Using RC motors on E-bikes [Archive]

[Fumesucker]

A few interesting components I found for an RC based E-bike.

First, a high voltage (for RC) 100A controller for $99.99:

http://www.hobbycity.com/hobbycity/...TURNIGY_Sentilon100A_HV_2-12S_(USA_Warehouse)

Next, a cheap high power motor that may be more efficient than some of the other cheapies $48.79:

http://www.hobbycity.com/hobbycity/...werPro5330-10T_679g/215kv/80A_Brushless_Motor

And last, a wattmeter with a built in servo tester, $35.95:

http://www.hobbycity.com/hobbycity/...roduct_Name=TURNIGY_Multi-F_DC_Watt_Meter_75A

 

[Fumesucker]

I liked it a lot. I thought it was a plus that when I was in the correct gear for pedaling the motor was in an efficient range. I would like to do a similar drive but I'm not sure if I can figure out a way to mount something similar on a conventional bike. More likely to work with small RC components.

That's the entire idea behind gearing, keep the motor in an efficient range. Powering the crank is a good way to do it, if the motor is appropriately geared to the crank then all you have to do is shift to keep your cadence and you are automatically in an efficient operating regime of the motor.

In addition to the fact it might not be strong enough this seems like an relatively heavy and expensive way to just to get a reduction 2:1 or 3:1 reduction. Why not a simple jackshaft with sprockets sized appropriately? Might be easier to mount also.

Agreed, for a single speed a simple jackshaft makes better sense. However, the point of using a geared hub as a jackshaft is to get the multiple speeds in a single unit that has already had the design and manufacturing setup costs thoroughly amortized. Designing and manufacturing a reliable multiple speed gearbox is a far from trivial task.

 

[nasukaren]

A few interesting components I found for an RC based E-bike.

First, a high voltage (for RC) 100A controller for $99.99:

Next, a cheap high power motor that may be more efficient than some of the other cheapies $48.79:

And last, a wattmeter with a built in servo tester, $35.95:

1) That 100A controller is most probably only for peak current draw and might blow with the kind of sustained current draw we use in our bikes. My first cheapo "100A" controller blew because it couldn't handle the back-EMF from when my bike was slowing down against the motor...

2) That little motor only has an 8mm shaft. I think it'd be fun to run a Razor scooter on, but I don't know if it'd be powerful enough for anything more than a very small bike....

3) That wattmeter/servo tester is VERY INTERESTING. Good find!

Sorry to be kinda negative 67% of the time but I've been (literally) burned by the buy-cheap mentality.

Karen

 

[johnrobholmes]

To add to the negativity:

I have a Turnigy Sentilion ESC. They are junk for bike use. They get out of sync if you sneeze at them, roll into the throttle too fast, hit a bump, go up a hill too steep. With pedal assist it wouldn't be too bad, but I guarantee it will be falling out of phase frequently. and programming them really sucks.

Motor- maybe a good find. The shaft size doesn't mean squat really. I have a motor being built for me with a 5mm shaft and 3.5kw of power. I know the shaft will take the power because I have seen the motors in action!

 

[Fumesucker]

Eh, if I try the controller and it doesn't work I could still get most of my money back by selling it on ebay.. Come to think of it, if I "sell the sizzle", I could maybe even make a profit, there are very few HV RC controllers on ebay and I sell a lot of stuff there and have gotten pretty good at writing ads that sell well.

One point about RC controllers, the prop on a plane acts as a large flywheel and makes the job of the controller considerably easier, one thing that could be done to make the motors run more smoothly would be to add some flywheel weight to the motor shaft. Make the flywheel thin in the center and middle with a fairly massive outer ring and you could get considerable angular momentum without a terribly large weight penalty.

I don't want direct drive to the wheels, I want a freewheel on the motor, so I wouldn't have the problem of back emf.

Motor shaft size really doesn't mean much, 8mm hardened steel is pretty tough, that's about 5/16"..

The electric RC plane market is quite mature, the electric bike market is barely nascent, products on the RC market are much more technologically advanced than the e-bike products and have the advantage of having R&D and manufacturing setup costs already amortized over a huge number of units.

