efficiency and pwm

[Alan B]

The lams need to be electrically isolated, or they have little value and might just as well be solid metal.

If you can take them apart, deburr them and re-coat them with a tough lacquer or other thin nonconductive paint that would restore the isolation.

 

[justin_le]

Any way to better isolate the smeared stator? I smeared one last night as the motor structure became loose and had a bit of rubbing with the magnets. I've had this before and the magnets are super durable but the stator will polish in a spot. I wonder if there's anything that can be done to reisolate. I'm guessing no.

What I did on stator that I had machined was slowly roll the stator in a tray with a thin layer of acid (HCl) to chemically etch away the corners and smeared over stator edges, then rinsed thoroughly in water and treated with phosphoric acid so that the freshly exposed steel was better protected against rusting. And believe it or not, the process worked flawlessly, and the motor after reassembly was back to minimal cogging.



Not an everyday repair process to encounter but it did the trick.

It seems the lams aren't fully isolated from each other anyway even at the start

They are never fully isolated from one another, they make contact on the core and at various points between the layers. A few spot points of contact between the lams doesn't make for a very effective current path for eddie currents between the layters, but the goal is to reduce the number of contact points so that you don't have that many large area loops that can conduct. Usually they don't bother putting a varnish or insulating coat on the steel because the loss of steel fill factor from doing that more than outweighs the benefit of having 100% isolation.

 

[liveforphysics]

Nice repair job! That's clever to acid etch the shorts away.

 

[Hummina Shadeeba]

I'd have thought the acid would eat the insulation on the lams and also the windings but it ate the steel. That's a surprise.


I'm wondering about what Justin said here:
"There is some marginal truth to this, that running a motor at say 24V and 100% duty cycle could be a tiny bit more efficient than 48V at 50% PWM duty cycle, but when I say tiny I really mean tiny. "

this situation reminds me of having a motor with a kv that's much too high...and then you'd alway be in a lower duty cycle and increased motor amps compared to what the pack was putting out...and therefore able to get the necessary amps to create torque with such a high kv. No?

 

[Hummina Shadeeba]

still don't know why you can't just use a much higher kv motor than you'd typically want if the duty cycle can be lowered and the motor can receive whatever current it needs.

with pwm on startup at lowest duty cycle there can be huge amperage, does this equate to tiny moments of huge torque?

 

[ferret]

As long as the controller and battery can handle the current required by the high KV motor...

Yes, Torque = Current X Motor KT

Avner.

 

[gogo]

still don't know why you can't just use a much higher kv motor than you'd typically want if the duty cycle can be lowered and the motor can receive whatever current it needs.

with pwm on startup at lowest duty cycle there can be huge amperage, does this equate to tiny moments of huge torque?

2nd question:
No, the motor windings smooth out the flow of electrons. If you're asking about the difference between high and low Kv ebike motors operated at the same power and speed, my guess would be that you wouldn't notice any difference in smoothness.

1st question:
A high Kv motor operated at a lower duty cycle generally requires a more expensive and physically larger controller to handle higher currents.

 

[Hummina Shadeeba]

so running a high kv motor with a high voltage can be just as efficient in the motor as long as the esc duty cycle is set to change appropriately?


and I think you're saying that at the lowest duty cycle pulses maybe be huge current but the coils being inductors take enough time to build up the magnetic field so there aren't extremely small moments in time with extreme torque, just amps.

 

[Alan B]

still don't know why you can't just use a much higher kv motor than you'd typically want if the duty cycle can be lowered and the motor can receive whatever current it needs.

with pwm on startup at lowest duty cycle there can be huge amperage, does this equate to tiny moments of huge torque?

As was pointed out, the motor winding inductance filters the current into a fairly smooth value. The huge current pulses are from the capacitor bank through the FETs.

In theory this works. In practice there is additional capacitor current capacity, and increases in the FET silicon and copper conductors required (and more engineering and development), plus there are additional losses and limits to how narrow the PWM can be and how small the interval of PWM adjustments are (making the throttle less smooth).

So the further you go in this direction the more it costs, the larger and heavier the controller gets, and the more losses there are in the controller. Like most engineering problems, there are tradeoffs and it becomes inefficient and expensive if you go too far.

 

[Hummina Shadeeba]

I've a friend who keeps insisting that the torque is not smoothed out and that at 765hz i'm experiencing 14,000 watts peak that is producing fractions of second with extreme torque due to the duty cycle. what can I say beyond the coils in the motor are smoothing the current out and it's not creating torque spikes.

