Struggling to understand consequences motor rpm

[knutselmaaster]

Hey everyone,

I am trying to figure out the consequences of different rpm versions of motors.
Besides the obvious higher speed means lower torque I would like to understand what happens with power consumption and efficiency.

For example
If I have two versions of the same motor that are both capable to run faster than 25km/h and I restrict the speed to that 25km/h would there be a preference?
Assuming I don't use full power to reach that top speed but an equal 100w for instance, there would be no (noticeable) difference in consumption when riding at (25km/h) top speed, would there? (assuming riding on flat terrain)
What would be the influence on overall power consumption in this case?

If I let a motor run at almost it's max speed, the power it can deliver will be close to none, but is the motor still consuming a lot and generating heat, like on too low speed?

Would it be more efficient to have a low rpm motor and use the controller's max A setting to lower torque to the level of the higher rpm motor (assuming that this is sufficient for the situation) than using the higher rpm motor with speed restriction?

In general, would it be better to take a motor rpm just on/over the max speed limit that will be set on the bike, or a higher rpm one where the less efficient highest speed will certainly not be reached ?

I have read and searched a lot and I think a simple answer does not exist because a lot more factors are in game here, but maybe someone has a satisfying "general" answer, or guidelines ?

Thanks

 

[arkmundi]

Good luck trying to understand it. I followed along with the Science, Physics, Math, & Myth thread for awhile. That dialogue was started to consider what was the best motor turn-count on a MXUS, higher or lower, 3T versus 5T. Whether the speed versus torque consideration was myth or fact, among other things. If you come to some meaningful theoretical conclusions, let us know.

When I bought a MAC motor, I chose a 10T believing it to be a best compromise between speed and torque. Based on information provided at EM3ev.com on choice question. When I bought a second motor, after further consideration of this thread, I chose an 8T, supposedly a higher speed, lower torque motor. I was curious. I got higher speed, but no noticeable loss of torque. I'm heavy, lots of hills and carry stuff. A loss of torque would be noticed.

I'm of the opinion that the amount of copper in a motor's winding matters most. Rather than turn count, maybe the weight of a motor, all other things equal.

The MAC, loaded at 36V:
6 turn high speed motor - 400rpm
8 turn standard motor - 320rpm
10 turn torque motor - 255rpm
12 turn low speed extra torque motor - 200rpm

 

[Malloot]

Your case is actually quite simple, you just need data to know how much it will matter.
lets say that you have a 26 inch wheel with a 2 meter Circumference. Making 25 km/h about 208rpm

Lets say motor 1 is the high speed one, it has a efficiency of 85% at 400 rpm loaded and a 75% efficiency at 200rpm
motor 2 is the lower speed one, it has a 85% efficiency at 300rpm and 82% at 200 rpm

we take it that keeping you on 25kmh costs 500 watt of power on the road, so its costs the battery (we have a 100% efficient controller) 500*1.25=625 Watt with motor 1
With motor 2 it would be 500*1.18=590Watt

So you win 35 watt for going with the slow motor

What you want is a accurate table of efficiencies at a loaded scenario comparable to a bicycle riding at 25kmh, and that is with the mostly Chinese stuff we are dealing with rather hard to get accurate data for. We can assume it is accurate but i know from experience that testing a electric motor for efficiency properly is not easy even for a bunch of uni professors.

 

[spinningmagnets]

I chose a 10T believing it to be a best compromise between speed and torque. Based on information provided at EM3ev.com on choice question. When I bought a second motor, after further consideration of this thread, I chose an 8T, supposedly a higher speed, lower torque motor. I was curious. I got higher speed, but no noticeable loss of torque

Thanks for posting that. I am of the opinion that the biggest difficulty in comparing someone elses experience to a potential purchase choice is that...my hills might be steeper than your hills. If you take two identical 48V E-bikes, with the exeption that one is an 8T and the other is an 10T. Run them side by side (with no pedaling) until they are at the normal amount of warmth. Then take them onto an uphill that is very long. If the hill gets progressively steeper, there will arrive a point where ther RPMs begin to noticably sag, and then...the controllers will add more amps in order to 'try' to get the RPMs up.

