12 kw rc motor

[dave.com]

Nobody likes the idea of a normally zero contact plain bearing?

Cheapest.
Almost zero drag (only windage loss).
Effectively infinite RPM range.
Strongest support for bell to prevent magnet/stator crashes when needed.
Only needs to be a smaller air gap than the magnet to stator air gap to function.
Reduce cost, noise, and maintenance.

By all means, this would be the cleanest solution. My hunch is the skirt bearing evolved for use in RC aircraft swinging giant props that can induce large radial and axial thrust loads due to gyro effects, etc. If there is a pilot bearing on the end of the shaft to take most of the belt/chain/gear radial loads, seems like the plain bearing may be the best solution. And I could make one on the lathe in 10 minutes. If our "safety bearing" does crash at 11,000rpm, I think I _do_ want a freewheel on my drive!

 

[liveforphysics]

Nobody likes the idea of a normally zero contact plain bearing?

Cheapest.
Almost zero drag (only windage loss).
Effectively infinite RPM range.
Strongest support for bell to prevent magnet/stator crashes when needed.
Only needs to be a smaller air gap than the magnet to stator air gap to function.
Reduce cost, noise, and maintenance.

By all means, this would be the cleanest solution. My hunch is the skirt bearing evolved for use in RC aircraft swinging giant props that can induce large radial and axial thrust loads due to gyro effects, etc. If there is a pilot bearing on the end of the shaft to take most of the belt/chain/gear radial loads, seems like the plain bearing may be the best solution. And I could make one on the lathe in 10 minutes. If our "safety bearing" does crash at 11,000rpm, I think I _do_ want a freewheel on my drive!

I agree.
Using a cheapy little thin Teflon insert for both surfaces should do the trick just fine for a general purpose solution. Teflon/teflon friction has a 0.04 kenetic friction coefficient, so even a vicious bell impact of 1,000N with force pushing the bell against the stator will only result in a split second reduction of torque by ~4nm. If you wanted to get extra fancy, pressing in a ring of SiC on each side would give a plain bearing that could run with very frequent contact events and not wear, but would be a higher friction even when/if contact occurs.

 

[katou]

What the heck is that? I just did a search on "normally zero contact plain bearing" and I couldn't figure out what it was.

Someone want to post a link? Are you talking about a oilite bushing type bearing?

Katou

 

[dave.com]

What the heck is that? I just did a search on "normally zero contact plain bearing" and I couldn't figure out what it was.

Someone want to post a link? Are you talking about a oilite bushing type bearing?

Katou

No, not like an oilite. The mechanical idea is not really as an active bearing at all, just a backup in case of primary bearing or structural failure that would allow the rotor and stator to come in contact. I found a description of a similar device used for "landing" magnetic bearings from extreme high rotational speed. Original link to white paper http://www.skf.com/files/081016.pdf

"For the purposes of this paper, a successful auxiliary bearing landing system is defined as a passive bearing system that will support the rotor duringa delevitated coast down from full speed without excessive rotor deflections or damage to any of the rotating or stationary components."

In our case, we're not levitating anything magnetically or otherwise, just trying to prevent a hard crash of the stator with the magnets.

 

[bigmoose]

Luke and guys, if you stay with a sealed ball bearing, wash out all the factory grease and put a match tip sized amount of DuPont Krytox grease in there and see what happens.

On the solid bearing idea, if it flys or floats most of the time this is a place for coatings. One surface hard and polished, the other coated with a baked on teflon like Xylan.

 

[Miles]

On the solid bearing idea, if it flys or floats most of the time this is a place for coatings. One surface hard and polished, the other coated with a baked on teflon like Xylan.

How about hard-anodised, or Keronite treated, aluminium, with a PTFE coating?

 

[Miles]

How about de-coupling the original skirt bearing with a clearance fit which is less than the motor's airgap? As with the above, the bearing would only be active in extreme situations....

 

[liveforphysics]

How about de-coupling the original skirt bearing with a clearance fit which is less than the motor's airgap? As with the above, the bearing would only be active in extreme situations....

Whoa. That's thinking outside the box my friend! I love that option!

 

[AussieJester]

Personally i would be more than happy with the motor as is Hal. I wont
be getting anywhere near the limits of it nor will anyone else here silly enough to use them on a off the shelf bicycle.

