Motor comparison spreadsheet

[crossbreak]

thanks miles, i will redo the math in a separate column and see if we get the same results. Once it matches, i'll put your chosen calc method in an image/pdf like above. Just to have it for reference. Next step is to include inductance as it shifts the line of best efficiency to lower rpm, which may be significant for some types of motors

John, thanks for contributing so many measured data. I am waiting keen for no load numbers of the RV120 I ordered a micro ohm meter on ebay for less then 50 bucks. hope it arrives eventually. A single tool for measuring all values (no load current, rpm, inductance) would be nice though. I was successful in using a simple multimeter (set for frequency) to measure commutation frequency at no load btw, this makes life easy

 

[Miles]

thanks miles, i will redo the math in a separate column and see if we get the same results. Once it matches, i'll put your chosen calc method in an image/pdf like above. Just to have it for reference

Thanks CB.

It wouldn't be difficult to add a torque input column to the spreadsheet, as John suggested. I never really intended this to be a motor simulator, though...

 

[John in CR]

thanks miles, i will redo the math in a separate column and see if we get the same results. Once it matches, i'll put your chosen calc method in an image/pdf like above. Just to have it for reference

Thanks CB.

It wouldn't be difficult to add a torque input column to the spreadsheet, as John suggested. I never really intended this to be a motor simulator, though...

Maybe it will lead to the spreadsheet calculating a realistic power rating at a given voltage. Even if variable torque doesn't stay I'll have fun gaining a better understanding of the effects of load on motor operation. Peak efficiency is interesting to see playing around with rpm, but peak eta is a point we may only rarely cross, so while it's a great number for comparing motors, maybe we can do better.

 

[speedmd]

Great work guys. I am stumped by what may effect the ways a particular motor design may run into efficiency losses at either end of the rpm range. Can one design be much better at having a broad efficient RPM band while another burn up quickly if ventured out of its sweet spot? What exactly drives a robust / broad / efficient power band? In the efficiency contour diagram below what factors move the vertical boundaries steeper and to the left and horizontal boundaries flatter and lower. Certain motor designs may be equally peek efficient, but hate to be stalled or run hard from a stop. Some designs may be better at extending the rpm range without major losses creeping in. Do some designs have more of a step off of this sweet spot of ideal operation. Would be critical in knowing in a direct drive vs mid drive application or depending on your operating goals.

 

[Miles]

Great work guys. I am stumped by what may effect the ways a particular motor design may run into efficiency losses at either end of the rpm range.

Peak efficiency will keep increasing with speed until the eddy current losses are significantly greater than the hysteresis losses. The only thing that would make it come down again, AFAIK, are windage losses, which increase as the cube of speed. Obviously, there's no point in having the peak efficiency power out higher than the continuous power capability, at that speed....

 

[speedmd]

there's no point in having the peak efficiency power out higher than the continuous power capability

Understood, but in the situations of tail winds / downhills and coasting along slightly off throttle but wanting to travel quickly as many would be doing in a high power build, what exactly is most important in allowing a motor to buzz along as freely as possible. I can understand that the more copper / lower resistance/ (inductance also?) helps at the bottom of the rpm range and puts the controller more at risk if not limited in some fashion.

 

[crossbreak]

you wont get a more definite answer to that than miles just did. windage loss is too low to really become an important factor, what limits power is heat dissipation in the first place.

now you could say that why dont people use water cooling+air cooling? Answer is they do, but this again increases weight / lowers sKm². And it reduces efficiency at some point as a big fan/pump may introduce windage loss that may no more be unimportant

As soon as you lower some sort of loss you will increase another (or raise cost). A good motor simply is a good compromise. The factors that shape the contour plot are those recently discussed and expressed in that formula. Hopefully i can come up with some little app that allows us to play around a bit with these plots without the need of installing Matlab

 

[John in CR]

What exactly drives a robust / broad / efficient power band?

Running a motor at low stress levels is the key. Take Miles motor for example. I have no doubt running it in its designed use that it will possess the qualities you mentioned. High on his list of qualities for the design is low weight and high efficiency, so it spins up moderately fast, but still able to handle the single reduction to the wheel. It has room in the spectrum for somewhat higher power than he's after by increasing voltage, but that's likely to require a 2nd stage of reduction. Its relatively high resistance means don't dare try to pump big current through the motor. If I wanted to build an ebike that was as elegant to pedal as it was as an electric machine, then his is the motor I'd want.