As long as the RPMs on the 8T remain high enough that the amps are low enough that the 8T doesnt overheat...monitoring the temps will show that the 10T will be slightly cooler. If the rider with the 8T has a flat stretch of highway where they want to get around car traffic, I am certain they will really appreciate having the extra top-speed the 8T provides. But...if the hills around OP are steep enough that the 8T will overheat, then the 10T would be a better choice (in the absence of serious added pedaling, or cooling mods [ventilation/oil]). The difference would be much more pronoiunced if we were comparing a 6T to a 12T, and not as much a difference for the 8T/10T here, but...there is a difference.

I think motor temperature monitoring is the single most important thing that is currently missing from 'most of' the E-bike world (it is an option for em3ev.com, ebikes.ca, and Cromotor, but very few others). the reason I say this is...if a motor is getting too hot on your commute, you can sell it for "something" and buy a more appropriate motor...as long as it hasn't fried. If you fry it, you're buying a new motor, then waiting 3 weeks.

 

[knutselmaaster]

Thank you all for your replies, it gives me something to read and a lot of food for thought.

I am not deciding for myself, that was rather easy, namely as fast as possible
I want to be able to find the right motor for each situation.
Given that most of my clients will choose for a fully EU road legal setup, all information I absorbed over time makes me conclude

-Choose the motor that can run a tiny bit faster than 25km/h (or other top speed chosen) based on wheel size and restrict its max A if the client is light / loves pedaling / rides only flat terrain
Do not restrict its max speed (as it cannot go much faster anyways)
This will result in less power consumption and less heat production, with descent power and smooth cutoff.

-Do not choose the fast motor (with fixed max speed), adding A to make torque acceptable, as there will be a large portion of the low speed that will be inefficient and the added A will be transformed in heat more than in power, giving bad battery life and risk of overheating.

(on geared motors)
The gears are of more influence than the windings if one looks to get higher torque.
Ex a q100 201rpm will have less torque than a q100h 260rpm (33% more rpm 25% reduction difference, still more torque)

Please correct me if I'm wrong...


It would be nice if controllers could be set to have a minimum speed, so that it will be impossible to get in a situation where you can overheat the motor.
On my high speed bike I use an Arduino solution to prevent high power demand on low speed, so that I don't have to worry when I am forced to drive slowly

 

[arkmundi]

The advert of EM3ev MAC motors is that they are "high torque". Supposedly, the motor magnets & windings are such to turn it optimal efficiency. Then the torque is geared out to hub. I can attest to the design, make and acceptable levels of speed & torque.

What is lacking in the industry are standard measured torque. We know what torque is and how to measure it..

Its just that it is not standard practice for the industry to report it, at least not that I've seen. We've come to expect that the motor rpm for various wheel sizes are reported and would be skeptical buying if not. But torque is just as important.

 

[knutselmaaster]

In the most cases here, torque is the only important factor, as speed is limited and hills are steep.
It would be great to have an industry standard for motor properties indeed.

 

[d8veh]

As a general guideline, choose a motor that can go about 30% faster than your cruising speed. If your bike is limited to 15 mph, presumably you will want to cruise at that speed. 15mph is 201 rpm in a 26" wheel, but that's at nominal voltage, say 36v. Your actual average voltage would be closer to 38v, depending on how deep you discharge the battery. I would therefore go for a motor somewhere around 240 to 260 rpm as a good compromise between power, torque and efficiency.

Highest efficiency comes at about 3/4 up the rpm range, where you've probably already passed the point where current is ramping down. Any higher rpm than that and you can only use less current, so although efficiency of the motor is less, the efficiency of the vehicle in terms of miles per wh will be higher. A 201 rpm motor will always use less wh per mile than a 260 rpm one, but you'll get less assistance above 12 mph.

