I know that the actual kV of motors is not necessarily what they might be labeled as but I’m wondering in practice how far off it can be and if anybody has had a case where it was really off. I recently upgraded a board from 6384 190 kV and 5.5 gear to 63100 170 kV and 3.45 gear.
With the old setup I was able to max out at about 57 km/h which is in line with calculations. With the new setup I was really expecting to be able to get over 70 km/h but instead I’m only able to get to about 62 km/h which actually works out about right for a 140 kV motor.
From Metr and Davega data the duty cycle hits the low 90s at times and the max battery and motor currents are not getting anywhere near their set limits. For example on a recent test ride the max speed was 61 km/h at 92% duty cycle and the battery current was only 18.4 A per side and motor current 22.9 A per side.
I’ve changed nothing else and the only thing I think of is that the kV of the motors is just not correct.
Could I be missing something else?
I saw a video on Pavel Garmas’s channel showing a way to check the actual kV of motors but it looks a little involved and I would have to remove the gears from the motor shafts.
Yeah I saw a video like that. I may have just have to do that. Unfortunately I would likely have to remove the gears off the shaft to do that which is a small pain as everything is all red loctited and the grub screws were a bit of pain to get in there.
As it is, I’m pretty happy with the performance but just somewhat surprised at the numbers.
You can calculate KV by revving your motor to max rpm and recording the measured erpm in vesc tool, and plugging it into a formula. (ERPM / 7 Pole pairs / (Battery voltage) = KV. Add 5% to the final KV result to compensate for vesc’s duty cycle limit.
Actually I’m on gear drives. This was moving between two different ratios of the Newbee drive (and changing the motors). Fortunately they only recommend red loctite. Breaking that is fine but the grub screws are a real pain to get in. As a matter of practice from now on I guess I should use that technique of checking the motor kV before installing it just to see what it is.
BLDC has a higher top speed than FOC… reasons are complicated. Mostly due to BLDC being able to sustain it’s peak voltage for longer each second due to the non-sinusoidal waveform. Other things to try: increase the battery current limit (if safe). 5% of the top speed loss might be caused by the 95% peak duty cycle of the VESC (it utilizes roughly 5% of the time for measuring current with no power).
So I measured the kV and sure enough, as I suspected, it’s close to what I thought it was. These motors are clearly labelled as 170 kV but the calculation comes out to about 144 kV and when I plug that into the calculations the top speed comes out pretty much to what I’ve been getting. So it all makes sense.
I saw someone said the actual kV could be up to 10% off spec. This is a bit more than that and a little surprising. I guess I’ll see what Flipsky has to say.
It is not necessarily an issue with the motor, it might be the ESC and the way it regulates power
If you run the motor in BLDC (square wave) mode you should get higher top end speeds because the RMS voltage 1 * duty cycle * battery voltage, we usually use FOC (sine wave) which has an RMS voltage of 0.707 * duty cycle * battery voltage
Based on experience running a board in real world conditions with FOC you will get 0.76 * kv * battery voltage as the max angular velocity (accounts for all loss and inefficiency)
Yes that’s what I’ve also been thinking. Fortunately I’m actually very happy with the setup but was just surprised at the numbers and now it all makes sense. Probably better for my safety that I’m not going around doing 75 km/h and I am actually loving the sound of the lower revs. Sounds a lot more relaxed.
170kv being “measured” as 144kv is only a 15% difference and 5% of that could easily be explained by the 95% duty cycle limit of the vesc and the other 10% could be explained if you measured it in FOC mode instead of BLDC.
What matters to me is how far off I am from what I calculated it would be.
For example, according to the 3DS esk8 calc, with my previous setup, calculated max speed was 68 km/h and I would typically get about 56 km/h. So that’s about 82% of the theoretical speed. Of course that was using 190 kV motors and I don’t know how accurate that was.
With the new setup (different gears and motors 170 kV as labeled), calculated max speed should be 88 km/h but I can only get to about 62 km/h which is about 70% of the theoretical speed.
If I use 144 kV in the calculations, this comes to 74 km/h which means the 62 km/h I’m able to get is about 84% of that.
As far as measuring the 144 kV, that already included accounting for the 95% max duty cycle.
KV is measured unloaded. Put the board on a bench, switch to BLDC mode, test the RPMs unloaded, then multiply the measured RPM by 1.05 and divide by the battery voltage, that will give you the true KV.
Another way would be attach a lever to the motor. Make note of the length of the lever from the center of the motor. Suppose it’s 25cm. Place the end of the lever on a scale. Apply 20 motor amps to the motor. Note the force applied to the scale. If its grams write that down. Then multiply the grams by 25 centimeters then divide by 100 centimeters to get the torque in Gram-meters. Convert that to Newton meters. Then divide by the 20 motor amps. That gives you Newton meters per motor amp or KT. Then take 1/KT this will give you the KV in radians per second per volt. Then convert that rad/sec per volt to RPM per volt. That’s another way you can make the KV measurement — by 1st determining the torque per motor amp and converting that to KV.
