Mars Electric LLC Axial Flux Motor

Miles said:
I meant that the core is spiral wound, as opposed to individual teeth stacks.
Click to expand...

individual teeth, as in salient?

this layout looks superficially at least similar to ur entry for the motor challenge.
do u think u might using this as the foundation for ur motor, it should save u from having to fab a bunch of stuf urself, particularly the lams.
in which case will u also keep the copper as is or will u b rewinding?
was the plan from the beginning that ur motor have a spiral wind?
 
What he means is that the stator is one big spiral of steel that is machined instead of being stacks of individual laminations. I have no evidence of the stator being one solid spiral, but considering the size it is not out of the question.


Did some tests with various controllers. The sensorless XieChang will NOT start up the motor. It takes too much force and the controller quits after three to five "pings". A Castle controller will start it up, but at low throttle it pulses a few times to get it going. At high throttle it just starts up. I can actually hear the PWM ramping up and down with the Castle while it is trying to detect the rotor position. Next up will be trying the motor with a sensored XieChang. It may be tomorrow before I get that done.


On 7.4 and 11.1v the no load current is 2.4 amps. I will test the sensored no load amps on 36v as well. There is a little bit of adjustment on the timing too, the faceplate has a bit of movement available.


I have to agree with Josh that this motor is noisy. It makes an efficient speaker. :lol: I can imagine that running it bolted directly to the frame would result in all kinds of noise.
 
I think the way these are usually made commercially is that the slots are punched in the lamination strip - spaced so that when the strip is wound into a spiral they are in alignment. Not really feasible for us. It might be possible to post machine a spiral - the other possibility was machining from a Somaloy blank. I started off with the modular approach. Cores for single rotor axial flux are tricky........

The Mars stator is a bit too big for what I had in mind. Anyway, I'm starting to get more interested in twin rotor designs, now :)
 
Some more feedback using a xie chang 6 fet. It starts up sensored just fine. The no load current at 36v is 3a even (with no gearbox). I don't have a tach around to test the speed, but it is pretty dang slow.

It is also super noisy with the Xie Chang. VERY loud. Much noisier than with the low voltage Castle controller. The last tests will see if an ICE 160 can put a damper on the noise as compared to the Xie Chang.


I would love to do further testing with the motor, but I don't have the time or the proper bike to put some load on it. On monday I will have it boxed up and ready to ship to Luke for some inductance, resistance, and whatever other testing he can do with it (if Luke wants to).



With no modifications (other than shaft) I think this motor would make a great 1 to 3 horse solution. To get it to work easier with sensorless controllers I think a higher magnet count would be needed to reduce the detent torque.
 
did not know about spiral laminations as a manufacturing technique but makes sense.
altho i think it would have to be machined as a block like u were planning to do.
if they are stamped first it certainly would be cheaper but how to get them to all line up at the end would be rather tricky.
i'll dig into it some more & appreciate the clarification, i learned something new.

as for the noise, do you think driving it with a sine wave is worth a try?
 
John, by any chance do you have pix to aid in the disassembly of this motor without doing hidden damage?
 
The only thing needed to disassemble the motor is to remove the front dust shield, and remove the C clip. After that it is just a matter of unbolting the two case screws and using a gear puller to get the stator removed from the magnets.

Upon reassembly I did not reinstall the C clip, but I did install an internal washer to keep the airgap in the motor. Either way the bearings see the axial load created by the stator to magnet attraction.
 
Thanks John. Appreciate the tips!
 
No problem at all! Glad to help out when I can, even though it is pretty little compared to you trained engineers and programmers :lol:
 
Final tests for me, Castle ICE HV 160.

No load amp draw 2.83a. Speed seemed to increase a bit too.

The motor is still very noisy. Changing switching frequencies didn't do anything for the noise, but the Castle did seem a speck less noisy than the Xie Chang. As a final test I spooled it up all the way and then let off the throttle while paying close attention to how it spools down. The noise is coming from the platter vibrating, plain and simple. Every time a magnet passes a pole there is a little thunk. Add up all the times that happens in one revolution, multiply that by the rpm, and we get a nice howl!