Reducing the throttle changes everything with a speed control controller because it effectively converts a high rpm motor into a lower rpm one. I'm still trying to figure out what happens with current control. I think it's less effecient than speed control for our type of riding.

It's different with a higher power motor, where it's difficult to get speed above the point of peak power because of the wind resistance at the higher speed.

 

[fellow]

When I choose the motor, I always start from the controller. I choose the one that is easiest to hide, repair and is cheapest. 6-fet controller for $20 is the obvious choice. When mosfets die, it is easier to desolder n mosfets than n*2 (Instead of n, insert your minimal number of mosfets here that can satisfy your power hunger).

This forces me automatically into 2 other choices. First, I need the motor that gives me the most Nm per A invested, as 6-fets are genereally weak. As of 2015, their limit is about 25A when used with hightech mosfets (Snickers pushed them to 53A just because he can ). That leads me to choose lowest KV motor that I can buy. This low amp thinking has several other positive sides - the main switch is in grams instead of kilograms (DC current is tricky to brake), fuses can fit into proper 6mmx30mm 125VDC standard (max 30A!, no dangerous car fuses here), cables can be made lighter, cheaper and everything can be hidden inside the frame...

Crap, max speed is 25km/h at 36V and the power is too low! That is easily fixable, I just use the highest voltage I can fit without killing myself. It can be 48V, 60V, 100V, or the number you are comfortable with.

Fixed voltage vs fixed current thinking:

vs

We have a lot to learn/steal from the RC people (article by Jim Bourke): http://www.rcgroups.com/forums/showthread.php?t=337735

As the above table demonstrates, the efficiency increases as voltage increases. Our model predicts that increasing voltage will always increase efficiency (up until the point at which we hit the RPM limit). A quick look at the power out formula will explain why:

Power Out = (V - Iin * Rm) * (Iin - Io)

As you can see from the formula, the input voltage is not factored in to any of the losses within the motor. The higher the voltage, the lower the losses in all cases.

And the legendary:

In addition, consider that the motor limitations cannot be exceeded. If they are exceeded, we can use the following equation to calculate output power:

Power Out = 0

So if we exceed the motor limitations our model predicts that we will get a big, fat zero for power coming out of the motor. In fact, once we exceed the limitations the motor is destroyed and will forever produce 0 output watts.

And Jim Bourke's conclusion, and I quote:"
*Power is a representation of how much work can be done in a given time.
*One horsepower is the same as 746 watts.
*Rotational power is expressed in terms of torque and RPM. Torque is specified in Nm and is a measure of how much "oomph" is behind the motor's shaft rotation.
*A motor's Kt value specifies its ability to produce torque for each input amp. The torque constant, Kt, can be directly derived from the Kv of the motor. It is impossible to increase the torque constant without proportionately decreasing the voltage constant of a motor (and vice-versa).
*A motor does not produce torque perfectly. There is a torque loss equal to the no-load current value of the motor. The no load current is symbolized using "Io".
*A motor is subject to two primary limitations: the RPM limit and the Torque limit. There is no power limit, per se.
*Output power can be computed by taking the RPM and Torque losses into account. Pout = (Vin - Iin * Rm) * (Iin - Io).
*Motor Efficiency is calculated by dividing Power Out by Power In.
*Increasing voltage always increases motor efficiency up until the point where the RPM limit is reached (when the efficiency drops to zero immediately!).
*Increasing current increases the RPM losses and decreases the torque losses.
*The point at which the RPM loss equals the torque loss is the maximum efficiency point of the motor."

 

[knutselmaaster]

Great stuff!
I am happy that I started this thread
Thank you all for your insights.

I didn't think about different voltages yet, I had just decided that 36v will do a better job than 24v and that 48v would be too much (bigger, heavier).

