Using RC motors on E-bikes [Archive]

[johnrobholmes]

I must also say that my comparison of the 14pole/12slot and 2pole/3 slot is holding the switching frequency for a given RPM constant. This in effect locks the 14pole motor into exactly 1/7 the KV of the 2 pole motor. This is why I stated that more poles = more torque, but if we vary the KV of the motor we can of course change this within reason.


I have now become overwhelmed with thoughts. :lol:


Don't get too bogged down in details, my point is that motor torque can be changed via magnet/slot ratio when the KV changes too. Kinda redundant to say, but obviously it stirred up some thoughts.

 

[johnrobholmes]

john, it seems that only using 1/3rd of the stator windings will lower the resistance by 3, so you would get 3X the current, but since there is 1/3rd the active stator poles, you get the same torque, the backemf would also be 1/3rd its original value, so the rpms can increase much higher, the motor seems to actually be more powerful in this arrangement. power is proportional to resistance for a given voltage.

On a 12 slot we would need to use 1/4 of the stator

If you held the wind count the same (I assume so with 3x less resistance in your example) the KV would triple but we would have much less copper fill. 4 times the KV, 1/4 the kT, and now we are back to square one. Yes, the motor would indeed pull more amperage. But, we would need to gear it back down to hit our 500w speed, and probably to keep it from blowing a wind. Having 14 magnets with only 1/4 of a wound stator wouldn't be effective however. A 2 pole three slot design would be much more effective use of copper and magnet.

 

[dirty_d]

Kv has nothing to do with the number of magnet poles, the number of magnet poles just changes how many commutation cycles there are per mechanical revolution. i think what you mean is that the 14 pole 16 slot motor would be 1/4 the Kv as the 2 pole 3 slot motor if the number of windings around each of the 16 teeth was the same as the number of windings around each of the 3 teeth in the 2 pole motor.

 

[johnrobholmes]

Kv has nothing to do with the number of magnet poles, the number of magnet poles just changes how many commutation cycles there are per mechanical revolution. i think what you mean is that the 14 pole 16 slot motor would be 1/4 the Kv as the 2 pole 3 slot motor if the number of windings around each of the 16 teeth was the same as the number of windings around each of the 3 teeth in the 2 pole motor.

Right, which is why I stated that all other design elements were held constant in my fist post. To correct your statement a 14pole, 16 slot 3 phase motor isn't possible. Here is a handy dandy chart to show common ratios and the corresponding commutation "geardown" http://www.powercroco.de/Kombinationstabelle.html

 

[eP]

Kv has nothing to do with the number of magnet poles, the number of magnet poles just changes how many commutation cycles there are per mechanical revolution.

It depend on configuration (stator/rotor geometry).

If more poles means faster magnetic flux change for the same rpm, so in such case Kv falling for the same turns count as faster magnetic flux change giving as a result higher voltage across these turns.
Higher voltage at the same rpm means lower Kv of course.

But for classic RC motor (radial flux, longitudinal) it is not the case - i have to admit that once again .

Radial, transversal is quite different story - here more poles means faster magnetic flux change and as a result lower Kv and higher torque. 8)
We should keep in mind more poles in that case doesn't means higher Rm.
For that kind of config (geometry) Rm don't depend on poles count in fact !!
That is the real reason why we should deeper investigate transverse flux motors advantages and disadvantages.

 

[johnrobholmes]

I clarified a few posts a bit, and added in how torque does change with winding IF gearing is also changed to keep a stated final wheel speed and wattage.

I guess my point is that we can select a motor design based on the geardown that we want. Then small tweaks can be made to wind count to adjust to the voltage used.

 

[eP]

Then small tweaks can be made to wind count to adjust to the voltage used.

I agree.
But small tweaks are useless if you want adjust the torque for longitudinal motors geometry.
So if you want to adjust the torque by poles count you need to go to different kind of motor geometry (transverse flux machine TFM).

That way you can get something like "magnetic gearbox".

 

[johnrobholmes]

While a LFM could be a great idea, you won't find a single one used in RC. We use permanent magnet DC brushed or 3 phase brushless exclusively, which are worlds easier to manufacture and get high efficiency from. If I need more torque than a design will give I can always use a bigger motor, volt up more, or gear down more (or both volt up and gear down).

 

[eP]

If I need more torque than a design will give I can always use a bigger motor, volt up more, or gear down more (or both volt up and gear down).

But what you can do if you need extremly high torque at very low rpm ?

RC world can be happy without TFM i know, but if small TFM become to the reality than these machines could be very competitive for some high torque, low rpm RC applications i'm sure. For the sake of low weight and high efficiency i'm sure. All material costs would be low too (for mass production).