 

[lawsonuw]

2) That little motor only has an 8mm shaft. I think it'd be fun to run a Razor scooter on, but I don't know if it'd be powerful enough for anything more than a very small bike....

An 8mm shaft is PLENTY. I bet it's not even hardened. Really, the reason Outrunner motors have such massive shafts is to stiffly support the magnet bell so hardening is pointless. (the stiffness of a material is largely independent of hardening)

Now, the 8mm shaft IS an advantage. These gearboxes are setup for a 8mm input shaft. The mounting pattern likely won't match. (easy to fix) Torque should not be a problem though, with an 86ft-lbf peak "do not exceed" torque it should take a couple kilowatts.

Marty

 

[Fumesucker]

The reason the shafts are hardened is so they will not bend so easily.. Putting a prop in the dirt is a not uncommon occurrence and when you do you stand an excellent chance of bending the motor shaft.

Hardened shafts are not stiffer but they are considerably more resistant to permanent deformation.

 

[John in CR]

Now, the 8mm shaft IS an advantage. These gearboxes are setup for a 8mm input shaft. The mounting pattern likely won't match. (easy to fix) Torque should not be a problem though, with an 86ft-lbf peak "do not exceed" torque it should take a couple kilowatts.
Marty

Marty,

You're the only one I know of with experience using these robot gearboxes. How confident are you that they can hold up to use on a bike with 2-3kw motors running over 10krpm? What about 4-7kw motors? Does your gearbox ever get hot after a long ride? Solving the gearing issue with an off the shelf planetary gearbox for $100 or so, and outputting to a normal sprocket sure is appealing.

With the low speed current issues of these RC motors and controllers, I'm now leaning toward something like a Bafang geared rear hub for low speeds, takeoffs, and hill climbing. Then at 15-20mph have an RC motor be my afterburner and take me up to whatever speed my bike can handle.
OR
Just go with 2 RC motors and 2 gearboxes (one geared high and one low), until someone comes up with an appropriate 2 speed transmission. I really like the idea of 2 motors and intermittent use.

John

 

[lawsonuw]

How confident are you that they can hold up to use on a bike with 2-3kw motors running over 10krpm? What about 4-7kw motors?

I'm pretty confidant that the P80 will hold up to a continuous 1-2Kw for ~100hours. (just look at the size of the gears in the spare parts page!) Lower average power will prolong this greatly, while exceeding the recommended output torque will kill the gearbox quickly. (btw, 100hous @ 25mph is 2,500miles, a rather long distance to ride) 4-7Kw motors are likely to have more torque that the gearbox can handle. (also the torque constant can be approximated using this equation: Kt[oz-in/amp] = 1352.4 / Kv[rpm/volt] ) Just, cranked through a few numbers and 100amps @24v through a ~400Kv motor followed by a 16:1 gear ratio produces about 28ft-lbf of torque at ~650rpm.

Does your gearbox ever get hot after a long ride?

Yes, these gear boxes lack needle or ball bearings on the planets. As such, I'd expect at most 95% efficiency per stage. (i.e. for a two stage gear box .95*.95 = .903) I ran a 36:1 three stage gear box initially, and it heated up noticeably after shooting me up a hill.

Marty

 

[drewjet]

I have the Turnigy 100 HV from hobby city. I am running it with a Currie rear wheel hub motor. Hitting peaks of about 1KW.

It does have a tendancy to loose sync, however I haven't got many miles on it yet and am still working on a battery solution. I will probably end up with a 12s3p of Emolis. As I recall setting it for s;low start and either high or low advance (my memory sux) made it work the best. It cannot start from a dead stop, you have to peddle first. Seams like about 5 MPH was all I needed.

Get the programming card for an additional $7 or$8. it is worth it. Neat find on the servo controller/Whatt meter. I may NEED one of those.

 

[Jeremy Harris]

OK guys, go gently as I'm about to propose what might seem to be a crazy idea.............

I can't help but notice that RC brushless motors are light, relatively cheap and seem reasonably reliable (I bought a couple to play with at my local model shop).

On the other hand, big motors, like the Etek, Mars, Perm etc are heavy and expensive, although they do put out more power (on a continuous basis).

