Motor comparison spreadsheet

speedmd said:
Just so I am understanding the two losses correctly, the hysteresis losses are from the irons magnetic pole direction switching in both the stator and rotor core steel and the eddy current losses are those currents induced from magnets passing by the stator's copper coiled steel teeth. Would not these eddy's in turn induce further hysteresis losses on their own and cause somewhat a small curve in the hysteresis loss plot. Thinking this may be small in the normal RPM range, but significant when pushed to high RPM's.

Both hysteresis and eddy currents are caused by the changing magnetic field. Hysteresis is the result of particle movement within the steel as they move to align with the magnetic field. Eddy current losses are the resistance losses from the eddy currents created by the changing magnetic field. To a lesser extent these losses also occur in the permanent magnets. No load current at a given voltage tells us the core losses (and parasitic losses, bearings, windage, etc) in total at that rpm, since the current is low enough that the copper losses are negligible. Give me the the no load current, phase-to-phase resistance, along with Kv, and I have almost everything I need to know. Give me a 2nd no-load current at another rpm, and then the spreadsheet gives us the real good stuff, because it provides enough info the give reasonably accurate slopes of the iron core losses.
 
Thanks John for the added insight.

Eddy currents and how they are handled by the controller is a very interesting subject. It appears to get very complicated when pushing high rpm's at least with my little understanding.

Was thinking the hysteresis losses are more like tire friction losses in a formula race car and the eddy currents more like aerodynamic drag. With proper control of the aero flows (deflectors and spoilers) you could exchange a bit of normal drag with rear spoilers- body work (behind the body- tires- tanks etc ), add some down force (with its associated added tire losses) with frontal facing wings for some added drag. You could lessen the total aero drag, or increase the drag in exchange for down force (traction, cornering speed), or a combination of both. In this example, smaller-narrower tires will be better/ easier to drive to higher speeds with regard to the aero losses. What exactly allows a electric motor to run up to 10,000 RPM without loosing it's cool?
 
Data for current Cromotor

Winding: 4T 30 Strands
16 pole pairs
38.6mOhm
Kv: 11.1 with trapezoid controller (x8m06c 12Fet)
0.35mm lams
233mm spoke flange BCD

No load data:
2.375 A
43.45 V
481 rpm

2.072 A
17.17 V
190.5rpm


Finally a motor with a Km of almost 20. Almost double of a MXUS3000 and Hubmonster was beaten, at least in terms of Km :shock: they managed stuff quite a lot of copper into this thing. looks very similar to the QS205 V3 if you ask me BTW

special thx to merlin who provided the motor. @merlin It would be very nice if you would upload a pic of the internals.
 

Attachments

  • cromotor.jpg
    cromotor.jpg
    87.8 KB · Views: 3,801
You mean KM2?
 
That just proves how little reliance can be put on Km and Km2. 2A of no-load at low rpm...LOL!!! What's the resistance anyway, since it's not in the sheet yet?
 
You still dont get that Km has nothing to do with no load current... do you? there simply is no connection. Km relates to resistive loss. No load amps relate to core loss. A different thing.

and you still dont get that a no load amps figure alone tells you nothing about the actual loss. At 1rpm, Hubmonster has 1.45A while Cro has 1.89 A...this is phase current
Can you compare this? No absolutely not. Their Kv or Kt is completely different. what you should compare instead it hysteresis torque. doesnt look too god for the Cro then. Here a LOL would be much more appropriate

still it is 4.5 kg more lightweight but has better Km. Definitely a motor that i'd say is recommendable, even for large wheels like 19". No load loss is much lower than the old Cro, while they could even raise Km. Seems like the lower 16p config pays off.
 
You my friend are clueless on what I get and don't get. Km and Km2 have meaningless units, which makes them useless for any performance calculation. You gave some core loss info and Kv, so I commented on it. I didn't say anything about Km, which is useless to me exactly because it contains no core loss effects and is a dimensionless number, yet you're so fixated on it you go off on a wildly wrong tangent. I asked for resistance, which is a useful number and still don't have it.

No, no-load current is not phase current. No, I don't need hysteresis torque (why leave out eddy current torque, do you not understand there are two components of torque to overcome to spin a motor?), especially when I have no-load current, rpm, and Kv (or Kt). To reasonably compare motors all I need is Kv or Kt, resistance, and no-load current at a useful rpm. The spreadsheet is great in that gives us predicted efficiency (the only truly useful number capable of standing alone for motor comparisons, because it includes everything), and we can manipulate rpm and torque for detailed performance info as well as apple and apples comparisons between motors.

Do yourself a favor and lose this fixation with Km. I know you're excited that you came up with Km2 as probably a better number than Km, but all the number manipulation is leaving you seeing the trees instead of looking at the forest.
 
Of course, no load current always is phase current per definition. Maybe we should talk about no load loss instead of current as many people don't understand the difference.

Eddy current torque looks very good for the Cro, so I didnt mention
 
crossbreak said:
Of course, no load current always is phase current per definition. Maybe we should talk about no load loss instead of current as many people don't understand the difference.

Eddy current torque looks very good for the Cro, so I didnt mention
your confusing us here.

Current flowing thought the motor is Phase current.

Current flowing from the battery to the controller is battery current.

