Please Critique My Motor Before I Waste $$$

Joined
Dec 15, 2010
Messages
152
Location
Brisbane, Australia
Hello all,

I'd be highly surprised if I get any constructive comments to this seen as this section of the forums seems to deal with modifications of out/in runners more than scratch built motors, particularly AF. No harm in asking though...

I have for the past 6 months or so taken a sudden and strong interest in EVs and given my technical background have become highly interested in the deep down inner workings of their drive train. As such I have decided to design and build my own motor. Having never done anything like this before I decided to start small. For reasons I can’t even remember I am pursuing Axial Flux motors. At this stage I just want to see if I can do it, so no serious thought went into this. I simply picked a rough maximum diameter of the motor and went from there. Hence I have no idea on performance – hell I have no idea what controller to use with it yet!

What I would like is for people who know their stuff to simply say “looks good” or “you have NFI what you’re doing”. Maybe even “You can improve it this way…”.

Some specs:

- 3 Phase, single stator Axial Flux Motor
- 9 Teeth
- 6 Poles (32mmx12mmx3.5mm N48H)
- 36 Windings per tooth (18AWG)
- WYE termination
- Hall sensors placed at 40 physical degrees to each other (360/3 = 120/3 = 40)
- Impeller ring designed to suck air in and over the coils
- The axle is hollow and all wiring will go to and from the stator through this
- 2 RHP 61802 Z Bearings (inner diameter = 15mm, outer diameter = 24mm)
- The axle (steel with a 12mm internal diameter) DOES NOT rotate – the rotors will spin on the previously mentioned bearings.
- The stator is fixed to this non-rotating axle.
- A bicycle gear will be attached to the outside of one rotor, and the motor will be attached to whatever it is sitting on via the axle on the other end. I guess…haven’t thought that through either.

Yes, it’s clearly based on the Launchpoint motor. Please bear in mind that this was designed on a strict “cheap as possible” method. If this works even remotely well I’ll be inclined to design a much more professional version and spend much more money on it. I have no commercial ambitions as this is purely for my own interests.

Below are 4 diagrams of the various parts of the motor. I will be sending these off to be laser cut (rough cost will be AU$140). Some small explanations are also included. Please note the 4 are not to scale as the "stator frame" is significantly smaller than the other parts. Infact the first 3 are drawn on A3 paper, and the stator frame is on A4 paper.

Rotor External. Made from 3mm thick Steel. The centre circle is exactly the same size as the bearing from the centre of rotation to the inside of the bearing’s outer sleave. The 4 small holes on the outside will have a bolt going though each to old the plates and internals together. 2 of these will be made obviously.



Rotor Internal. This is a 3mm thin piece of MDF with shapes cut from it to support the magnets, and the impeller blades. This will be glued to the inside of the External Rotor plates. The bolts will also hold this in place (I am self drilling these). The centre circle is the same size as the outside of the bearing. Again 2 of these will be made.



Impeller Ring. The impeller has 12 small plates stuck sort of perpendicular to it in the half cuts that adorn it’s inside edge. These blades mate with the slits seen on the internal rotor plate and should provide perfect spacing between the 2 rotors the whole way around the motor. Again the impeller ring will have bolts going through these but I will be self drilling them. Only one of the rings and 12 of the blades will be cut. The impeller ring and blades are made from 2mm Aluminium.



Stator Frame. The stator frame holds the 9 coils. It is made from 6mm thick MDF. The coils sit in the 9 large slots and will be glued in place. Holes for the Hall sensors will be self drilled. All wiring will be recessed into the surface of the stator frame (using a Dremel tool or similar) and fed to a hole cut in the axle. The whole lot will be sealed with an epoxy of some sort…I haven’t thought that far ahead.



I’m kind of hoping it will be powerful enough to move a push bike, but if not then I’ll just convert it to a generator instead once my testing is over. Though if the modifications to my design to make it powerful enough for a pushy are simple I may do that as well…

I am still iffy about a few deisgn points, but these are finishing touches (how to hold it all in place on the axle for example - my current "wedge it on tight" method does not satisfy me) and can be worked out easily. I guess this is a "will it work?" from the pure motor perspective.