High voltage is good for efficiency, but so would a parallel array of low voltage brushless motors. Building on the excellent work you guys have been doing, I've come up with the idea of making a multiple motor array, with in-built toothed belt primary reduction and individual motor controllers, that would emulate something like an Etek, but run from multiple low voltage batteries.

I've sketched out a design, using two sandwiched CNC machined alloy plates, with the belts fitted inside the sandwich. The sides of the sandwich would be sealed and fed with cooling air from a small high pressure fan. The motor cans would protrude from one plate and act as vents for the cooling air. The controllers would be mounted inside the case, so that they were cooled as well.

A few quick sums shows that such a design would be cheaper, lighter and more powerful than an Etek, yet spin an output shaft at a similar speed, albeit with a lower supply voltage. I've based my calcs on using cheap 2.7kW motors, derated to 1.8kW for reliability, and run at around 30V, 60A using 100A controllers (with forced air cooling for the whole lot).

I can fairly readily get the side plates, spacers, output shaft and bearing housings machined up and would use off-the-shelf belts and pulleys.

What the heck have I missed?

Jeremy

 

[Fumesucker]

I've sketched out a design, using two sandwiched CNC machined alloy plates, with the belts fitted inside the sandwich. The sides of the sandwich would be sealed and fed with cooling air from a small high pressure fan. The motor cans would protrude from one plate and act as vents for the cooling air. The controllers would be mounted inside the case, so that they were cooled as well.

A few quick sums shows that such a design would be cheaper, lighter and more powerful than an Etek, yet spin an output shaft at a similar speed, albeit with a lower supply voltage. I've based my calcs on using cheap 2.7kW motors, derated to 1.8kW for reliability, and run at around 30V, 60A using 100A controllers (with forced air cooling for the whole lot).

I can fairly readily get the side plates, spacers, output shaft and bearing housings machined up and would use off-the-shelf belts and pulleys.

What the heck have I missed?

Jeremy

What kind of total power output are you considering, and for what purpose?

An e-motorcycle or a very small e-car would need something like that, an e-bike, not so much.

And there are some brushless RC motors that have a built in cooling fan, the helicopter ones I think mostly.

If you are talking about using a single belt with multiple motors, the first thing that comes to mind is getting a decent wrap on the motor pulleys, you'll have to have idler rollers to increase the belt wrap to get enough grooves on the motor pulley.

 

[johnrobholmes]

You have missed nothing. It is just plain cheaper to use RC equipment.

 

[recumpence]

This looks, on the surface, to be a good idea. However, multiple motors do not give a direct 1 to 1 power increase. Typically doubling motors gives roughly 25% to 30% more power than one motor (depending on the arrangement). I know this sounds strange, but, I have run dual motors in RC helis and RC cars. It is pretty consistant. That is why I have gone to single larger motors in all my helis, cars, and my bike.

There are many reasons for this issue;

#1 Added drag from the transmission (gears, or belt) to drive them together.

#2 Double the number of bearings in two motors versus one.

#3 There is some issue related to doubling motors and ESCs (you cannot drive two motors with one sensorless ESC). The electrical efficiency goes way down with two motors as well.

I was frustrated when I did dual motor testing in my helis and finally was told by the guys at Castle Creations that they came to the same conclusion (very low additional power from a doubling of the motors).

Matt

 

[Jeremy Harris]

Thanks for the interesting contributions, much appreciated.

I was thinking of using separate belts for each motor, so the belt wrap issue shouldn't present a problem.

I'm also looking at making a "motor module" that would replace something like an Etek, so probably more power than most would want on an ebike.

The plan was to use one controller per motor, fed with a common PPM signal, as I'd already gathered that running a couple of motors on one controller was fraught with potential pitfalls with regard to timing etc.

The apparent drop in efficiency for multiple motors on a common drive system is intriguing, as it can't be readily explained by the very small additional losses in bearings etc. I can't think of a logical explanation as to why this would be, off the top of my head, unless it's related to the controllers getting a bit screwed up as a result of synchronisation problems (although I can't see quite why this should be).