No load current is measured from the battery so its NOT phase current.
 
no it's not battery current. If you post battery current figures here, these will be completely off if not at full rpm without flux weakening. Only if you do a measurement at full rpm, then battery current equals phase current. Only then your figure is valid. Or measure no load loss and calc no load current from that using Kv constant. You get the same results both ways
 
Punx0r said:
Curve ball: It doesn't really matter since it's only no-load POWER that is important?
yep exactly. This way we avoid the - more or less meaningless - phase current/bat current discussion.

Time to add these changes to the spreadsheet, hope i find the time tomorrow
 
crossbreak said:
no it's not battery current. If you post battery current figures here, these will be completely off if not at full rpm without flux weakening. Only if you do a measurement at full rpm, then battery current equals phase current. Only then your figure is valid. Or measure no load loss and calc no load current from that using Kv constant. You get the same results both ways


That's what I am saying, When you ask for no load current its measured at the battery. That's how we have all been doing it. And that's how I got the figure for no load current for the leaf motor
 
crossbreak said:
no it's not battery current. If you post battery current figures here, these will be completely off if not at full rpm without flux weakening. Only if you do a measurement at full rpm, then battery current equals phase current. Only then your figure is valid. Or measure no load loss and calc no load current from that using Kv constant. You get the same results both ways
Battery current is not equal to phase current at full rpm...
 
Punx0r said:
Curve ball: It doesn't really matter since it's only no-load POWER that is important?

+1
 
Lebowski said:
crossbreak said:
no it's not battery current. If you post battery current figures here, these will be completely off if not at full rpm without flux weakening. Only if you do a measurement at full rpm, then battery current equals phase current. Only then your figure is valid. Or measure no load loss and calc no load current from that using Kv constant. You get the same results both ways
Battery current is not equal to phase current at full rpm...

I use battery current less the no-load controller current. How big can the controller losses to convert be at no-load spinup? Do I need to run a controller in an insulated box to prove it's insignificant? Oh wait, it doesn't matter, because our results are conservative since we're tossing some controller losses in as iron core losses.

Now if the spreadsheet is computing something wrong, then that needs fixing. The info available for input is gonna be battery current. Kv is rpm at 100% throttle with no load or wheel divided by battery current. Since the motor isn't doing any work other than the tiniest bit of wind, no-load current X voltage (both from the battery) = core losses since current is so low that copper losses are insignificant.

Hopefully that puts me in the camp with Punx0r and Liveforphysics. :mrgreen:
 
Astro motors are 80% motor and 20% toaster oven.

I think the reason for that 20% toaster oven is a valid point because yes its a good thing the motor and internal can handle 200c , but the question is why would you run it that hot as it significantly reduce the efficiency of the motor.

All motors have copper winding so every degree of heat causes the temps to increase and hence resistance to go up which reduce efficiency. Compared to a normal bigger motor that runs average winding temps of 80 deg to one that has 160 deg avg temps i.ce astro that's a few 2-3% points less efficiency.

Instead you could use a normal style motor that is a bit bigger and yes heavier or vent a normal motor with air cooling.

The other main negative of Atro motors for ebikes is that their RPM is a lot higher then needed and in most cases for a mid drive you would need 1 extra stage of reduction which add 3-5% efficiency losses of the drievtrain.

So combine these two things and its not such a good motor for an ebike after all, you can see it discussed in this thread that the 1680w cyclone motor probably has the same or more efficient on an ebike with similar weight... I guess this makes sense as the astro was optimised for ultra light weight in a plane use which is not what an abike is....

is it true that the astro motor has been optimised to suit an aeroplane and hence minimise the weight, which is a compromise when used on an ebike as a mid drive ? i.e the weight is so low their is no thermal mass and the RPM is so high that you need another stage of reduction which adds efficiency losses and the overall efficiency is no better then the1680w cyclone motor ?
 
crossbreak,

I'm still waiting for the resistance on the motor you think is better than HubMonster by some metric, and the full details into the spreadsheet. 30 strands wound on fewer teeth than the 8yr+ old design so commonly copied sounds interesting, but since it could be smaller gauge strands phase-to-phase is the best info.
 
crossbreak said:
Data for current Cromotor

Winding: 4T 30 Strands
16 pole pairs
38.6mOhm
Kv: 11.1 with trapezoid controller (x8m06c 12Fet)
0.35mm lams
233mm spoke flange BCD

No load data:
2.375 A
43.45 V
481 rpm

2.072 A
17.17 V
190.5rpm

thanks for doing the measurements. thumbs up!
have you also done measurements on previous cromotor and if yes, could you post them?

ridethelighting took some pics of the internals of his QSV3 motor. in this thread: https://endless-sphere.com/forums/viewtopic.php?f=31&t=65972&start=125#p1047031

this newest Cromotor (it's called V4 i think?) IS QS 205 V3 - nothing else, just like previous Cromotor was a QSV2 ;)
as a hint: cross section of one single strand should be 0,2mm² (same on QSV2 and MXUS 3k and i believe on many other motors as well)
 
Nice improvement over previous versions. A big improvement in resistance, which boosted crossbreak's favorite number. Thanks for editing the resistance in there. It's still a bit rpm limited for my taste, and 3 phase is a detriment, but it's certainly getting closer.
 
from the pics aus the QS Topic....yes....looks like the V4 is a QS with "zelena" axle....

Cromotor_V4%20%281%29.jpg


Cromotor_V4%20%282%29.jpg
 
We're drifting off topic, but look at that unused space that should be filled with copper, and instead its got fiberglass or bamboo sticks. If I order some samples, I'll be pushing for another strand or two per turn.
 
Back
Top