Good, bad or in between? I must admit I’m very nervous to see what people think. Go easy on me please. :(

Regards,

- Matt
 
I haven't read your description in all detail, but I'm under the impression, you want to build the rotor and stator plates from wood? (3mm and 6mm mdf)
You have to make them from steel for stability and for the magnetic backpath.
Also the 6 magnet design isn't the best choice for a motor with a huge diameter.
-Olaf
 
Sorry there are 2 parts to the rotors. The outer 'shell' (first picture) that the magnets back onto. This is made from steel. The inner part is made from wood but all this does is act as a template for spacing everything perfectly and then helping it all stay in place while it's running.

In regards to the 6 magnets I only went from these to change the winding scheme (as I am sure you are aware from my other posts) and wanted to go down in numbers so as to reduce the cost. My other option is upping the number to 12 (so as to keep the ABCABCABC winding scheme) but this doubles the cost of the magnets :( (and slightly increases the cost of the laser cutting - though that's negligible)

Would the performance increase be worth it? I'd imagine it'd see a lot more torque.

Actually could maybe tightening the layout (instead of adding more magnets) work as well? Bring it all closer to the center of rotation, thus reducing the total diameter but keeping the same number of magnets, coils etc? Has this got anything to do with magnet coverage? I'm not sure what the principles are but for some reason the phrase "smaller air gaps is better" is going around in my head...
 
I presume you have seen Axel Borg's motors: http://endless-sphere.com/forums/viewtopic.php?f=30&t=14134&start=15

You might buy his book, before dropping big $$$: http://www.amazingdiyprojects.com/
 
So from what you have said, can I assume there is more benefit to having 12 magnets of a certain area, as opposed to 6 magnets of the same total area? Smaller magnets would not be as strong as the larger magnets. But on the other hand having twice as many magnets would negate that downside some what yes?

This has nothing to do with 'airgap' because thats the distance between magnets and coilheads.

Apologies for that comment - I'm still getting used to using the correct phrases and 'airgap' was the first thing that sprung into my head when I was trying to decide what to call it. I do understand what an airgap is in electric motor terms!
 
TylerDurden said:
I presume you have seen Axel Borg's motors: http://endless-sphere.com/forums/viewtopic.php?f=30&t=14134&start=15

You might buy his book, before dropping big $$$: http://www.amazingdiyprojects.com/

I have it lol. It goes ever so slightly into some theory but is really just a step by step guide to building your own motor (so exactly what I need lol?). I should point out I am reasonably confident that this motor will work. I'm just after some affirmation of the fact really. Crossing my t's and dotting my i's so to speak. Or maybe just nervous.

Really from what I've read thus far there is nothing technically wrong with my design that will prevent it functioning. Just some odds and ends I can do to improve it. Personally mine is a bit more professional than Axel's. Then again his entire motor is scratch built while I'm sending parts of mine off for laser cutting!!! (Probably a bit too extreme - but I don't have the gear to do the work manually unfortunately)
 
I don't really have a target kV in mind. Obviously with my low turns and thick wire (for the size of the motor) it will have a high kV (for it's size anyway). Maybe I should drop to 0.5mm wire to increase the torque? This will make it more feasible for use in a bicycle if it actually works.

To be honest, as with most aspects of this motor, I am building it based on the physical constraints of the rotor dimensions I set. The stator was to only be 6mm thick (the coils were going to be 6mm x 6mm - 1mm wire wound to that size is roughly 36 turns), but given a few comments here I've started the design from scratch using circular magnets to increase the magnet surface to 80% (because I don't want to spend money on custom magnets yet).

Same basic design but re jigged to better performance and efficiency.
 
I think you want the steel backing to be about as thick as the magnets but I'm not a motor guru so I could be wrong. Looks like a fun project.
 
Yeah that's about right - the ticker the better though. I've changed magnets to ones that are 3.175mm thick so the steel backing is slightly closer to the thickness of the magnets. The air gap between the faces of each opposing magnet has also been reduced to make up for this slightly.
 
My other option is to make the holder from Aluminium. It's just as thick as the magnets themselves. That's really all I can do but I think an aluminium holder will be able to resist the magnets pulling out.
 
It's complicating the build somewhat. In re-doing my design I've gone from 4 different sets of materials to 8 (same stuff different thicknesses and so on) so I've taken a step back and tried to simplify it back down.

I'm struggling to see how a 14mm x 3mm thick piece of MDF wrapped around the magnets will break apart at high speed. The 14mm section is the part taking the most punishment as it's perpendicular to the axis of rotation (the 3mm section is obviously parallel). Granted I have no experience with electric motors so I guess I will learn the hard way?

How about 2-pac epoxy? (Araldite...I have no idea what it's called i Germany - resin) That stuff's stroooooong.
 
Back
Top