Only ones that will need these super dooper kryptonite based platinum infused teflon coated Keronite treated
fan-dangled things as far as i can tell are those using them in e-motorcycles and/or in multiple numbers in
in a car setup?.

Why can't they be sold to those that want them "as is" under understanding there is zero warranty
and zero blame to Hal if they do pack up? I totally appreciate Hals time and efforts with the testing (don't get me wrong there)
for those that want the data on the motor but i am sure i am not alone at not understanding what the hell most of it means
nor caring, If its ~85-90% efficient great, pissing about to gain 1/2 or 1% makes no sense to me, add another lipo pack
if your worried and/or go buy a frock motor...oh wait, their even less efficient Hal if you would be
so kind to pm when they are available please i would be very appreciative... ..... /jm2c Carry on

KiM

 

[SilverSurfer]

How about de-coupling the original skirt bearing with a clearance fit which is less than the motor's airgap? As with the above, the bearing would only be active in extreme situations....

If the bearing race isnt spinning with the bell, when contact is made the friction and impact would be very detrimental, an air gap of teflon friction bearing would be better in my opinion. As stated, a skirt bearing isnt neccessary 100% of operation only during the most extreme adverse situation. More of a crash bar. Impact of the skirt on a non spinning bearing surface thats not properly lubricated would be bad. SS

 

[Miles]

If the bearing race isnt spinning with the bell, when contact is made the friction and impact would be very detrimental,

Detrimental to what?

As stated, a skirt bearing isnt neccessary 100% of operation only during the most extreme adverse situation.

I don't think we know how well dynamically balanced the rotor of this motor is, yet...

 

[SilverSurfer]

If the bearing race isnt spinning with the bell, when contact is made the friction and impact would be very detrimental,

Detrimental to what?

As stated, a skirt bearing isnt neccessary 100% of operation only during the most extreme adverse situation.

I don't think we know how well dynamically balanced the rotor of this motor is, yet...

Detrimental to the bearing and the skirt! If the skirt spinning at say 5,ooorpm where to come into contact with a bearing race not spinning at all, the resulting contact in the Machining industry would be classified as a crash! I am not an EE, I am a machinest. The sudden friction and attempt to bring the race up to speed with the bell would result in disasterous chatter, once the chatter ocsilations start they will likely multiply themselves to the point of total motor failure! This failure could be as little a breaking a part of two maybe siezing it up if your lucky, to internal parts exploding resulting in complete breakdown of the motor while spinning and under torque, you could get very hurt! Of course, maybe Im a dork and there is no DETRIMENTAL risk of letting the skirt spin freely around a non contact non lubricated bearing race. Perhaps someone should invest there money into a motor, put it under their ass, and try it out. Please, feel free to prove me an idiot in my assessment of the situation.

The dynamic balance of the bell is certainly critical in any case, even more so with a air gap bearing situation. Perhaps if it is an issue that becomes hard to get around an internal or even external harmonic hyrdraulic balancer could be helpfull. As in a liquid filled balancing ring that would help to dampen vibration similar to ICE.

 

[Miles]

There may well be a problem but, if you use unorthodox solutions, the only way to find out is to test, as you say. There's no reason why the contact faces between the bearing and the stator structure shouldn't also be lubricated, is there?

 

[SilverSurfer]

There may well be a problem but, if you use unorthodox solutions, the only way to find out is to test, as you say. There's no reason why the contact faces between the bearing and the stator structure shouldn't also be lubricated, is there?

Yes, lubricated is better than dry contact. However the zero contact teflon inserts may just be the ultimate solution if they have decent balancing, or if we can fabricate some kind of harmonic balancer. Normally outrunners dont even have skirt bearings, but this is a unique application with extremes to be sure. Even if in my application I were not to get much past the 5000 rpm point, I would still want to do better with that issue than the bearing they have! The efficiency hit with that bearing is nuts! Not to mention cost of teflon inserts would be incredibly affordable solution. cant wait to get my hands dirty! LOL SS

 

[Miles]

If one were starting from scratch, I guess it would be be more sensible to choose an "aspect ratio" that did away with the need for a skirt bearing, even as back-up.......

