Cyclone ebike motor efficiency 97%

[avandalen]

According Cyclone, the maximum efficiency of the mid-drive motors is 97%, but I don't believe that, this is impossible. Maybe it's a typing error: 79%. To create the efficiency graph, we just have to know the motor resistance, motor constant kv and the no load current. Who knows these values or can measure these?
See more here:
http://www.cyclone-tw.com/dc24.htm
http://www.avdweb.nl/solar-bike/hub-motor/measuring-motor-parameters.html

 

[amberwolf]

See the science, math and myth thread for discussion on how to measure some or all of those, and Miles' follow-on thread (linked in a post there) for data on various motors.

 

[Miles]

I don't have any data on the Headline motor. It would be good if someone could provide the data on these as they're so widely used. I think you'd be pretty safe to assume the test results on the Cyclone site were provided by "Any Spec U Like Testing Inc.". It seems to be a motor designed more for ease of mass production than performance.....

 

[spinningmagnets]

Here's a thread from the "technical reference" section. I will add any info you find out.

http://endless-sphere.com/forums/viewtopic.php?f=16&t=62589

 

[Miles]

This is the relevant page at Cyclone: http://www.cyclone-tw.com/dc24.htm

Peak eta is stated as 97.6% (-15degC. to 40degC.) for the 360Watt motor.

 

[liveforphysics]

That cyclone reaching 97.6% seems about as believable as pedaling your bike at 97.6mph unassisted and without streamlining.

 

[Animalector]

I often wonder why people make these claims on their website.. Are they deliberately being misleading, or did they do some test (incorrectly) and ended up with a value of 97.6%??? Surely they can't just publish the values knowing it's a lie...

Andy

 

[amberwolf]

Why not? Any buyer that would know to question it probably wouldn't want that type of motor anyway; would probably know what they are looking for in a motor design-wise and parts-wise, data.

Companies lie all the time; it's called "marketing". Sometimes the lies are outrageous, sometimes they're almost innocent.

We don't have to like it, but it's common practice in many (probably all) industries, and most of the time it's hard to call them on it and get anything actually done about it. :/

 

[rscamp]

Wild guess. Maybe someone in the marketing department thought it would be okay to put the efficiency of the gear reduction there.

 

[Miles]

Wild guess. Maybe someone in the marketing department thought it would be okay to put the efficiency of the gear reduction there.

As far as one can tell, the graph appears to show efficiency peaking at around 96%....

 

[Animalector]

Wild guess. Maybe someone in the marketing department thought it would be okay to put the efficiency of the gear reduction there.

At least that has some truth to it.. and therefore not a flat out lie.. I don't believe that marketing is about blatantly telling lies. it's about selecting the information that is most appropriate to the target market. In Australia we have laws against false / misleading marketing.

But anyway...

 

[avandalen]

Bill Bushnell has done measurements, the efficiency is a poor 74% instead of 97%:
http://mrbill.homeip.net/albums/cyclone_motor/efficiency_curves/cyclone_360_eff.pdf

This low efficiency does not surprise me because the fill facor is very low:

 

[avandalen]

Maybe I use the Cyclone gearbox with a Astro 3205 motor:
http://www.astroflight.com/3205
I've done simulations with the Astro 3205 in Excel, the efficiency including the gearbox would be greater than 90% in a large working range.

 

[Gab]

Bill Bushnell has done measurements, the efficiency is a poor 74% instead of 97%:
http://mrbill.homeip.net/albums/cyclone ... 60_eff.pdf

if you read on the bottom of bills page he estimates that the motor to rear wheel losses where he measures the power is 13% i think this is quite optimistic for a 3 stage reduction ! i.e for 9.33 gearbox, 14 to 48t chainwheel then 48t to rear sprocket. i would thinks its more then 13% drivetrain losses?

but anyway so the motor efficiency is at least 13% better hence 74+13 = 87% and more. On the cyclone graphs the 97% is for the motor only and due the extra drivetrain losses i would think it could come close to the 90 to 95% at the motor.

 

[avandalen]

Bill Bushnell has done measurements, the efficiency is a poor 74% instead of 97%:
http://mrbill.homeip.net/albums/cyclone ... 60_eff.pdf

if you read on the bottom of bills page he estimates that the motor to rear wheel losses where he measures the power is 13% i think this is quite optimistic for a 3 stage reduction ! i.e for 9.33 gearbox, 14 to 48t chainwheel then 48t to rear sprocket. i would thinks its more then 13% drivetrain losses?

but anyway so the motor efficiency is at least 13% better hence 74+13 = 87% and more. On the cyclone graphs the 97% is for the motor only and due the extra drivetrain losses i would think it could come close to the 90 to 95% at the motor.

