Jump to content

Understanding motor power better (ELI5)


Recommended Posts

Hey everybody!

Beginner question, and sorry if I didn't search enough and the answer is already out there.

I'm trying to figure out what matters when it comes to motor power. The King Song KS14C I'm testing advertises a motor that gives 500w continuous power and is advertised as having 3000w peak output. I just want to understand better what this means in the context of what you guys know is out there.

In particular, I'm wondering what kind of advantage the 3000w peak output might have, or if it just doesn't matter. My hope of hopes is that it might help protect against one of the many kinds of unexpected shut off we're working on? <fingers crossed>

I know you guys understand this better than I do. What do you think?

Link to comment
Share on other sites

Unfortionately it tells you just nothing - but it sounds good...

It depends mainly if just the motor can (theoretically?) deliver this peak power or the whole system (including battery pack and mainboard) and if how often and how long....

So in the best case this very nice power figure could be used to handle bumps, sudden accelerations, etc and increase the felt performance and the overall safety. In the worst case its just a marketing number. The power which the motor could deliver for a short period of time - but if this time is to short, not often enough usable per time intervall or more sadly not beeing able to be delivered by the mainboard or battery pack its just nothing....

So for the answer you would need to compare it with different other cycles with 500W continous power and less peak power and "see" the difference or get a trustful contact to a development engineer of kingsong, i fear.

Beeing a realist it assume, that this number means nothing, since a customer cannot (edit) easily prove it. And especually cannot get refund for the cycle, if it is not usable but just marketing...

Regarding cutoffs it has no relevance. A high figure for battery back peak power output would be good for this (edit: including mainboard and bms)

Link to comment
Share on other sites

As Chriull says the peak power number on it's own means very little unless you know it is actually being delivered for enough time to be useful. It could help if it is available for as little as 2 or 3 seconds as that would be enough to recover from hitting a large bump or pothole.

Hopefully if Jason McNeil gets his test equipment up and running we may be able to test the claims in the near future.

Link to comment
Share on other sites

While I can't say I have wrapped my head around all the things that affect how a DC motor works and behaves, I do have one real world scenario that happened to me, and probably would have lead to a face plant with a lower maximum rated power or batteries/mainboard that can't provide enough power to the motor:

I was riding the Firewheel maybe a week or two after getting it around one of the bikelane-routes that I frequent now, and there's a short but fairly steep uphill. In the middle of this hill is some sort of asphalt-patch, that sticks out somewhat, so there's sort of an edge to it, and it's causing some humps in the hill. I had lots of speed coming to the hill, to ease with getting up, and as I was climbing up, I hit the edge at fairly high speed (no idea how high, I'd say nearer 20km/h maybe, I think the first warning starting at 18km/h was playing at the time). The wheel tilted forwards A LOT really fast at that point, and I nearly shat my pants. But, and I think this is where the maximum (very short term) motor power steps in, it didn't shut down, fall down or anything like that, but just corrected the position back to upright almost immediately and I continued uphill without even losing much speed. No idea how much power that required, but considering I was already riding at fairly high speed, going steep uphill and the wheel needed to correct a fairly large tilt (I was riding in sports mode, so it keeps the pedals level at all times, except apparently in situations like this), I'd expect the rated 550W(?) power alone couldn't do it, considering that when we were making some physics calculations before, moving my (m)ass uphill a 10 degree slope at 10km/h steady speed without acceleration needed AT LEAST around 350W, and that was without any air drag, rolling resistance or extra balancing...

Btw, this post has a list of motor powers by dhlee3, I have no idea if they're correct or not: http://forum.electricunicycle.org/topic/264-speed-vs-torque-margin-of-safety/?page=2#comment-3205

Does anyone know what the motor size is for the Firewheel? I have an F260 and contacted them to find out what the motor size was and they responded "Rated power 550W, maximum output power 1350W".  I'm not sure how to compare that to other glidewheels - like the ones below. Do all manufacturers have a rated power and a max output power?  Is the value they publish the rated power? 

Brand & ModelMotor (Watts)
AirWheel X3400
GotWay M0400
GotWay M10 260400
GotWay M10 340400
Ninebot One C450
Ninebot One C+450
AirWheel X5500
GotWay MCM 14500
Ninebot One E500
Ninebot One E+500
Firewheel F779550
AirWheel Q1800
AirWheel Q3800
AirWheel Q5800
AirWheel X8800
GotWay MSuper1000
IPS181 Lhotz1000
IPS191 Lhotz1000
IPS192 Makalue1000
Solowheel Classic1500
Solowheel Extreme1800
Firewheel F132550?
Firewheel F260550?
Firewheel F528550?

