Understanding motor efficiency

[veloman]

In general, how efficient is a DD hub motor like a 9c? Let's say it's at it's optimal rpm range, not climbing a hill.

Maybe this a bad question, there are quite a lot of variables possible.

For arguements sake, let's talk peak efficiency. I've got a brushed Unite 450w motor, peak efficiency of 84%.

Could we say that a DD 9c is 80% efficient?

 

[dak664]

When I switched from a brushed crystalyte 409 to a brushless 9C front wheel hub I saw a 40% reduction watt-hours on my 11 mile yuba mundo ride (220->130 WH on the charger killawatt meter). I think the main difference was going uphill, the 409 liked >13 mph, below that drawing lots of current without much torque (in fact I melted a brush holder following my wife up a long hill at 10mph). The 9C has more torque at all speeds, don't know in detail how this translates into efficiency.

 

[John in CR]

I think that Unite motor had a dream that it ran at 84% efficiency.

Low to mid 80's is what I've seen on the motor test reports for almost all brushless hubmotors. I'm anxious to try my 9c at 150V to see if the higher voltage/lower current results in greater efficiency like I think it will. I've got my new bearings and one cover balanced, and just need to finish the other cover before trying the mid drive high voltage alignment. The improvements were a must, since it will be running at over 1krpm. Even if it's no more efficient, I won't have to worry about overheating with the outstanding centrifugal ventilation I'll get at high rpms. Plus geared down about 2:1, it should give me both mountain climbing ability and healthy speed.

 

[LegendLength]

I've tried searching the forum for this because I know it must've been asked plenty of times, but I'm just not sure what search terms to use.

I'm trying to understand motor efficiency. I get that if a motor states it's 80% efficient at 2000 rpm then it's going to produce 20% as heat at that rpm. But that is unloaded right? What about going up a hill, where say a hub motor would drop down to 50% efficiency. Is there any way to calculate that sort of thing from published figures or is it trial and error?

 

[amberwolf]

There are simulators that can even show you this in graphs for some motors already tested and such, like the one at ebikes.ca. There is another one by Swbluto here on ES, IIRC, should be a thread just for it around someplace.

 

[Spence /1/]

Am I right in saying that a brushed/brushless DC motor is only 5% efficient at 48V or (30mph) as the power required is about 550W (for a 325lb bike @ 30mph) and that is only 11.5A and looking at other motor performance charts; the motors are only 5% efficient so the actual required power input is more like 1100W at 5% efficiency?

This is non-hub also, which is why it is pulling a lower amount of amps cause it needs less torque.

 

[amberwolf]

Am I right in saying that a brushed/brushless DC motor is only 5% efficient at 48V or (30mph) as the power required is about 550W (for a 325lb bike @ 30mph) and that is only 11.5A and looking at other motor performance charts; the motors are only 5% efficient so the actual required power input is more like 1100W at 5% efficiency?

This is non-hub also, which is why it is pulling a lower amount of amps cause it needs less torque.

If a motor was only 5% efficient, it would be turning 95% of the input power into heat.

So if you had 1100W going in, you'd have 1045W of heat, rapidly cooking the motor, and 55W of motive power, which wouldn't do very much motivating.

If OTOH you had 550W of motive power coming out of the motor, at 5% efficiency, you would get 10450W of waste heat in the motor, quickly smoking just about anything made for an ebike, and possibly turning parts of it into slag. It would take 11KW of input power to get that.


I'm not aware of any motor that anyone actually *uses* in anything that is even remotely as bad as 5% efficiency.


Now, if it were only 50% efficient, then the numbers you give (1100W input, 550W work done) would make sense, but would still be incredibly awful, with half the energy (550W) being wasted as heat in the motor, again quickly cooking it.


Keep in mind also that the current you're talking about (11.5A) is only *battery* current, and doesnt' have much to do with the motor's phase current, in a brushless motor. In a brushed motor, it is directly related to the PWM duty cycle.

 

[Warren]

For the motor guru's:

For a geared motor, is higher motor rpm always more efficient? Two examples of a real motor setup I am considering. A "500 watt" MAC, geared hub motor, run as a mid-drive, on 12 series, 100 Ah, LiFePO4 cells. It will be pulling 350-1200 watts, about 875 watts most of the time, on this setup. A 10 turn motor versus a 6 turn motor. Both would be geared for the same speed. Would the lower turn, faster motor be more electrically, efficient? If so, why, and by approximately how much?

Thanks, Warren

 

[dogman dan]

Well, depends? Either motor could be run through the wrong external gearing on a mid drive thus running too slow, stalling it, and making heat instead of motion.

You'll have to figure out how to gear whichever you choose, and get the motor rpm you want, that is not too slow.

 

[Warren]

As I said, both would be geared to produce the same bike speed/output wattage. I am wondering if the faster spinning motor has lower internal electrical losses.

 

[neptronix]

They would be spinning at the same speed internally.

You'd just be using more amps to get the same RPM on the 6T, and you'd be using more volts to get the same RPM on the 10T.

The difference in the motor is the winding.

If you want to understand this better, go play with the ebikes.ca simulator for a while.

