Miles' DIY motor thread
- Thread starter Miles
- Start date
woot!
liveforphysics
100 TW
Hooray!!! It's definitely past-due to get Miles applying that extensive motor function/operation/design knowledge and CAD skills. 
Any parts you need CNC'd for your motor Miles, make the code for it, tell me what alloy, and I will CNC it for you and send it back for free.
Any parts you need CNC'd for your motor Miles, make the code for it, tell me what alloy, and I will CNC it for you and send it back for free.
Malcolm
10 kW
So how will you make the cores Miles? Or have you found another motor you can borrow them from?
Thud
1 MW
Looking like some fun there Miles,
I can generate any cnc code any one needs. let me know if there is anything you need.
(I have generic posts of most current controllers also).
Looks like I need to get cracking on the Dyno'mometer......
Cores for axial flux motors are a problem. These will have to be made from scratch.Malcolm said:
Originally I was going to machine from a laminated block - like this:
Then I saw the method that Shane Colton was using and thought I'd try adapting that to my case. You start with a set of identical stamped or laser cut laminations (as with a radial core). Here's the fabrication sequence:
Attachments
Malcolm
10 kW
That's a clever way to use the same lamination. Does it require some compromise in copper fill since you don't have a tapered web? (Excuse my stupid questions. I know little about motor design apart from the scraps I've picked up reading here, but I'm fascinated by these motor builds).
How do you shape the stacked laminations? I imagine it's something a little more refined than jamming them in a vice and taking a file to them
I can't see any change between image 3 and image 4. Is it out of sight or just me missing the obvious?
Edit: Forget that last one
How do you shape the stacked laminations? I imagine it's something a little more refined than jamming them in a vice and taking a file to them
I can't see any change between image 3 and image 4. Is it out of sight or just me missing the obvious?
Edit: Forget that last one
You would increase the percentage of end turn and the amount of iron by making the copper run closer to a radial direction. I'm not sure exactly what the trade-offs would be?
I'm not even sure how I will hold the laminations together, yet... I think the block could be cut reasonably cleanly with a slitting saw.
I'm not even sure how I will hold the laminations together, yet... I think the block could be cut reasonably cleanly with a slitting saw.
Thud
1 MW
Some nice figured Harewood would make lovely ends Miles :wink:
12p3phPMDC
1 kW
- Joined
- Mar 16, 2009
- Messages
- 462
The eddy currents will be frozen in time if you use curly maple venner as your lamination insulation
and will buck eddies in the iron!! :wink:
But seriously, the dovetails would be a slick way to attach to the stator.
and will buck eddies in the iron!! :wink:
But seriously, the dovetails would be a slick way to attach to the stator.
enoob
1 kW
birds eye maple. i hear it has fantastic eddy current handling capability's.
i like the dove tail , with grub screws through the back into each section (or some kind of retaining ring) you would be able to wind each pole separately and install after winding . very 8)
i like the dove tail , with grub screws through the back into each section (or some kind of retaining ring) you would be able to wind each pole separately and install after winding . very 8)
Reagarding securing the lams together, a quick dip in some insulating spray (like the stuff meant to coat magnet wire to keep it from vibrating loose in motors and transformers) and then compressing them together with a vice and woodblocks should work (to squeeze out excess liquid and ensure similar distances between all lams).
The dovetail is an interesting way to do it--I think enoob is right about how it would probably work (not sure if you intended it that way).
Regarding the bearings with axial and side loads, if you use two bearings (one to take the side loads) it should help, right? If the side-load bearing presses up against the loaded item at a point that it takes the side load before the axial bearing would have to, at least.
FWIW, the axial radiator fan motors I used on DayGlo Avenger only had axial-load bearings, and were probably well-used as fan motors before I got them. Then using them as direct friction drive motors for a few months resulted in no obvious bearing problems. (the shaft broke at the circlip groove on one, but that has nothing to do with the bearing itself )
The dovetail is an interesting way to do it--I think enoob is right about how it would probably work (not sure if you intended it that way).
Regarding the bearings with axial and side loads, if you use two bearings (one to take the side loads) it should help, right? If the side-load bearing presses up against the loaded item at a point that it takes the side load before the axial bearing would have to, at least.
FWIW, the axial radiator fan motors I used on DayGlo Avenger only had axial-load bearings, and were probably well-used as fan motors before I got them. Then using them as direct friction drive motors for a few months resulted in no obvious bearing problems. (the shaft broke at the circlip groove on one, but that has nothing to do with the bearing itself
)Thanks for your thoughts, amber.
Yes, bonding the lams together is a possibility. Not sure what the core insulation coating will be, yet, though... The core assemblies will also be coated in epoxy, after machining.
To be honest, I hadn't envisaged that the cores would need any more than the dovetail to secure them to the plate. I'll probably put a spacer in the exposed outer end of the dovetail-way, too, though.
For bearings, probably 2 radial bearings and a thrust bearing or maybe some sort of angular contact or tapered roller bearing. Need to find out which way is the least lossy....
Yes, bonding the lams together is a possibility. Not sure what the core insulation coating will be, yet, though... The core assemblies will also be coated in epoxy, after machining.
To be honest, I hadn't envisaged that the cores would need any more than the dovetail to secure them to the plate. I'll probably put a spacer in the exposed outer end of the dovetail-way, too, though.
For bearings, probably 2 radial bearings and a thrust bearing or maybe some sort of angular contact or tapered roller bearing. Need to find out which way is the least lossy....
Hadn't thought of the tapered rollers for this, although I have a partial idea for using some from a motorcycle steering post to create my "ultimate" cargo-bike custom made rear axle and wheel from. Not sure if it's a good idea or not but I have them already.
BTW, I said "axial" bearings when I meant "radial" because I am an idiot. So pretend I said the right thing when you read my post above.
BTW, I said "axial" bearings when I meant "radial" because I am an idiot. So pretend I said the right thing when you read my post above.
liveforphysics
100 TW
Wouldn't this design have a pretty poor copper fill percentage due to the inner diameter to outer diameter difference in available space for windings?
You know what the solution is Miles. Larger diameter rotor.
You know what the solution is Miles.
Larger diameter rotor. John in CR
100 TW
liveforphysics said:You know what the solution is Miles.
Ah, my solution to everything for these motors. I guess that darn weight limit of the challenge gets in the way.
BTW Miles, put me at the top of the list for a transmission sight unseen as long as the price isn't too crazy.
AussieJester
1 TW
Very nice and all Miles all above my head...as usual LoL ...so, can i please have a peek at the bicycle you have hidding away your going to mount this motor on pleeeeeease....
KiM
KiM