We can easily calculate the motor efficiency if we know the winding resistance and the motor constant, who can measure that?
See here:
http://www.avdweb.nl/solar-bike/hub-motor/measuring-motor-parameters.html

 

[sigimem]

don't the cyclone motors have a planetary gearbox 1 stage reduction?
edit: just read total reduction. A bicycle chain is supposed to have 98% efficiency. So two means 0.98*0.98=0.96. 4% loss 0.87/0.96= 90.6% efficiency for the planetary gearbox seems ok for me considering it has straight gears.

edit:
if the reduction system has 13% loss the motor efficiency is 74%/87% so 0.74/0.85 =0.85 85% just saying.

 

[mrbill]

I found my old notes where I had measured the phase resistances of two different Cyclone motors, both of them the standalone (w/o internal controller) motors that have the same characteristics as the "500 watt" motor, but without the internal controller. The "500 watt" motor looks the same as the "350 watt" motor except that the "500 watt" motor is longer axially by about 10mm.

The "500 watt" motor has a kv of approximately 125 rpm/v. As I recall the "350 watt" motor has a slightly higher kv, maybe around 135 or 140.

Motor 1:
phase 1: 68 mOhm
phase 2: 67 mOhm
phase 3: 67 mOhm

Motor 2:
phase 1: 73 mOhm
phase 2: 71 mOhm
phase 3: 76 mOhm

Paco of Cyclone Taiwan told me once that the Cyclone motors are Wye wound, although I did not verify that.

don't the cyclone motors have a planetary gearbox 1 stage reduction?
edit: just read total reduction. A bicycle chain is supposed to have 98% efficiency. So two means 0.98*0.98=0.96. 4% loss 0.87*1.04= 90.5% efficiency for the planetary gearbox seems ok for me considering it has straight gears.

if the reduction system has 13% loss the motor efficiency is 74%*113% so 0.74*1.14=0.836 83.6% just saying.

If we're looking at a system efficiency (0.74) and trying to back out the effect of the drivetrain (eff=0.87) so that we isolate the motor's efficiency, we'd calculate: motor eff = 0.74 / 0.87 = 0.85. This is a reasonable figure.

Somewhere in some thread on E-S Jeremy Harris remarked that he had tested a bunch of different mass-produced BLDC motors in the 1/2 - 2 hp range appropriate for e-bike use. He discovered that the peak efficiencies of his motley collection were all within about 10% of each other. Based on my testing over the years I'd have to agree.

 

[jk1]

85% is a pretty good motor efficiency ??

Is the 500w motor with the internal controller the same as the high power one with the external controller ? i think they are both similar size 10mm larger then the 350w ?

On this thread here their some internal photos of the external controller version.

http://endless-sphere.com/forums/viewtopic.php?f=30&t=66144

interestingly they added 4 turns to the motor and reduce the KV by 10%, could you work our from this how many turns the motor is standard? the motor phase resistance seems quite low compared to your figures only 44mohm ?

Bill would having these 4 extra turns increase the efficiency much from 85% ? or would changeing from internal IRF3205 8mohm mosfets to some new 2.5mohm mosfets make much difference to this 85% motor efficiency?

 

[jk1]

85% is a pretty good motor efficiency ??

Is the 500w motor with the internal controller the same as the high power one with the external controller ? i think they are both similar size 10mm larger then the 350w ?

On this thread here their some internal photos of the external controller version.

http://endless-sphere.com/forums/viewtopic.php?f=30&t=66144

interestingly they added 4 turns to the motor and reduce the KV by 10%, could you work our from this how many turns the motor is standard? the motor phase resistance seems quite low compared to your figures only 44mohm ?

Bill would having these 4 extra turns increase the efficiency much from 85% ? or would changeing from internal IRF3205 8mohm mosfets to some new 2.5mohm mosfets make much difference to this 85% motor efficiency?

 

[avandalen]

I found my old notes where I had measured the phase resistances of two different Cyclone motors, both of them the standalone (w/o internal controller) motors that have the same characteristics as the "500 watt" motor, but without the internal controller. The "500 watt" motor looks the same as the "350 watt" motor except that the "500 watt" motor is longer axially by about 10mm.

The "500 watt" motor has a kv of approximately 125 rpm/v. As I recall the "350 watt" motor has a slightly higher kv, maybe around 135 or 140.

