Motor Material Choice

[IanFiTheDwarf]

I’m back on my axial motor build again and have just received the laminations for the coils, so I now need to build some of the structure to support the coils.

Basically I need two discs which will be about 280mm diameter with holes milled in them to slot over either end of the coils then the whole assembly will slot into the tube of the motor housing and be screwed in place from outside with quite a lot of screws.
I was going to use aluminium about 8 mm thick but I am concerned about eddy currents being induced and heating everything up reducing efficiency. So now I am thinking about various plastics polycarbonate is the strongest non reinforced plastic I know of but I also thought about carbon fibre, glass fibre or Tufnell

Carbon fibre should be the strongest but the carbon is electrically conductive so might give problems with eddy currents also

Next strongest is glass fibre (FR-4) but from my experience it can delaminate when machined (but maybe that was just a poor grade cheap material)

Tufnell I have used quite a few times and it has always served me well but I have never thought of it as a modern high strength material

Do you thing the aluminium will be a problem and if so what would you suggest?

The cylindrical section and the two end plates of the motor will be aluminium.

Cheers
Ian

 

[Miles]

Hi Ian,

I wouldn't use aluminium in the path of the magnets (even with iron core).

I don't know enough about the electrical conductivity of CFCs. It's not simple...

Tufnol might not give you enough thermal headroom (unless you use one of the high temp. types)
http://www.par-group.co.uk/high-temp-insulation/Tufnol-Industrial-Laminates.aspx

Glass reinforced Polycarbonate?

 

[Thud]

I am using Baltic plywood...with some fiberglass re-enforcement/high tem resin.

no metal in the swinging magnet path other than copper windings & iron cores.

 

[Kingfish]

Agreed: Don't use CF (or Aluminum) anywhere near electrical fields or you will be conducting levels of alternating current relative to proximity.

I like the idea of Plywood as an alt to FR-4 :wink:

Clever! However - I wonder what the thermal limits would be. I suppose it would be at least to the curing temperature. Then there's the question of what occurs when it becomes too hot and begins to char. At what temp does cellulose break down? :? Brief investigation says paper burns at 451°F/233°C (I should have known that from Bradbury :lol: ) and that wood ignites at temps even greater. So perhaps woody composites (glass fibers, concrete, etc.) would raise that substantially higher. Then were back to the limits of the epoxies - and well beyond concern.

Most illuminating! KF

 

[IanFiTheDwarf]

Temperature shouldn’t be a problem as the coils will be liquid cooled.

I am more concerned about structural strength as my target is over 600NM torque.
If I have 32 screws at 140mm radius that works out at 145N shear force on each screw and there could be spikes of much more than 600NM. An M 8.8 screw has a static shear load of 4212N so should be more than 1000N dynamic (I have read dynamic should be between ¼ and ½ static load) which is more than enough but I am not convinced plastic can hold that torque in such a small area.
I think I would be best to helicoil all the screw holes and possibly go for 64 or more screws but that means putting screws into the weaker parts of the disc where the cut out is for the coil.

I could also go for 10mm discs and M5 screws.

Glass reinforced Polycarbonate looks very good but I’m betting it’s not easy to get hold of and pretty hard to machine, I will see if I can get hold of FR-4 first the failing that Tufnell and cross my fingers.

Cheers guys

 

[speedmd]

The FR4 is much better material than the glass filled polycarb I have used. Chemical, heat and warp resistance are much greater and well as holding screws /threads. You will want to seal all surfaces so you don't get into moisture jacking it's way into the fibers and splitting/delaminating on any high fiber content material. GF Nylons are also good as long as it stays either wet or dry. It tends to swell a bit with moisture.

 

[Kingfish]

OK, I have to ask: What are you building that requires 600 Nm of torque?

~KF

 

[IanFiTheDwarf]

It’s for a car actually.
I had been thinking about building a motor for a couple of years and researched several different technologies and came up with a few ideas of my own, and then I picked up a non-running Mazda rx8 for £300 so my design grew to be able to power a car. It kind of happened without me realising, I just kind of found parts which came together to make a bigger motor then I started tweaking everything until the figures suited the car. Now I have got the Mazda running so I will probably sell it and spend some of the cash on building the motor. It’s more about the motor to me now but there is always RX8’s going cheap with knackered engines if I ever had enough cash to buy the batteries I would need for a conversion.

The motor is based around a car wheel hub with two break discs for the rotor’s
Outside dimensions are about 300mm diameter and 110mm deep
My latest iteration should be able to produce around 450 nm all day long with over 650nm for long enough to get most cars from standstill to over 100mph. that’s from 400v and 600A peak phase current

 

[Kingfish]

So lemme try and understand:

Are you going to have one monolithic motor with several rotors transmitting through a transmission, or direct-drive rotor(s)/wheel?

Diameter looks good. Sharp-looking car: Always liked Mazda rotary engines, although the seals were a problem.

Excited for your conversion! KF 8)

 

[Punx0r]

I'm not clear on how your screws will be arranged/loaded, but the general rule is that threaded fasteners should never see shear loads (especially cyclic). Shoulded bolts being something of an exception. Rather, the friction between the joint faces should be sufficient to carry the shear force.

I think your main issue with threading into a plastic or composite will be achieving sufficient preload in the screw to achieve a good joint clamping (and also keep the screws from working loose). Regarding that I think you're probably on the right track using lots of small screws, so you get a good ratio of thread length to diameter, giving sufficient thread engagement that the composite can take the load from the screw thread. Ideally the thread will be stronger than the screw, although this isn't always practical.

Check out "thread shear area" to work out how much thread length you need in the composite to take the preload of your chosen fasteners. Unless the material manufacturer has already published minimum thread engagement values

Apologies if this is trying to teach you how to suck eggs. It was something of a realisation for me when I learnt that sometimes the solution to a sheared bolt is sometimes to fit a smaller, not larger bolt

 

[IanFiTheDwarf]

Single motor with two rotors, each rotor has 16 arc magnets with 16 rectangular magnets between them making a hallbach array. The average magnetic field across the magnets is a little over 1.1T, which is where the high torque comes from. From my calculations the motor should weigh in at about 30kg.

If fitting in an RX8 it could be mounted direct to the diff doing away with the engine, transmission, prop shaft, exhaust, fuel tank and quite a few other bits and bobs.

All the RX8 conversions I have seen have the motor in the engine bay and cut the boot floor out to put half the batteries in there. with the motor in the rear and everything else removed there should be room for about 300kg of batteries in the transmission tunnel, in place of the fuel tank and under the bonnet (hood).




I might have to rethink my fixing method? For a production motor I would probably cast the coils in resin straight into the housing with some recesses milled to give some keying. For now I want to be able to disassemble the motor as far as possible, maybe keys is the way to go?

 

[Kingfish]

Gosh, that’s a lot of torque upon two rotors. Single stator or 3?

>> I might have to rethink my fixing method?
For myself, I have spent a lot of time looking at various composites and layouts. The ideas ebb and flow… though I should not complain coz sometimes nature gives you a dang good answer if you look hard enough. If this was easy – we’d be driving them already!

OT: How far is Fife from Glasgow?

From where I’m standing… the wind blows around me.
On you go! KF

 

[IanFiTheDwarf]

Single stator, it’s very over engineered

http://endless-sphere.com/forums/viewtopic.php?f=30&t=54548

I’m about 40 miles from Glasgow, about an hour give or take depending on traffic.