 

[gogo]

I've a friend who keeps insisting that the torque is not smoothed out and that at 765hz i'm experiencing 14,000 watts peak that is producing fractions of second with extreme torque due to the duty cycle. what can I say beyond the coils in the motor are smoothing the current out and it's not creating torque spikes.

Why does he insist? Is he measuring torque pulses?

 

[Alan B]

I've a friend who keeps insisting that the torque is not smoothed out and that at 765hz i'm experiencing 14,000 watts peak that is producing fractions of second with extreme torque due to the duty cycle. what can I say beyond the coils in the motor are smoothing the current out and it's not creating torque spikes.

Do the math.

If this were happening the motor would buzz with a loud sound. A motor is also a sound transducer (speaker), some ESCS make various tones in the motor during arming, for example.

Calculate the L/R time constant of the motor and compare that to the period of the PWM pulses. It takes about 5 time constants for the current to go to zero (it drops to 63% each Tau). If the current is decaying significantly between PWM pulses then the PWM frequency is wrong, it is chosen to avoid significant current decay.

Make a simple spice model and you can see the current ramping up and down at the L/R rate, making a small ripple on top of the average constant current.

 

[Hummina Shadeeba]

yea he said it was creating torque pulses. thanks for your help on this I have the answer i need.

 

[liveforphysics]

I've a friend who keeps insisting that the torque is not smoothed out and that at 765hz i'm experiencing 14,000 watts peak that is producing fractions of second with extreme torque due to the duty cycle. what can I say beyond the coils in the motor are smoothing the current out and it's not creating torque spikes.

Your friend should read up on 'inductance' and 'buck-converter'.

 

[Hummina Shadeeba]

http://www.learnabout-electronics.org/ac_theory/dc_ccts45.php

the answer to if a motor gives torque pulses in time with the pwm is no because of the inductance time constant I think. Ripple being fluctuating inductance maybe.

this reveals a possible inefficiency in a motor, is it common? is it common for a frequency to change?

 

[liveforphysics]

http://www.learnabout-electronics.org/ac_theory/dc_ccts45.php

the answer to if a motor gives torque pulses in time with the pwm is no because of the inductance time constant I think. Ripple being fluctuating inductance maybe.

this reveals a possible inefficiency in a motor, is it common? is it common for a frequency to change?

It's not loss. It's not inefficiency, and it's not PWM that makes the torque ripple, it's commutation freq.

 

[Hummina Shadeeba]

My question is why wouldn't the frequency of the pwm create something similar to torque ripple. torque ripple seems to explain the level of torque based on the amount of poles and the distance between them I'm talking of with much smaller time periods the frequency of the pwm. At half duty cycle you can have a high frequent or low just as long as measured within the time period it in on half the time. According to what I posted the inductance time constant reveals the time it will take the coil to build its magnetic field. If it were to take less time to build the field, especially during low duty cycle when there are possibly many amps fed to the motor in pulses, the magnetic field would build and collapse in fractions of a second and the motor would therefore produce pulses of torque. But the magnetic field is produced slower than the frequency and the rapid pulses are smoothed out. No torque pulses and also according to the link the higher the frequency the better in this regard. That's what I got from this, this is how I understand it, what's missing?

I did the math to figure how long it takes with my motor for the field to build and collapse, don't have the numbers now but it's like .00006 seconds or something per period, and the frequency I believe is .00002 or something , so no torque pulses as the field takes 5 periods to get 99.5% complete

 

[liveforphysics]

My question is why wouldn't the frequency of the pwm create something similar to torque ripple.

The motor only makes torque in response to phase current, not voltage. The PWM makes a voltage waveform, that when filtered by the motor's inductance makes a current which is super smooth, and only that current makes a magnetic field that imparts forces on the motors rotor.

 

[Hummina Shadeeba]

Yes. Was wondering if the current has time to build and collapse a magnetic field during its on or off period based on frequency : just how smooth does the current end up.
After doing the math the frequent of the pwm is 5x higher than the LR time constant allowing no torque pulses. Smoothed as is being said.




LR Time Constant = H / Ohm
I'm my motor the inductance time constant is:
0.0006573 S = (0.00002713 H) / (0.0415 ohm)

With the motor at 24% duty cycle the frequency ON is .00003 seconds (assuming a bunch of stuff)