It's also the only source where my "information" that Firewheel has 550W rated / 1350W max motor comes from  ;)  I suspect that at least some of those numbers in the table are maximum output and not rated (I don't think Solowheels are RATED at 1500W/1800W for Classic/Xtreme).

I don't know if the 1350W max for Firewheel is correct, but at least the batteries on F260 and above should be able to provide it, as the Sony US18650V3's have 10A continuous discharge current, don't know the pulse discharge, maybe 15-20A?  On F260, 16S2P =>  2 * 10A * 60V = 1200W continuous, higher temporary pulse current -> temporary max must go over 1350W from the batteries, whether the mainboard/wirings can handle the currents and whether the motor is really that high max rated is another story... ;)

Link to comment
Share on other sites

What I'm getting here is that the numbers mean nothing until someone who knows more than I do has smashed the thing to pieces, or at least done some tests. It takes all quality parts to make the numbers really matter, not just one.

But if everything is quality and working together, a higher peak wattage might give you a chance in those sudden bumps on steep hills. Eh?

Link to comment
Share on other sites

Higher peak power is very important for EUC's to compensate for sudden balance issues like the high-speed bump Esaj was talking about. Of course, just having a powerful motor is not enough, the whole system must be able to provide that power.

I was talking with Shane Chen this weekend (the inventor of the Solowheel) and he mentioned how the Seqway's have about 3000W power (!) but when they tried to play Polo on the Segways, the players would still often fall -- when they reached out for the ball, the motor often just could not compensate for the severe leaning forward. As far as I understood, during testing it seemed that for a EUC, 3000W is very robust but they cant make that yet due to the weight etc. Also, to really always balance a 90kg person even if you severely lean forward or take high-speed bumps, you need about 10.000W peak power (!).

Anyway, long story short: the more peak power the wheel can deliver, the safer you are B)

Link to comment
Share on other sites

More or less, that's it ;) But there also should be experienced people who can just answer your question - at least at the development departments of the euc companies should be some...

From the discussions about the batteries available and used in some wheels, the 3000W peak power should not be a problem.

If you want to try if the kingsong can bring the 3000W peak power on the road get someone who weighs together with the kingsong about 90 kg, drive 20 km/h on a flat road and go on a ~30° slope. If the kingsong keeps the speed of 20 km/h the peak power is feasable.

Would be interessting what happens the way up the slope - How long the the kingsong can keep the 20 km/h until it either just gets slower (at 3-4 km/h it should just need its 500 W rated power), or starts smoking/burning/ falls apart ;)/shuts down.

Just from the momentum the euc should come up to a height of ~1,5m. So thats about 3 m up the slope.

The transient from the flat to the slope should not be to hard, so the euc has a chance to deliver the power.

And maybe someone should take a look at these numbers, before you send a crashtest dummy on this mission. I came to about 80 Watts for the airdrag, ~100 Watts for the rolling friction ( 90kg, 20 km/h) and ~2700W to go up the slope...

Link to comment
Share on other sites

While when it comes to motor power, "bigger is better", why it's hard to say much with just something like the stated rated / maximum power of the motor has many reasons, this is based on what I've understood of the motors so far:

Different motors have different powercurves, here's some example curve from the MIT DC motor-page:

This is just some small dc motor, as you can see the, the peak power is only little over 5W. The torque drops linearly (in ideal case, real world might be different) from the full torque at "Stall torque" (maximum torque, but the motor isn't actually turning at all at this point, will burn the motor if used for longer while), to torque disappearing entirely at "no-load RPM" (no load would actually mean that there isn't anything attached to the motor). Still the maximum power is reached around the center of the torque-line. For definitions of these terms and others, see for example http://www.robotshop.com/blog/en/how-do-i-interpret-dc-motor-specifications-3657

Stall torque: This is the maximum torque a motor can provide with the shaft no longer rotating. It is important to note that most motors will sustain irreparable damage if subjected to stall conditions for more than a few seconds. 

No load RPM: This is how fast (angular velocity) the final output shaft will rotate assuming nothing is connected to it. The motor’s RPM is proportional to the voltage input. “No Load” means the motor encounters no resistance whatsoever (no hub or wheel mounted to the end). Usually the No Load RPM provided is associated with the nominal voltage.

To make it more difficult, even two motors with same rated and maximum power can be different, if you think of the different versions of 18" Gotway MSuper, there are three versions: high torque, medium and high speed. While (I think) all the three motors have same rated and maximum power, the torque and max speed characteristics are different.

To add to the injury, if the wheels use "standard" off-the-shelf -brushless DC motors, the power curve/torque/max RPM -characteristics might not be optimal for the use. Even if the motor could theoretically run at very high RMP (high top speed), it's of no use in the wheel, if the torque isn't enough to move the rider that fast. If the wheel also has limited max speed (pedal tilt-back or otherwise), it could stop the motor from accelerating while it could still run faster, in which case a motor with more torque / less max RPM should have been chosen.

