Performance data for several hub motors, and features from vendors
 

I was in the market to get a hub motor conversion kit for my Trek 7500, but found it pretty hard to judge in advance how powerful, how fast, how torquey etc. the various hub motors are. I wrote to most of the vendors on the net (including many on Ebay), and a few of them sent me data sheets for their motors.

I found that all the data sheets I could get, only list full-throttle performance. Since I was mostly interested in 48V 1000W motors, that's what I mostly asked for, though I got one for a 36V motor too. Also, the data sheets seem to only list performance at the higher RPMs, which you would get at 30mph or faster with a 700c wheel.

They were a little cryptic, since they listed Efficiency, RPM, Watts in, Watts out, etc., as a function of torque. So I picked off a number of data points, and re-listed them here, as Torque, Efficiency, Watts etc. as a function of RPM at full throttle.

At the highest speed, that's what you get with no load: Lift the wheel off the ground and open the throttle wide, and the wheels spins freely, without pushing anything in particular.

At less than top speeds, that's what you get on a very slight downhill slope, or on level ground, or going up a long hill, all with full throttle all the way.

These charts don't tell you what the motor's performace is at part throttle, which is actually where most e-bike riding is really done. Your e-bike might be able to go 30mph at full throttle on level ground, but that doesn't mean you keep it blasting along at full throttle every minute. You probably back off on the throttle and go maybe 20-25 mph, or less, most of the time, right?

Still these charts give a hint of what the motor can do, which one is more powerful, which uses more battery power, etc. I'd like to get data for part-throttle performance too, but haven't seen any yet. I may have to build a dyno and test motors myself... with all the spare time and money I have (yeah right).

I also looked thru many websites and Ebay ads, and made a list of what features and toys each one offers with his motor (charger? Battery bag? thumb throttle? etc.).

These charts helped me when selecting a motor (I went with the 48V 1000W from YXM Corp in China, now we'll see how their delivery and after-sale service is). Hope they help you, too. Every little bit, is better than no info at all. Click the thumbnails below to get the full-size chart.

.

They only give full throttle data because any performance data at less than full throttle.....is completely determined by the throttle. :)

Of course.

So what's to prevent them from doing the following?

1.) With full throttle applied, apply enough brake on the test stand, to slow the motor down to, say, 90% of its unbraked speed. Call this full-throttle braked speed, the motor's 90%-braked speed. Then reduce throttle a little, to slow the motor down to 90% of THAT speed, and measure all the torque, amps, efficiency, and the rest. The reduce the throttle a little more to 80%, measure torque, amps etc. Then reduce throttle to 70%, and keep doing that till you're down to about 100RPM or so.

2.) Open the throttle all the way. With full throttle applied, apply enough brake on the test stand, to slow the motor down to 80% of its unbraked speed. Call this speed, the motor's 80%-braked speed. Then begin reducing the throttle to 90% of that speed, measure everything, reduce throttle to 80% of that speed, measure, etc. This is the same as Test (1) above, except the brake is set a little harder and is dragging the wheel down a little more.

3.) Open the throttle all the way. With full throttle applied, apply enough brake on the test stand, to slow the motor down to 70% of its unbraked speed. Call this speed, the motor's 70%-braked speed.

And keep going like that.

You'll probably have to take recesses every few tests, to let the motor and test brake cool off, and maybe to swap batteries.

Soon you'll have a large set of test points, for LOTS of different throttle settings, loads, and speeds. I'd like to try this myself... if I ever have time and opportunity (my standard whine).

If I have a 750 watt motor/controller, all I can ever have is 1 hp.

All other data test points are irrelevant, as they will always be less than 1 hp, and are entirely determined by the throttle up to that 1 hp point.

Sounds like a lot of work for no real comparison data.

Suppose your ebike with 750 watt motor/controller, has a top speed of 25mph on flat ground with wide-open throttle.

If you give it enough throttle to go 15mph on flat ground, how many amps will it use? What efficiency is the motor producing?

The data I described will tell you that.

Well now, here's something kind of weird.

Given several E-bikes with the same weight riders and same voltage batteries running 15mph over flat ground with the same type motors (say brushless hub motors) of different wattage's (say 250 to 750)....they will all be pulling about the same amps.

Could be true. But that's what I'm wondering about. That's part of the reason for these tests.

At full throttle at least, different motors have their peak efficiency at different RPMs.

The black 3-band 48V 1000W motor from luckybbd, has its peak efficiency (at full throttle) at around 36mph.

The black 3-band 36V 750W motor from sannykung, has its peak efficiency (at full throttle) at around 26mph.

What efficiency do these motors run at, at PART throttle, with the throttle backed off enough to run at 15mph on flat ground? The data I have so far, doesn't say - it only gives data for full-throttle operation.

There are ways to determine how much power it takes to push a given rider and bike at 15mph on level ground. Once you know that number, you have to divide it by the motor efficiency to get the number of watts the motor must consume from its batteries.

