 utilizing it's thermal rollback feature as well, if you are concerned about frying another motor.