 

[johnrobholmes]

If I want high torque at low RPM I will choose a large diameter motor, possibly an outrunner. Right now there aren't any RC motors that spin at 700rpm tops that we could use for a 1:1 drivetrain, generally they top out at around 15,000 for a decent outrunner or 30,000rpm for a decent inrunner. Personally I would rather use a high RPM motor geared down if I am going for power to weight ratio. A motor spinning 34,000rpm with a 50:1 gearbox would be perfect for 16" tires. That is only two 7:1 reductions, or three 3.7:1 reductions.

 

[eP]

Personally I would rather use a high RPM motor geared down if I am going for power to weight ratio. A motor spinning 34,000rpm with a 50:1 gearbox would be perfect for 16" tires. That is only two 7:1 reductions, or three 3.7:1 reductions.

Why not to eliminate one noisy high rpm stage and as a bonus get a higher efficiency at the same or lower weight ?

 

[johnrobholmes]

Why not to eliminate one noisy high rpm stage and as a bonus get a higher efficiency at the same or lower weight ?

Show me a ready made motor for the task that does not sacrifice efficiency and I will buy one. Right now high speed inrunners trump low speed outrunners on efficiency and power to weight ratio. I may use this motor for my mid sized motorcycle http://www.castlecreations.com/products/neumotors/nm2200.html , the 2230 1.5y. If I can get a special run of a higher pole count for greater torque I would certainly use it, but as of right now it would be custom made.

 

[Miles]

What we need are a few builds up and running, then we can find out what the total system efficiencies really are.. I'm looking forward to Matt's testing with the eLogger, as a start.

 

[eP]

What we need are a few builds up and running, then we can find out what the total system efficiencies really are.. I'm looking forward to Matt's testing with the eLogger, as a start.

Costs are important too. TFM have one unique advantage: there is possible to use the same laminations for different diameter motors.
So you could use the same laminations for 80mm motor as well for 280 mm in diameter too.
Large RC motors are very expensive as they use large and unique laminations. As a result cost per diameter rising sharp above D=40 mm.
Efficient 70-80mm outrunners are crazy expensive.

 

[johnrobholmes]

Motors billed for typical EV use are ridiculously expensive too, but much heavier.


Just got word that a 130x80mm PM inrunner motor is in beta stages, perfect for a motorcycle. Top speed should be 9,000rpm, with a 24slot/8pole construction. I will probably get one of these for version three, but it can hardly be billed as an RC motor even though the design comes from an R/C background. It would perfectly remove one stage of 3.7:1 reduction.

 

[eP]

Motors billed for typical EV use are ridiculously expensive too, but much heavier.

A little OT - lets look how cheap and powerful they coud be:
http://www.osti.gov/energycitations/servlets/purl/885638-y9YVQx/885638.PDF

 

[johnrobholmes]

They do not state the quantity being produced at that price. I would assume production runs of at least 1000, but more than likely 10,000 to meet that price. Double to quadruple the price for retail sale depending on market conditions and consider the price of both fuel, copper, and iron have gone up considerably in the past four year since the report was written.

I fully expect to pay $1000 for a motor that is lighter and higher efficiency as compared to most EV motors, assuming 7-10k watts constant.

 

[eP]

I fully expect to pay $1000 for a motor that is lighter and higher efficiency as compared to most EV motors, assuming 7-10k watts constant.

 

[johnrobholmes]

Totally missed that table, too many pages for me to read :lol: Pretty sweet pricing, but again they don't state what quantities need to be made and that is still OEM price and not low qty retail. Figure an average of 50 to 200% markup on top of that depending on market. I still see the motors having a retail price ranging from $500 to $1400 or more depending on model.

 

[eP]

Totally missed that table, too many pages for me to read :lol: Pretty sweet pricing, but again they don't state what quantities need to be made and that is still OEM price and not low qty retail. Figure an average of 50 to 200% markup on top of that depending on market. I still see the motors having a retail price ranging from $500 to $1400 or more depending on model.

I like study such docs
Keep in mind these motors are for full EV vehicles. We need only 1/20 of power and 1/10 of material weight , so material costs should be fractional (negligible).

 

[johnrobholmes]

Material costs still figure in pretty high, but labor contributes to a larger share of the total cost on smaller motors. I know because I have been designing and buying motors for a few years now :wink:

I think I could bring affordable and high power motors to you guys, two different designs of course. First I will take care of myself and do some more testing. Off to work for now...

 

[John in CR]

eP,

I take it that pic is the guts of a TFM that you keep wanting to get discussion rolling about. If you need beta testers or group buy participants for something with innards like that, put me firmly on the list. At this point a TFM applicable for our use isn't available except maybe to someone where price is no object and is something maybe a handful in the world can DIY.