So, you need to know what throttle setting it takes for your motor to push with, say, four pounds force at 15mph. And how efficient the motor is when doing that.

Same deal for a bigger or smaller motor. All of them can push you with four pounds of force at 15mph on level ground. The luckybbd 48V 1000W motor would have the throttle maybe 1/4 open, while the sannykung 36V 750W motor might have it 1/2 open. And a 250W motor might hav the throttle nearly all the way open.

It's unlikely that all of those motors would be running with exactly the same efficiency, at the different throttle settings each would need to push you with four pounds force at 15mph on flat ground. So some motors would use more watts than others, to do the same job.

That's why I want to get the data I described, for part-throttle operation while pushing a load. But no manufacturer I know of, publishes such data for their motor. They only publish for full-throttle operation, at pretty high RPMs. Testing is necessary to get the rest (the "really usable" part of the data).

One of the reasons nobody has done this is because it takes a lot of time to measure, and it takes a lot of effort to put the information out in a way that makes sense to everyone who's looking for the information. Not that many people are interested in this level of detail in motor information.

That said, I think a lot of people would like to learn what their motor's behavior is like (that is, power, torque, and efficiency) at 65% throttle and 100% throttle.

Less information-overload than giving separate graphs for each 10% throttle interval.

Cerewa, I fully agree.

But that extensive testing and documentation, is part and parcel of designing and selling a motor to people, or even to a manufacturer who is building e-bikes on a mass production line.

Keep in mind I didn't say to make a graph of the raw data you'd get from the kind of testing I described. It would be a complete mess, next to useless, just as the full-throttle data sheets I've gotten from the manufacturers, are next to useless. I rearranged the data from those full-throttle sheets, into the lists at the top of this thread, and now they are a *little* more useful, though very incomplete.

I described ways of taking a lot of raw data, over the entire spectrum of the motor's operation, from very-light-throttle to full-throttle, from slow to fast, and all combinations of those. And at every point, you measure Amps, Torque, and RPM, which then lets you calculate Watts-from-battery, Watts-produced-by-wheel, Efficiency, and Watts-Wasted (becomes motor-heating-up). Once you have that mass of raw data, you then put together some graphs of selected data.

One graph might have four curves on it, showing showing Amps from Battery as a function of RPM, at 25% throttle, 50%, 75%, and 100% (wide open).

Another graph might show Efficiency as a function of RPM, at 25% throttle, 50%, 75%, and 100% (wide open).

Another might show steady speed (on flat ground with no wind) as a function of throttle-position, several curves - one curve for a "typical" 5'2", 100# rider, another curve for a "typical" 5'6", 140# rider, another for 5'10", 180#, another for 6'2", 210# rider, and then a fifth curve for fat guys like me, 5'10" and 250# or so.

That last graph is pretty subjective, of course. There's no such thing as a "typical" rider. Two people who are both 5'6" and 140#, probably won't be the same shape. One may have wider shoulders, while the other is skinny except for a pot belly. They will have different amounts of aerodynamic drag, which is the chief factor that uses up bicycle power at medium to higher speeds on flat ground, so their top speeds won't be the same for the same conditions.

If a tall, heavy guy is considering getting a 36V, 500W motor to cruise his bike at 20mph, that last graph might show him that it's unlikely that size motor would maintain him at that speed, and he needs to get a 750W or 1000W motor, with larger batteries to match. At the same time, a small person might see that a 1000W motor is much more than he/she needs to cruise at slow to medium speeds, and so they would get the smaller motor instead.

As you correctly pointed out, testing and taking the data, is only half the battle, though it's a large half. The other half of the battle, is deciding what data should be presented, and how, to make it the most useful for people who are either buying a bike motor, or designing an e-bike. The purpose here is to AVOID the information-overload you mentioned.

A major use for the final graphs, might be to give e-bike salesman the info he needs to tell a big customer, "No, for the hills you mentioned you were riding, you'll need this 48V 1000W motor, and the 20Ah batteries, just to maintain speed on that long, steep hill, and then get the rest of the way to your destination." If the customer asks him how he knows that, the salesman can get out the appropriate graph. But even if the customer doesn't challenge the salesman, the salesman can be secure in knowing the customer won't bring the bike back two weeks later, snarling that "This damned thing can't even begin to get up that hill after you told me it could".

The salesman can also tell a medium-sized customer that between two bikes, identical except one has a 500W motor while the other has a 1000W motor, both will cruise fine at 15mph on flat ground with him on it, but the 500W motor will have 25% longer range than the 1000W at that speed. (Actually, I don't know if that's how the numbers come out. One of my reasons for wanting these tests, is to find out!)

Yes, testing at part throttle is very involved, and figuring out how to use the data in a way that makes sense, is even harder. But there are a lot of questions people have about e-bikes, that presently can't be answered, excep by buying a bike and trying it... and maybe finding you bought the wrong bike. This kind of testing, won't answer all questions 100%. But it will go a long way toward filling in the present gaps in knowledge.