Motor 1:
phase 1: 68 mOhm
phase 2: 67 mOhm
phase 3: 67 mOhm

Motor 2:
phase 1: 73 mOhm
phase 2: 71 mOhm
phase 3: 76 mOhm

Paco of Cyclone Taiwan told me once that the Cyclone motors are Wye wound, although I did not verify that.

don't the cyclone motors have a planetary gearbox 1 stage reduction?
edit: just read total reduction. A bicycle chain is supposed to have 98% efficiency. So two means 0.98*0.98=0.96. 4% loss 0.87*1.04= 90.5% efficiency for the planetary gearbox seems ok for me considering it has straight gears.

if the reduction system has 13% loss the motor efficiency is 74%*113% so 0.74*1.14=0.836 83.6% just saying.

If we're looking at a system efficiency (0.74) and trying to back out the effect of the drivetrain (eff=0.87) so that we isolate the motor's efficiency, we'd calculate: motor eff = 0.74 / 0.87 = 0.85. This is a reasonable figure.

Somewhere in some thread on E-S Jeremy Harris remarked that he had tested a bunch of different mass-produced BLDC motors in the 1/2 - 2 hp range appropriate for e-bike use. He discovered that the peak efficiencies of his motley collection were all within about 10% of each other. Based on my testing over the years I'd have to agree.

Hi Bill,

Thank you very much for the figures.

I have done simulations with your values. Here are the 2 motor graphs, in the same power range 0 to 1000W, and without gearbox. I don’t know the friction of the Cyclone motor, so I take the same value as the Astro 3205 (0,02Nm).

Surprisingly, the Cylone motor is comparable to the Astro 3205 in the range 0 to 500W (with the assumed friction of 0,02Nm). But note that the Astro needs a larger gear ratio which has a larger loss.

However with a Cyclone gearbox, the efficiency is strongly dependent on the gearbox losses which I don’t know yet. But this is a problem for both motors of course.

I measured the Cyclone gearbox friction without load; it was 0,4Nm on the output shaft without the motor mounted (without sun gear).

Do you know anything about the efficiency with and without the planetary gear?

Albert

 

[mrbill]

85% is a pretty good motor efficiency ??

85% is a reasonable figure. Whether that is "good" in a qualitative sense is up to you.

Is the 500w motor with the internal controller the same as the high power one with the external controller ? i think they are both similar size 10mm larger then the 350w ?

Yes.

On this thread here their some internal photos of the external controller version.

http://endless-sphere.com/forums/viewtopic.php?f=30&t=66144

interestingly they added 4 turns to the motor and reduce the KV by 10%, could you work our from this how many turns the motor is standard? the motor phase resistance seems quite low compared to your figures only 44mohm ?

My measurement of around 70 mOhm includes the leads exiting the motor, about 1/2 m of extra 12 ga. cable, although that would not explain all of the 26 mOhm of difference. I measured phase resistance between each pair of motor leads by running a steady current (approx. 8 Amps) through the circuit, then measuring voltage drop across the motor leads and calculating the resistance. I did not measure the inductance.

Bill would having these 4 extra turns increase the efficiency much from 85% ? or would changeing from internal IRF3205 8mohm mosfets to some new 2.5mohm mosfets make much difference to this 85% motor efficiency?

Winding an extra 4 turns would be a big hassle. You'd have to splice in the wire such that it doesn't short out on an adjacent wire when wound tightly on a tooth, then wind it by hand, taking care not to make an error. But, the Kv of the motor would be reduced, which might throw off pedaling the ergonomics, unless you also increase the voltage accordingly.

Swapping out the controller's MOSFETs for some 3077s or similar is also a hassle, but one with more predictable results (assuming, of course, that your workmanship is clean). Efficiency gain would probably be greater at lower throttle settings. Expect up to 3% improvement in overall efficiency at half-throttle.

At some point all the extra tweaks and mods just aren't worth the hassle, and it's easier to use a different motor or to buy a new battery that gives you the extra range without weighing more than your old battery.

 

[mrbill]

Do you know anything about the efficiency with and without the planetary gear?

Hi Albert:

I wasn't able to measure the efficiency of the Cyclone/Headline gearbox. I was only able to measure the Matex planetary gears used in the Currie and Lashout gearbox

http://mrbill.homeip.net/albums/lashout_planetary_gearbox/index.html

because that motor uses a standard 8mm output shaft to which I could attach a sprocket to measure the motor without the gearbox. That gearbox's efficiency varied from 93-97%, highest when loaded.

http://mrbill.homeip.net/albums/mac_bmc_600w_motor/efficiency_curves/M1_gearbox_study.pdf

My gut feeling is that the Cyclone/Headline gearbox has lower efficiency, especially at low loads. The Cyclone/Headline gearbox teeth are larger, the reduction is twice as great, and the unloaded friction (as you measured) is greater. On the other hand the Cyclone/Headline gearbox appears to be machined to a tighter spec with planet gears held firmly in their carrier so that they run true (which may explain why unloaded friction is higher). I don't want to hazard a figure without being informed by data.

http://mrbill.homeip.net/albums/cyclone_motor/index.html

 

[Gab]

Surprisingly, the Cylone motor is comparable to the Astro 3205 in the range 0 to 500W (with the assumed friction of 0,02Nm). But note that the Astro needs a larger gear ratio which has a larger loss.