RC stuff, on the other hand, has a competitive market and is readily available, so I've no choice but go that route for now. I don't know where John's crazy expensive comment came from, but how can $56 for a 2.5kW 63mm motor or $150 for a 6.5kW 80mm motor be beaten? It seems crazy cheap to me. These things have to be pretty efficient too, or the heat would melt them apart. It's not like they are complex either, with just a handful of windings around a stack of laminations, a couple of bearings to support the only moving part, and some magnets. It seems not to hard to get right, and with close tolerances. To top it off, I find the simplicity extremely attractive.

Maybe TFM's are better, I don't understand enough to even formulate an opinion. If they are, then my attitude is that someone needs to get busy making them, and in a lot of different sizes. Electric motors run everything except transportation, and that will change quickly with the cream will rising to the top. Just imagine the Ebike motors we'll be able to get in 10 to 20yrs. The only sticking point I see is increased demand causing copper prices to skyrocket.

 

[johnrobholmes]

There are plenty of cheap motors available for sure, and you must keep in mind that the specs are normally inflated a bit (sometimes a lot) to make the motor more attractive. The outrunner I am running is a very nice high efficiency (mid to high 80's) 3.5kw continuous version, but the retail is only $170. You could also use a cheapo HXT brand outrunner like so- http://www.hobbycity.com/hobbycity/store/uh_viewItem.asp?idProduct=5141 Only $140 for well over 2 pounds of motor and a stated 6kw of power (probably peak input). The cheaper outrunners can have horrible efficiencies, sometimes in the mid 60% ranges, but sometimes creeping into the low 80%. Nice brushed motors can beat cheap outrunners on efficiency, believe it or not. To put it another way the 6kw outrunner may only put out 4kw of power under most efficient load.

The rather expensive motors I have my eye on are handbuilt in the US, and carry efficiencies in the low 90's. Of course I could save a lot of money and just buy a larger cheaper outrunner, but I have already done that. The plan is a 10kw continuous system that is a very fat 9lbs with motor controller.

 

[eP]

I take it that pic is the guts of a TFM that you keep wanting to get discussion rolling about.

No, it is not a TFM.

Maybe TFM's are better, I don't understand enough to even formulate an opinion. If they are, then my attitude is that someone needs to get busy making them, and in a lot of different sizes.

I didn't said they are better, as they almost non exist yet.
But if we are going to high power area we need efficient motor as well as efficient high rpm transmision too.
So my point is: we need to evaluate if TFM is not a better option for that case. For DIY at least now.
If they were able to spread this way maybe they could become as a regular option who know.
If we will not check that virtual option then we will not be aware what maybe we wasting now (how viable way). For do proper costs analysis we should contact scientists designers i suppose. At some high power level TFM should be cost competitive as a DIY option i'm sure.
Scientists are the same people like we are, so we shouldn't afraid looking for good professional advice.

Best regards

 

[John in CR]

There are plenty of cheap motors available for sure, and you must keep in mind that the specs are normally inflated a bit (sometimes a lot) to make the motor more attractive. The outrunner I am running is a very nice high efficiency (mid to high 80's) 3.5kw continuous version, but the retail is only $170. You could also use a cheapo HXT brand outrunner like so- http://www.hobbycity.com/hobbycity/store/uh_viewItem.asp?idProduct=5141 Only $140 for well over 2 pounds of motor and a stated 6kw of power (probably peak input). The cheaper outrunners can have horrible efficiencies, sometimes in the mid 60% ranges, but sometimes creeping into the low 80%. Nice brushed motors can beat cheap outrunners on efficiency, believe it or not. To put it another way the 6kw outrunner may only put out 4kw of power under most efficient load.

The rather expensive motors I have my eye on are handbuilt in the US, and carry efficiencies in the low 90's. Of course I could save a lot of money and just buy a larger cheaper outrunner, but I have already done that. The plan is a 10kw continuous system that is a very fat 9lbs with motor controller.

It makes perfect sense that the smaller outrunners could have horrible efficiencies, because the magnetic gap tolerance is a larger percentage of diameter. I'm not convinced of yet with the larger ones. It seems to me the larger ones would fail almost immediately if they have very low efficiency, because those losses can turn only into heat, and that much heat will quickly ruin the magnets. All are likely hand wound, expensive or cheap. If the silly expensive ones somehow have a smaller air gap, then I'd like that explained to me. Better magnets and/or bearing yes maybe, but not enough to justify the price difference. That leaves the windings and since we're only talking about 8-10 turns, there's now reason the cheap labor can't do that just as well as expensive labor.

10kw continuous in 4kg including controller is a whole different animal, and if it's highly efficient across the power band then please count me in for a group buy/beta test/whatever. That sounds just like what will fill my needs, both for the catamaran I'm building and for the bicycle/motorcycle crossbreed that I want for myself.