So a 36v 201rpm motor @ 48v would run better (more efficient) than a 48v 260rpm motor @ 48v?? (looking at it, they just might be the same motor??)

The power that is showing on the lcd, is the raw power coming from the battery, right?? (controller input V*A)
Or is it controller output power (losses in controller not included)

 

[d8veh]

The LCD shows the power through the controller's shunt/s, which is basically volts x amps from the battery. There's a few watts to power the controller that don't go through the shunt, so are not included

 

[knutselmaaster]

Seeing how my controller gets hot, it seems to be more than just a few watt that get lost in the ozone layer

 

[dogman dan]

When they say the low rpm motor is higher torque, they are lying. But the lie is true, under certain specific conditions. If what you sell meets those conditions, then it can make sense to tell this little lie.

By now, you might be understanding why this lie is told. Just try to explain the real truth to a guy just looking into an ebike. Hard enough to get amps vs amp hours explained in that first conversation. You simply have to dumb it down some. What I do, is try to get info about what the guy needs to do, what he weighs, etc. Then I either tell him our "high torque" motor kit will suit him, or I tell him it won't. If the guy really needs 2000w, I'll tell him that.

But here is a truth. If you need torque for some typical situations, AND you won't be running 2000w or more, then the low rpm motor is a better choice for heavy riders, cargo bikes, or off road use with normal weight but steeper hills. The slow motor will run in a more efficient rpm at a slow speed, than the fast motor will. Torque may be very close to same, but motor heating will always be very different, with the fast motor running hotter and less efficient.

Both motors will pull almost the same torque, if the same wattage is flowing and rpm is not too slow. After all, the magnets are same, the copper fill is very similar, and gears if any, are the same.

But the catch 22, is the higher rpm motor will pull more watts for a longer period. So it could have more torque. (would with more power on both) Often it does not though, because the reason the low rpm motor will pull less watts measured at the battery, is that it reaches efficient rpm sooner. This is why a heavily loaded slow motor running 1000w will accelerate better than a fast motor with the same load and wattage. On a hill, it will be slower, but the rpm will be less close to stall, and until the hill gets too steep for the 1000w, it will climb a lot cooler. If the load is not so heavy, it can be very hard to measure any difference between the two. If both run efficiently enough, both will perform very similar when running the same speed.

The key thing with a fast motor, is feed it enough watts. Then it stops stalling. Same motor will perform a lot better if you give it 2000w. (assuming the motor can take that for long of course) On 1000w, a fast motor with a 190 pound rider is just stalling for the first 20 feet. But with 2000w, the same motor barely stalls at all. It just gets up and goes.

To actually change your torque, you change the gear ratio. For a hubmotor, the only way to do this is make the tire smaller diameter. Those really interested in good torque run a 20" rim instead of a 26"

Well, once again, that explanation sucked. If this then that, but if that then something else. But that is just how it is. If torque is what you crave, it's simple enough. Motor that easily takes 2000w, smaller wheel, now you got plenty of torque baby. 8)

If you are stuck with the same motor, and the same 1000w, then a 20" wheel will give you a HUGE improvement in torque.

 

[knutselmaaster]

Thanks for your reply

It makes me laugh (respectfully) to see the difference in approach between USA and EU, it is like with cars.
You just make it bigger, and put in huge power, while we (forced by legislation) try to squeeze the most out of our small motors and batteries.

Still, this difference in approach is very good for me to help me see things from another angle.
Please don't stop

 

[fellow]

So a 36v 201rpm motor @ 48v would run better (more efficient) than a 48v 260rpm motor @ 48v?? (looking at it, they just might be the same motor??)

(48V/36V)*201rpm=268rpm at 48V. It is the same motor, allowed for measuring error of 3%. Efficiency is about the same.

 

[knutselmaaster]

But shouldn't a 48v motor (with the same amperage) be "sturdier", with thicker wire for the windings ?
If it is just the same motor, why do they all sell them as different models?
Just marketing ?