 

[John in CR]

Warren,

I question the ability of that motor to push a bike/trike heavy enough to carry a 4kwh pack. That said, you did mention changing the gearing, so as a mid-drive and geared for low enough speed it will be fine if your power estimates are correct. Geared hubbies are already well matched in terms of motor rpms running in their prime band of operation, so the efficiency difference would be small if any. The decreased copper losses of the faster wound motor would probably be offset by increased iron losses due to the higher frequency. Running the same voltage, the speed wind motor will definitely be capable of higher power. Also, for a given power it will cool itself better due to the higher rpm. With a DD motor and the different gearing you mentioned, the speed wind would be more efficient on top of being more powerful and more able to dissipate heat. The difference in efficiency may be small, but careful use of Ebikes.ca simulator shows it's there.

edit- After a bit more thought, since those geared motors run at 48V without issue, the faster wind motor is quite likely to be better matched for 36V operation and therefor more efficient. Also, because the lower turn count motor is capable of more power, it will be able to accelerate more efficiently. Acceleration is likely to be where the biggest efficiency gain is, but would be difficult to quantify.

Now the hard part, why these statements are true:
Let me start with another true statement- Otherwise identical brushless motors with different numbers of turns in the windings are capable of identical performance (power, torque, rpm etc) and efficiency. There's a nice performance summary of Astro motors that clearly demonstrates this. While it seems to contradict my statements above, identical operation requires voltage and current to be different for the different turn count versions to get to the ideal rpm range of the motor itself. We don't have that luxury with hubmotors. That means for a given voltage, like your fixed 36V, the faster spinning motor is better, because you can gear it down for the speed you want. As you speculated it's all because of lower copper losses.

The statement about better cooling is simply because the convective heat transfer both inside the motor housing and outside will be greater at higher rpm.

I did make some assumptions, that the motor choices have identical copper fill and that losses in the gears are negligible since those friction losses increase with rpm.

Hopefully my wording is helpful and not as clear as mud, because this subject has widespread misconceptions.

John

 

[dogman dan]

Hard to understand stuff, but John usually does a great job of explaining it.

My point was just that any motor could be put on the bike with a too tall gearing sprocket, resulting in stalling the motor a lot. People do this constantly with cyclone kits. They fail to downshift enough on a steep hill, and fry thier motors and gearboxes. Since they are not feeling the effort like they would pedaling, the tendency is to stay in higher gears to climb the hill faster. The motor bogs and cooks.

Whichever winding you choose, don't go and gear it to run at too low an rpm. Clearly you get it, just saying it for the lurkers who read.

 

[Warren]

Thank you, John! I don't understand motors that well yet. But I figured two identical motors, with the same amount of iron, and copper, and the same power going in, turn it all into heat at lockup, regardless of turns. So the quicker, and the further you get from lockup, within the motor's practical rpm range, the less waste heat.

I had understood that I could get this benefit by running higher voltage, and gearing lower, but it had not dawned on me that I could do much the same by changing the turns. What got me thinking about it was seeing that the 100 Ah GBS cells had the lowest weight per Ah of their range, because of less packaging.

Thanks again, Warren

 

[Warren]

dogman,

Yes. That is one advantage of no crank freewheel. I gear the bike so when my legs are happy, the motor is happy. I never forget to downshift, because I don't want to pedal a 75 pound bike from a stop in too high a gear.

 

[John in CR]

Thank you, John! I don't understand motors that well yet. But I figured two identical motors, with the same amount of iron, and copper, and the same power going in, turn it all into heat at lockup, regardless of turns. So the quicker, and the further you get from lockup, within the motor's practical rpm range, the less waste heat.

That's a good way to look at it, though the guys with the slower wind motor would counter with the true statement that those motors get to their prime rpm range at a lower rpm. One main differences is the shape of the power and efficiency curves, which are broader and overlap better with the speedier wind motors. The other part is the greater power of the speedier wind version at that voltage. What you're planning is the equivalent of a comparison of running a 10t in a 26" wheel and a 6t in a 16" wheel, and the performance comparison is no contest with the same top speed, but a big difference in how you get there.

You wouldn't think it, but more power on tap can actually result in greater efficiency. The crazy high power bike I've had on the road for the past 2 weeks ends up with better or the same consumption as the cargo bike I've had for years over my normal routes as long as I restrict it to the same cruising speed. That's with high power launches every chance I get. The new bike's motor isn't any more efficient in terms of peak efficiency, the regen % is a lot lower, it's a much more upright ride so less aerodynamic, and the hard launches result in a significantly higher average speed especially due to the numerous short sections where the cargo bike never gets up to speed. For the amusement park type ride to use the same energy as the relaxed cruiser on the same routes despite those handicaps can only mean that the shorter time accelerating has to be significantly more efficient.

Admittedly, it is hard for me to let off the throttle so soon for the same cruising speed, so now that I'm more comfortable with the power my cruising speed has gone from around 35mph to more like 45mph. It's not that I like going faster, it's more that I love accelerating around cars. Going from following a car at his speed to being past him in an instant is priceless, but costs maybe an extra penny per ride in electricity. I can only imagine how much fun it would be to play in traffic with LFP's bike.

John