The only thing that could be done to lessen the noise IMO is to reduce the detent torque (aka cog force). A higher magnet count would do this, although I cannot guess at how much. Other design changes could reduce this as well.

Other changes I would like to see to a production motor (although I would like to stress that this motor would work fine as is).

1. Reduce detent torque
2. Increase magnet count.
3. Reduce the shaft diameter by around .03mm to ease bearing to shaft interference.
4. Reduce material in the rotor, could probably remove 300 grams there.
5. Make the winding tidy, or possibly use a different winding scheme. As is there is a LOT of end turn copper.
6. Balance the rotor. There was some vibration with each motor revolution.

This motor could easily be sub 5 pounds with minor modifications. All in all, I like it!
 
Cool Man,

Thanks for checking it out. I appreciate your feedback and interest.

I am not sure about the noise issue, We were surprised to hear it crying out under load.

What were the frequencies that you tried. Had you said something about trying a much higher frequency for the switching?

I have one going into a trike right now for some hands on, Cycle analyst equipped testing. John had made one up with the wires
coming up out of the side, instead of out on the end, and shortened it up to 3 inches thick!

Do you have Luke's contact info, has he shown interest in testing this unit out, or should we send it to Big Moose for some investigative work?


Thanks again, Josh K.
 
I tried on "outrunner" mode, 8k, 12k, and 18k switching if I recall correctly. The noise didn't change a bit and it didn't want to run right above 12k anyway. I do think it is just the platter rumbling past the poles, and not the controllers that are making the noise for the most part.


Wherever you want me to send it. I am waiting for a contact from Luke. He may be in the middle of a move right now.
 
Sounds a lot like my 80-100-B did on the HV-110 (before I blew it up)... except for the growly freewheeling at the end of power-off spin-down.

"Detent", however, in American - is a sound I've never heard before! :mrgreen:

DEEE'tnt!

The Francais would say "day-tawnt"... which is what I'm used to up here in Canada.

No offense intended, just enjoying the myriad of accents around the sphere.

Watching this thread with great anticipation, too. This motor has a lot of potential, I think.

Thanks to all for the great work and info!
 
Well, thanks so much for testing it out, Let's get it out to Big Moose for a little more evaluation.

In the mean time, I am installing one into a trike, equipped with 24 or 36 volts, cycle analyst, and temp sensors...

Maybe we can get some real world testing in soon, and bump it up into the 1,500 watt area for a wee bit...

Perhaps I can get a quote from John Fiorenza for a group buy of a few of these, once we figure out what we really want...

This is one of the best collaborations I have ever done, thanks to this here cool interweb site...

Peace, Josh K.
 
That thing is loud. My guess is that something's amiss and there is some play in there somewhere. Any way to check the runout?

Plus the rotor may not be very well balanced... it's for a floor-scrubber, after all.

We should compare it to a duplicate unit.
 
The bearing to shaft fit is so tight that there is no play that I can measure, the shaft surely does not move axially. I am quite sure that it is the rotor platter rumbling when the magnets pass the stator, merely an effect of the motor construction. The detent torque is quite large, and when the shaft is turned manually (it takes a lot of force!) it makes the same rumble with no power to the motor at all.


This does have me thinking that a sine wave controller could make it run quiet, but alas I have none to test with and they aren't exactly an item I can buy cheaply or easily.
 
johnrobholmes said:
The bearing to shaft fit is so tight that there is no play that I can measure, the shaft surely does not move axially. I am quite sure that it is the rotor platter rumbling when the magnets pass the stator, merely an effect of the motor construction.
Click to expand...
The platter is what I'd like to get a dial-gauge on.
 
You must log in or register to reply here.

Similar threads

E
    • Like
Eldridge Axial Flux Motor/Generator
Replies
19
Views
4,453
Logic11
L
E
    • Like
Homemade Axial flux PM motor
2
Replies
27
Views
12,310
Erik952
E
T
Multistage axial flux motor design and prototyping
2
Replies
25
Views
6,652
larsb
L
spinningmagnets
Axial Flux Reluctance motor, pondering DIY layout
Replies
7
Views
3,264
everythingisawave
everythingisawave
P
Designing an axial flux motor
Replies
5
Views
2,395
Kingfish
Kingfish
Back
Top