I'm at the other end of the spectrum and want motorcycle performance and range. Big power pushing my big self and a big battery means a 15kg motor is acceptable, because it's relatively immaterial to the total load. What doesn't show up yet in the spreadsheet except in the Km fields is how HubMonster is significantly more motor than MidMonster (Yes I'm finally seeing some value in Km, but it's not the "great equalizer" except that it makes low efficiency designs look better than they are.) The difference between HubMonster and MidMonster should be readily apparent as we load them down requiring higher torque/current.

We still haven't addressed heat rejection yet, and that's our overriding limitation. The less heat we make though, the less we have to reject.

 

[John in CR]

there's no point in having the peak efficiency power out higher than the continuous power capability

Max continuous power is not a fixed number. It varies with ambient temperature, how the motor is installed, and can even vary with rpm. Once I start running with my Analogger installed I'll be able to demonstrate this with real data.

 

[Miles]

Obviously, there's no point in having the peak efficiency power out higher than the continuous power capability, at that speed....

Max continuous power is not a fixed number. It varies with ambient temperature, how the motor is installed, and can even vary with rpm. Once I start running with my Analogger installed I'll be able to demonstrate this with real data.

Sure. The principle stands, though....

 

[Arlo1]

there's no point in having the peak efficiency power out higher than the continuous power capability

Max continuous power is not a fixed number. It varies with ambient temperature, how the motor is installed, and can even vary with rpm. Once I start running with my Analogger installed I'll be able to demonstrate this with real data.

Your going to either need a weak motor or a big battery to prove that john.

When Nissan tested the leaf motor they run it for an hour at its rated continuous power to prove it doesn't increase in temperature. They run it at lower levels for 1 hour increments to see where the temperature equalized and kept upping the power until they got to 80kw and the temp stabilized at a number that's warm but still safe for the motor.

http://energy.gov/sites/prod/files/2014/03/f13/ape006_burress_2013_o.pdf

 

[speedmd]

A good motor simply is a good compromise.

Perfect! Understood. You optimize best you can and simple is best. FOC can have some lowering of losses at the top end. Varying the amounts of back iron on the permanent magnets side would effect iron losses also. Comparing, permanent magnet motors to induction motors, the induction motors give up power and torque density in exchange for more control over both sides of the motor? Guessing this is why some have gone that route on high rpm apps.

 

[Miles]

How about something like this, as an additional section?

 

[crossbreak]

i had that in mine already


also interesting: Take a motor, put current in it at stall (DC). Look how warm it gets, try to find an equilibrium. This way you can estimate how much loss is acceptable. For example i dont like my MAC motor to get warmer than 110°C. But this is just a number i grabbed out of thin air. I just think this is fine for me. Someone who wants to last it longer wont want the 110°C that i do. The MAC can dissipate around 200W at that temperature without turning at all. Guess that dissipation increases with rpm

 

[Miles]

i had that in mine already

In that case.. Can I cut and paste from yours?

 

[crossbreak]

sry i'm at work

 

[Miles]

Only joking.. It's not much work....

 

[Miles]

Separate section added to allow for user input of torque output and loaded speed values. The new speed column is independent of the existing speed column.

Ok, John?

 

[John in CR]

Thanks Miles. Christmas came early with a new toy to play with. I'm already seeing interesting relationships, some expected and some not.

Time to really populate the sheet.

John

 

[Miles]

I ordered a micro ohm meter on ebay for less then 50 bucks. hope it arrives eventually.

Let us know what it's like.... One of these? http://www.ebay.co.uk/itm/OLED-Digital-LED-Micro-Ohm-Resistance-Meter-Micro-Ohm-Meter-Milliohm-W-test-Clip-/381126896698?pt=LH_DefaultDomain_3&hash=item58bceeec3a

 

[crossbreak]

yep. i have some 5mOhm/ 1% shunts that i can use to validate precision. maybe i find some even better ones in the university lab

 

[Miles]

Great. I'll order one for myself, if they're any good....

 

[John in CR]

How would you guys use the micro ohm meters? Are we talking about for measuring phase-to-phase resistance or the IR of batteries? If it's a tool I need I'm more than happy to get one, but is it better than the voltage drop at a know current method? If I'm going to provide measurements of 15-20 different motors for the spreadsheet, I want to do things properly.

My biggest concern with the meter would be getting good connections to avoid including resistance of the connector or connection. Is that the primary use, being able to go through all my wiring and connectors to look for higher than necessary resistance?

 

[crossbreak]

exactly.

to avoid the influence of the test probe contact resistance, you can use the 4-wire method with such a micro ohm meter. The meter supplies it's own known test current of around 100 mA

 

[crossbreak]

looks promising. shunt is a 1% type. the meter is this one: http://www.ebay.de/itm/121505331829?_trksid=p2060353.m2749.l2649&ssPageName=STRK%3AMEBIDX%3AIT