The only example of radial bearings being used "uncoupled", I could think of, was as blade guides on bandsaws.

 

[johnrobholmes]

Agreed. A shorter motor with no need for a skirt bearing would be ideal. Increase the stator diameter by a factor of 1.25 and decrease the length by 1/2.

Otherwise a 5 or 6k rpm limit will have to suffice. I do like the idea of a teflon or ceramic bushing however.

 

[liveforphysics]

Agreed. A shorter motor with no need for a skirt bearing would be ideal. Increase the stator diameter by a factor of 1.25 and decrease the length by 1/2.

Otherwise a 5 or 6k rpm limit will have to suffice. I do like the idea of a teflon or ceramic bushing however.

If you make it shorter and bigger around, then you end up with lower speed limits from the eddy losses, and higher percentage copper losses on end-turns.

Radial motors generally get more power dense and efficient as they get long and thin.

But... if you're going to go through the work of a total custom motor, going axial flux is a no-brainer if you're looking for performance, best power handling from good passive cooling design options, least losses from parasitic effects, outstanding torque vs volume and torque vs weight, etc etc.

 

[johnrobholmes]

Would eddy losses increase if the pole and slot count were not modified? End turn losses would increase, that is agreeable.

 

[Biff]

Would eddy losses increase if the pole and slot count were not modified? End turn losses would increase, that is agreeable.

If pole count were not modified, the rotor iron gets heavy.

As the magnets get larger, the flux carried through the rotor ring between the magnets gets larger, and therefore the ring needs to get thicker. Thicker ring, with larger diameter means significantly more mass. If you have more small magnet poles, the flux travelling between the magnets is lower so the ring can be thinner.

 

[SplinterOz]

ok with all this talk about the skirt bearing I thought I would ask a question about mounting one of these.

If I mount the "skirt" to something solid and place a sprocket to drive chain on the shaft, wouldn't that already place a sideways load on the shaft and bearing? Will I need to mount a bearing holder and bearing to place the shaft into so the shaft is held perpendicular to the chain?

Not sure if I am making sense today.

 

[AZDoug]

A floating magnetic bearing has already been invented.
It's called a halbach array bearing, and livermore labs is the holder of the patent.
It's possible an electric motor could also be made with this kind of magnetic configuration, but that's out of the scope of this thread.

Here's one link to the lab showing it being used in a linear version.
https://www.llnl.gov/str/Post.html

 

[geoffct]

I am interested in purchasing one of these for testing as soon as they're available. (if you've started making an order list)

Keep up the great work.

Thanks,
Geoff

 

[SplinterOz]

ok with all this talk about the skirt bearing I thought I would ask a question about mounting one of these.

If I mount the "skirt" to something solid and place a sprocket to drive chain on the shaft, wouldn't that already place a sideways load on the shaft and bearing? Will I need to mount a bearing holder and bearing to place the shaft into so the shaft is held perpendicular to the chain?

Not sure if I am making sense today.

Ok so I wasn't making sense. I forgot that the shaft now comes out from the skirt end. Man I want to get my hands on this.

 

[markobetti]

i am going to put today this.. but just to explain before everybody jumps. First controller that could give enough power is dad... bad wiring , Hal kicked the shit..out of his new kelly riding the albino so he was dead before he could be used... my kelly was in error mode from 1 week ago when i made a mistake of programing and running kelly at the same time , so we tested this with some chinese infineon base controller with only 800 watts of power... So the controller gave 200-800 watts of power to the motor at somwhere around 74 volts of lead acid batterys, max 4 -12 amps . Anyway the reson i am showing this to you is because tommorow we are going to move Dompas car with kelly: KEB72601,72V for Crystalyte 5000 Series that we revived from error mode and we will finally see the load effect on this motor. I think this is not bad for start . This is Dompas car : Volkswagen Beetle . He made connection to the motor , so anybody having a Beetle here please spam Dompa for advice :)) There was no straight line to go on ,so The car couldnt get up on a hill because of not so powerfull controller ...

http://www.youtube.com/watch?v=CAw2vqY51oo

 

[liveforphysics]

I love it! Great work getting the motor paired up to the bug tranny.

I can't wait to see it ripping around when you get a real controller hooked up.