I'd also thought of using a single big motor, but it looks like there aren't too many HV controllers around for the sort of continuous power that I'm looking for, plus I'm not sure that the really big motors (like the 6.5kW ones) will deliver this sort of power for any length of time reliably, in a non-model aircraft environment. (I may be wrong on this, I lack direct experience with them and am going largely on instinct).

I'm sorely tempted to buy a couple of the cheaper big motors and controllers and just hook up soemthing to experiment with and make some measurements, if only to understand a bit more about the apparent power loss problem.

Jeremy

 

[recumpence]

Yes, there are high power motor options out there. The issue is more of a ESC problem than a motor problem. There are 15kw RC motors out there (though they are $1,000 motors). But, finding an ESC that will drive them is another matter. Castle Creations is developing their industrial controllers (should be out by end of year) that should cure this issue.



Here is the information I have compiled over the years related to multiple motors;

#1 If two completely separate, decoupled drives are used, the efficiency is not too bad (such as a boat with two motors running two completely separate props without being coupled together).

#2 You CAN run two motors off one ESC if they are timed together perfectly. This is very difficult to do, however, and is hard on the ESC. So, essentially, you must run two ESCs.

#3 Running two separated belts is more efficient in that you do not need idlers to back bend the belt. With stiff belts, this is an issue, but not with all belts.

#4 The most efficient way to run two motors together is to gear them rather than use belts. You would use one large spur gear driven by two motors.

Yes, you will get more power out of the twin motor setup. Again, the power is not 1 to 1, unfortunately.



Now, the benefits are numerous, however;

#1 You have a fail safe if one setup dies (system redundancy).

#2 More power (again, not 100% gain, though).

#3 Far better heat dissipation of both motors and ESCs.

#4 If you use a oneway clutch setup, you could run just one motor at a time if you want to. There are various benefits to this.



So, it is not a bad idea to run dual motors. However, the power increase is not linear to the amount of motor added.

Oh, I remember back when I was a teenager. I scratch built an RC monster truck. That truck ran 3 motors to one gearbox. It only had a tiny bit more power than two motors, but they ran alot cooler.

Anyway, food for thought!

Matt

 

[vanilla ice]

The examples above make it sound like that multi-motor RC stuff was riding the battery's current limit. Like Jeremy says, I can't see that huge a loss from what we're talking about here.. 70% loss?? No way. Even more on a 3rd?? WTF mate.

if it is true that sucks. I've been eye'in those r/c motors and thinking about a multi motor setup for moped or motorcycle. Be a nice alternative to the big motors. I was picturing like half a dozen pretty little annodized RC motors humming along in harmony...

 

[johnrobholmes]

The added weight could be a very large factor in RC multi motor use, as the motor and controllers are a much larger % of the total weight. I would want to see the motors able to work independently of another, with freewheels or simply being on front and rear wheel. They will work more efficiently when worked together as the load is lower for each motor.

 

[Miles]

I was picturing like half a dozen pretty little annodized RC motors humming along in harmony...

Something like this?
http://www.youtube.com/watch?v=umZpn610AkE&feature=related

 

[Fumesucker]

Separate belts for each motor are going to lead to a very "thick" module, inadequate shaft length on the motors could become an issue as well. Not to mention that the commercially available pulleys are fairly expensive to get in "onesies and twosies". Without considerable machining to lighten them the commercial pulleys I'm familiar with weigh quite a bit too.

It seems to me that running two motors from one battery pack you are more limited by the pack than anything else. I've been scratching my head for a while trying to figure out why two motors wouldn't produce say 90% or more of the power of a single motor and I really can't think of any reason at all. I know that back in the days of brushed motors, two motor setups driving a single prop were not that uncommon, I even made one myself for an "indoor flyer" type of model, it seemed to work pretty well. A buddy of mine had a 12 foot wingspan "Stick" that ran two Astro 40 motors wired in series and geared 5:1 to a single shaft, we called it "MonoKote overcast". That model worked quite well and could stay aloft for over half an hour at a time on the 36 cell x2 parallel nicad pack he had for it, it flew at not much more than a jogging pace most of the time but would go vertical if Jesse gave it full throttle.