The reason the cyclone is as efficient as an ASTRO motor at these low power levels and hence low RPM is due to the motor design, it has an IRON core stator that hlep increase the torque at low RPM, the ASTRO is ironless which is only good at high RPM and high power. Also the magnets are IPM designs versus surface mount magnet motors and i have been reading they are higher efficiency and higher torque per mass then when the magnets are mounted on the rotor surface, and they also allow higher RPM as this also stops the magnets flying out.

Hi Bill so the motor has 85% efficient and with gearbox overall 74% in stock cyclone form with grease in the gearbox ? i wounder how much more efficient it would be with some of the improvements mentioned above or with AFT EMAX version of the cyclone , seems they got some big efficiency gains with these changes:

-Gearbox Grease to oil conversion
AFT custom low friction oil with high concentration Mos2 NANO particles. “Moly Gear Oil Additive considerably reduces wear and lowers peak temperatures. Moly Gear Oil additive smoothes out tooth faces and ensures almost wear-free, smooth running operation as well as less harsh shifting”. Can extend gear life by up to 3 times compares to grease (customers have achieved more than 4000miles on one set of gears using the EMAX oil conversion).
-AFT chain guide upgrade ( no chain tensioner)
-3 x AFT hybrid ceramic gearbox bearings
- 2 X AFT Motor hybrid ceramic bearings

http://www.aft-ebike.com/why-choose-aft.html

 

[jk1]

Swapping out the controller's MOSFETs for some 3077s or similar is also a hassle, but one with more predictable results (assuming, of course, that your workmanship is clean). Efficiency gain would probably be greater at lower throttle settings. Expect up to 3% improvement in overall efficiency at half-throttle.

Going from the IRF3205 RDS(on) = 8.0mΩ to some IRF3077= 2.8mohm is an improvment but they are still very old design ? What about something newer like these FDP020N06B=1.6mohm , they are 60v and 100A so ideal for anything up to 48v. Would they be a strait swap to the IRF3205 in the cyclone motors ? or is some of the values to different to work ?

https://www.fairchildsemi.com/datasheets/FD/FDP020N06B.pdf

How come the efficiency gain is only at partial throttle ? even at full throttle all of the motor power is trying to get through those mosfets at 8.0mohm so surely going to a 1.6mohm should have a large efficiency gain at both full throttle and partial throttle ? as its 80% less RDson resistance ?

 

[mrbill]

Hi Bill so the motor has 85% efficient and with gearbox overall 74% in stock cyclone form with grease in the gearbox ? i wounder how much more efficient it would be with some of the improvements mentioned above or with AFT EMAX version of the cyclone , seems they got some big efficiency gains with these changes:

-Gearbox Grease to oil conversion
AFT custom low friction oil with high concentration Mos2 NANO particles. “Moly Gear Oil Additive considerably reduces wear and lowers peak temperatures. Moly Gear Oil additive smoothes out tooth faces and ensures almost wear-free, smooth running operation as well as less harsh shifting”. Can extend gear life by up to 3 times compares to grease (customers have achieved more than 4000miles on one set of gears using the EMAX oil conversion).
-AFT chain guide upgrade ( no chain tensioner)
-3 x AFT hybrid ceramic gearbox bearings
- 2 X AFT Motor hybrid ceramic bearings

http://www.aft-ebike.com/why-choose-aft.html

Some of those modifications look interesting, such as the gearbox shroud to cut down on the noise, and the magnets to attract gearbox wear debris. It's possible that they wring as much performance from the Headline motor and gearbox as is possible. But, I doubt overall efficiency is increased by more than a few percent. Some of the mods look like a hassle what with an oil tube and reservoir to keep a lubricant thinner than grease flowing through the gearbox and not leaking out. The stock bearings cause very little energy loss. The bulk of the energy is lost where the most heat is generated, and that's in the coils. Pick the low-hanging fruit first.

Btw, I put somewhere around 7000 miles on a stock gearbox (run on a "500 watt" motor), only once cleaning and replacing the grease inside. That gearbox was still running smoothly when I sold it as "used" a few years ago.