 

[fellow]

Are you refering to Q100H or some other motor? 48V 260rpm Q100H does not exist. There is a 36V 201rpm, and 36V 260rpm if we are talking about the Q100H?

Yes, you trade Nm per A for sturdiness (in absolute A, not Nm), and torque for speed. There is no free lunch there. Low kv motor gives you most Nm/A for the controller you own, high kv motor gives you most km/h per volt for the batery you own.

By choising the kv, you choose if you prefer to overheat your controller or motor. The ideal is both .

 

[knutselmaaster]

I was not referring to a particular model, I was just wondering if motors sold as 48v are built differently (and should thus be heavier) from 36v models or are just adaptations in number of windings.
If it is the latter, one should be extra careful regarding overheating.

 

[fellow]

Insulation voltage is usually high, but they expode at extreme high rpms. Never heard of high voltage sparks destroying them because of insulation breakage. High current melts them. Heat is a problem, liquid cooling with external cooler is prefered if possible (especially if you live in Val Thorens ).

ATF cools 4 times better than air, and is lubricating and anti corrosive.
50%/50% antifreeze cools 6 times better than air.
100% pure water cools 12 times better than air.

 

[knutselmaaster]

I must be misunderstanding..
Current has more influence than voltage on heat production is what I read...
But isn't my calculation right:
Efficiency 80% = 20% heat
36V * 10A = 360W gives 64W heat
48V * 10A = 480W gives 96W heat

 

[fellow]

It is better to measure losses in Watts than in percent, it confuses less. Current makes loss, voltage helps efficiency. We have to look at it as the complete system, not only motor side. 100V*3.6A=360w, 90% efficiency, only 36W heat. Same power as 36V, less heat.
200V*2A=400w, 95% efficiency, only 20W heat.

That, AND you have 6.4A*100V=640W more in reserve that you don't have in 36V example(lets say that motor burns up at 10.1A and doesnt explode at high rpms).

It simply makes less heat at 3.6A than 10A in the controller, cables, switch, connectors and so on regardless of voltage.

 

[knutselmaaster]

Si I should correct
36V * 10A = 360W gives 64W heat
48V * 10A = 480W gives 96W heat
To
36V * 10A = 360W gives 64W heat
48V * 7,5A = 360W gives 54W heat(64W heat - the difference in efficiency between 10A and 7,5A, assuming 80% @10A and 85% @7,5A)
Will that result in exactly the same torque/behavior on the same motor ? (without counting for the higher possible rpm)
Is the difference in efficiency that big between 7,5A and 10A? (is the 10% difference between 3,6A and 10A from your example realistic ?)

 

[knutselmaaster]

/Double posting intermezzo

I've been working with electric motors for all of my life, in many different applications, as a tool/machine mechanic.
But the ebike approach is so different, I have to rearrange everything I've learned.
So fascinating, so cool!

 

[fellow]

No, I think 10% is unrealistic. I've used it as proof of concept. Take a look at that RC motor, constant current picture above. It gained double power and +3% efficiency at double voltage. It's a small motor, but it's better than nothing.

You have to trade rpm for Nm. This only works in geared motor like q100h 201rpm or mac 12T. Otherwise you have gutless motor at low rpm, and nervous wheelee monster at high rpm that is too fast and self-destructive to be usefull. This is kind if F1 thinking, cooling is essential. No burnout power, but it catapults one close to redline. If you need raw burnout power in a hub, you need big motor, big controller, big battery and high amps. The cost: increased weight. This is what our US friends do. In europe, we end at mental institution with anything bigger than Mac in our bicycles.

 

[knutselmaaster]

I'm going to play around with the ebike calculator to see if it helps me visualise things.
The curbs one by one I understood but after this very instructive discussion I should now be able to get a more global picture where I understand the influences and interactions better.
Too bad that there are so limited geared motors on there...

Thanks 1000x for all answers.
I think this discussion will help more than just me