A gigantic version of this model:

 

[Fumesucker]

Multiple motors work quite well with IC engines, witness the Summers Brothers "Goldenrod":

 

[recumpence]

I'm not trying to start an arguement here. I am merely trying to save someone time and money.

I totally agree that you would not expect so much loss. But, again, this has been my experience as well as the counsel I have since recieved from others in the industry.

Now, here is the kicker, it is not a direct efficiency loss, per-se. There is something more going one. I will explain it this way;

3 years back I built a twin motored RC heli (Lite Machines Corona with two Mega 16/15/3 motors driven by two Castle Creations Phoenix 25 ESCs on 3 lipo cells). That heli only had a small amount of added power wtih bith motors, however, they ran alot cooler and my flight time only dropped a small amount. There was not alot of heat buildup in anything. The pack was not over stressed nor was anything else. I merely went from one motor and ESC to two driving the same shaft. My power output went up noticelably, but not doubled. But, my flight time only dropped a small amount.

So, as I said earlier, it is not an efficiency issue only, there is also an output issue. For whatever reason, linked motors do not want to put out as much.

It is possible this is mostly a sensorless motor issue.

It is hard to say.

Again, I am not trying to start an arguement, or prove anybody wrong. I am just trying to share some experience.

Matt

 

[Fumesucker]

I don't want to argue either, I'm not sure that your experience really is transferable to what the previous poster whose name I forget is trying to do.

Did you significantly change the gear ratio of the two motors over a single motor in your chopper? If not, then there is no way you were going to get double the power out of two motors.

To get double the power at the same voltage you would have had to significantly raise the numerical gear ratio from motor to rotor blade. If you started at say 10:1 with a single motor you would have had to go to maybe 7:1 with the dual motors (just examples I'm pulling out of thin air).

You will get the same effect with dual or multiple motors on an ebike, the gear ratio will have to be numerically higher for multiple motors in order to completely utilize the power available from each motor.

 

[recumpence]

Understood.

I have gone up in ratio, stayed the same, run more voltage, less voltage, etc. I was really trying to get that twin setup to work well. I have done the same in RC cars as well, and same thing, very small added power. It was definately noticeable, but not doubled.

Now, here is one thing I never tried, but it may be very telling;

I have never tried going from one motor to two smaller motors as a test. It is possible that my applications running twin motors were such that the single motor was puting out plenty of power, but able to handle the job and by adding another motor, I merely shared the load with two motors, rather than really pushing the two. That is possible. However, again, I have done this with more than one application.

The last time I asked anyone was 5 or 6 weeks ago. I spoke with Bernie from Castle Creations about my bike and the blown ESC. I asked him if I should just go with two smaller motors and two ESCs to prevent frying another ESC. His response was--------- "Running double motors typically nets a 25% gain versus one motor. So, that is normally not the best option unless there is no other way to gain more power." At that point, I remembered all my past experience with double motors.

Now, that being said, I agree that my experience may not transfer over well to bikes. The load from a bike is far different than an RC model, no matter what the type. So, it may indeed be worth a shot. I just wanted to share my experience with twin motors.

Lastly, I figure, heck, if I am spending X amount of money on twin motors and ESCs, why not just spend that same amount on one better motor and ESC?

Anyway, that's the way my brain works. :wink:

Matt

 

[Fumesucker]

You put two motors geared together on a dyno and there is no doubt in my mind that they will produce very nearly double the output power of a single motor, if, and it's a big if, everything is optimized for the single motor setup and then the double motor setup.

I have an RC10L with an old style brush motor, 14 turn quad with a Tekin variable current limiting, regenerative braking controller and seven 1700 Sanyo matched low resistance nicad cells. Honestly I don't think doubling the power would be noticeable since the thing is uncontrollable at full power already (at least by me ). When I run it in the quite smooth street in front of my home, by the time I have the throttle all the way open it's time to think about getting on the brakes, I can basically keep the back tires spinning for about 2/3 or more of the battery charge if I turn the torque limiter all the way up.

When you already have more power than you know what to do with doubling it doesn't really help.

Not to mention that doubling the power on a vehicle doesn't even come close to doubling the top speed.