Then on top of it all, come the varying movement resistances affected by rider weight, uphills, air drag at high head wind etc, and finally, to be really able to get "everything" out of the motor, all the other parts need to be able to give out the power all the way to the motor. So there must be enough current output from the batteries (more parallel packs => higher current output, as the load is "shared" by multiple series of cells), the wiring must be able to withstand the current without melting, and the mosFETs in the mainboard must also be high power-rated and cooled enough (with heatsinks or otherwise) to not burn under high current. Some of the power is lost (mostly as heat) even before it actually gets all the way to the motor. The maximum power of motor and pulse current from batteries are always meant to be used only for short burst, not continuously, as they heat up the parts and can lead to battery cell failing/wires melting/electrical components on mainboard burning and/or the motor coils or other motor parts burning.

Also, I have used the nominal 60V voltage for calculating the power output from the batteries on the Firewheel, but the sad reality is that when the voltage starts to drop under high load, something has to give: either the output power must come down, or the current must go up to keep the power output the same, just as an example:

1200W / 60V = 20A
1200W / 56V = 21,43A

So if the batteries couldn't give out 21.43A when the voltage drops to 56V, but "only" 20A, the output power would drop:
56V * 20A = 1120W

I haven't found anything matching the Firewheels motor from DAAO Electric's standard catalog (http://www.daaomotor.com/html/Motor/), so it would seem that the motor is purpose-built for it (hard to say if it's optimal even then, there isn't much info on the pages about the motors). DAAO seems to build motors for at least Firewheels, (some) IPSs, generics and probably older Gotways (the AMK-marked motors seem to be DAAO).

I still haven't had the chance to test any other wheels than my own, so I can't speak much from experience. There is a huge difference in power between the 14" generic (350W/xxxxW?) and 16" Firewheel (550W/1350W?), based on riding experience.


And maybe someone should take a look at these numbers, before you send a crashtest dummy on this mission. I came to about 80 Watts for the airdrag, ~100 Watts for the rolling friction ( 90kg, 20 km/h) and ~2700W to go up the slope...

Going by similar simple equation which I used before (no resistances taken into account, plus probably the balancing uses some extra power):

m*g*sin(theta)*v  (mass * gravitational constant * sine of the hill angle * velocity):

=>  90kg * 9,80665m/s2 * sin(30 degrees) * 5,5555555... m/s

I got around 2450W just for moving constant 20km/h speed up the 30 degree hill for 90kg. So your numbers sound correct, assuming my simple equation is correct. ;)  If you know the air-drag and friction coefficients, and rider surface-area against head wind (which might be 0, but there's still airspeed due to groundspeed), I found this equation that should probably give more accurate values (http://www.sportsci.org/jour/9804/dps.html), it's again for cyclist/bikes, but same principles are at work even when riding an EUC with electric motor:

W = krMs + kaAsv2 + giMs

where W is power, kr is the rolling resistance coefficient, M is the combined mass of cyclist and bicycle, s is the bicycle speed on the road, ka is the air resistance coefficient, A is the combined frontal area of cyclist and bicycle, v is the bicycle speed through the air (i.e. road speed plus head wind speed), g is the gravitational acceleration constant, and i is the road incline (grade; however, this is only an approximation, as the sine of the road angle to the horizontal should technically be used).

Link to comment
Share on other sites

  • 2 weeks later...

The table of unicycle powers in previous posts is not correct.

We should make difference between maximum power and nominal power. Maximum power can be delivered instantly, and nominal power can be delivered continuously without heating of the components. If maximum power is delivered constantly, you would damage your system. Since the users have no knowledge about electricity, some producers can show maximum power where other producers show nominal power which makes a great difference.

Nearly all unicycles have NOMINAL power between 300-500W which matches with a MAXIMUM power between 900-1500W.

Only Gotway Msuper produces 1000W nominal and 3000W maximum power unicycles. Other unicycles are not able to do that because the wheel must be bigger than 14-16" to put a big 3000W motor in the wheel. You know: unicycles have in-wheel motors, so the bigger the wheel, the bigger the motor power. It is the Ninebot factory responsible himself who has revealed this information to me. I had asked for a more powerful ninebot one. He said that it is the biggest motor (=500W nominal=1500W maximum) that they can put in a 16 " wheel.

Ninebot, Solowheel, IPS121 announces that their unicycles are 1500W (=maximum), however their power is the half (!!!) of the power of 1000W (=nominal) Gotway Msuper 18". Unicycle with 1500W NOMINAL (!) power does not exist. This is the MAXIMUM power.

Link to comment
Share on other sites


This topic is now archived and is closed to further replies.

  • Create New...