Efficient pedaling with Cyclone 36/48v

[EstebanUno]

Help me choose the right voltage, please.

I'm considering a Cyclone kit to tackle the mountain on which I live. Going anywhere entails a 2000 ft climb over 4 miles to get back home. I like the idea of the power of 36v so I can climb at speeds as high as 15mph, but I do like the exercise and efficiency of pedal assisting. Does the 960w kit have a smaller diameter driving sprocket option, or bigger diameter chainring to better match the human cadence of 70-90 rpm? Or some other way of compensating for this?

I can't find an efficiency chart for the 960w motor, but from what I've read on the forum it is the same motor as the 500w but the controller runs it 36v, not 24v. Without some other modification, wouldn't it spin 1 1/2 times faster than the 500W? If so, the efficient rpm range would fall in the 3300 to 4000 rpm, which would require a reduction to the crank of about 45 to 1. This compares to about 30 to 1 for the 24v, which I assume matches cadence properly in the efficient range. I calculated about 75-80rpm given 44-48 chainring and 14 tooth motor sprocket. 1/12 times that is too high!

Something has to give.

Esteban

 

[Dingo2024]

Hi Estoban,
from what I understand the planetary gearbox cyclone motors are all the same 500w, 960w 1200w just dependant upon voltage.
I run a 1200w (48v) set up on my commuter bike and spent lots of time calculateing the correct gearing and pedal cadence. I ended up having to get a custom sprocket made to bring it down enough (88rpm) Take a look at my build thread as I think its explained in there how I went about it.

hope this helps,

Ian

 

[boostjuice]

http://www.endless-sphere.com/forums/viewtopic.php?f=28&t=13907&start=15#p228350

By replacing the standard cyclone components with a 60T chainring and a 13T motor freewheel, pedal cadence can be reduced to ~75rpm when matching the best compromise motor rpm @ 36V [3250/(9.33*(60/13)) = 75.5rpm]

60T, 130mm BCD chainrings for US$20 http://www.vueltausa.com/components/chainrings/se-flat-chainrings/se-chain-ring-130mm-38-60t.html
You can adaptively mount these to a solid plate Cyclone chain-ring by drilling 10mm holes through it and using chainring bolts. Use the chainring itself as a template for accurate hole positioning.

EDIT: 60T 110mm BCD chainrings are also available for US$42. (More easily adapted to the standard Cyclone crankset) http://www.vueltausa.com/components/chainrings/se-flat-chainrings/se-chain-ring-110mm-34-60t.html

Odyssey 13T 30x1mm bore Freewheel http://www.danscomp.com/461085.php
30mmx1mm bore freewheel removal tool http://www.danscomp.com/811035.php

EDIT:
ACS Crossfire 13T http://aebike.com/product/acs-crossfire-freewheel-13t-gun-metal-sku-fw1250-qc30.htm
ACS Crossfire specific removal tool http://aebike.com/product/acs-crossfire-freewheel-removal-tool-sku-tl1260-qc30.htm

(The ACS Crossfire series of freewheels have just been released and are preferable over the Odyssey due to reduced freewheeling drag, lower noise, less bearing play and better machining tolerances. They are a huge step up in quality from the old ACS Claws series.)

 

[VSkudarnov]

Largest chainring you can get on the MTB pattern is 52T

 

[LI-ghtcycle]

Largest chainring you can get on the MTB pattern is 52T

Frame permitting yes, but on a Trek 930 I have successfully mounted a 60T Chain Ring, they are custom, but they do exist.

However, most modern frames won't allow this big of a ring because it will rub against the frame.

 

[EstebanUno]

from what I understand the planetary gearbox cyclone motors are all the same 500w, 960w 1200w just dependant upon voltage.

Ian, I read through your build thread. What a great looking/performing bike! I learned some things reading through it.

I have the same understanding regarding the the cyclone motors being the same. I picked that knowledge up from a great post by boostjuice here: http://endless-sphere.com/forums/viewtopic.php?f=28&t=13423&p=199887&hilit=long+post#p199887

I wonder why it is that the efficient rpm point on the 36v is 3250 (from boostjuice's comment in this thread) rather than around 3750 that simple math would give? boost? Are there any efficiency charts for the motor at 36v out there?

Thanks for all the information everyone. I see there are lots of ways to go with these kits. I'm slowly processing them.

Esteban

 

[Dingo2024]

thanks Esteban, I am sure if you get gearing set up correctly you will have a lot of fun on the cyclone setup.......I would think for hill climbing 36 or even 48v would be better though?

Ian

 

[boostjuice]

I wonder why it is that the efficient rpm point on the 36v is 3250 (from boostjuice's comment in this thread) rather than around 3750 that simple math would give? boost? Are there any efficiency charts for the motor at 36v out there?

Esteban

Yeah it seems a little unintuitive but it's because of the controllers inherent current limiting (at least in unmodified trim where the motor can demand more current than the controller is set to provide even at intermediate rpm when BEMF has risen substantially). If nominal line/battery current wasn't capped, then by applying 1.5times the voltage, you would indeed expect the rpm point at which power and efficiency to be in 'best compromise' to increase by ~1.5X when going from 24V>36V. In reality though, because nominal current is limited to the same amount whilst voltage is increased, the point at which peak power occurs is held a little lower in the rpm range relative to efficiency which actually does climb ~1.5X later (skewed toward higher rpm). The result is that the 'best compromise' rpm is lowered a bit relative to the voltage increase, because torque/power diminishes sooner relative to increasing efficiency. The 3250rpm figure I gave is just a very rough calculation based on known static gearing, cycle analyst power-in and speed readout together with seat-of-the-pants acceleration. I'd put an error on it of probably +/-250 motor rpm.

If you raised the controller current limit (by replacing or mounting additional shunt resistor/s in parallel with the existing ones) then best compromise rpm would rise. Which model/rating of FETs you have in your controller, will determine if it's safe to do this, and by how much.
Also, the info I reported Here on FETs/Caps utilised in the two different models of Cyclone external controllers no longer seems to hold true. I recently ordered another external 24V/36V controller and the FETs installed were a major downgrade on older batches. They are TO-220 package FETs rather than the beefier TO-247 used in older batches, and of a model number that I couldn't even find a datasheet for on the web!! Fkn penny pinchers...


EDIT: It just dawned on me that all of the above only really affects you assuming your trying to maximise load on the motor through accelerating hard and travelling at top speed most of the time - Which I is my style of riding . So if you're the sort of guy to just cruise lightly on the flats and accelerate lightly most of the time, then you probably won't be hitting the current limit ceiling very often. Therefore the best compromise motor speed in this case would actually be around 3750rpm. Depends how you ride I guess..

 

[Ebikeman]

Because I was just checking for wear in the gearbox, I thought I would make some measurements. So here is a video of me testing the Cyclone 650-1200W motor with the gearbox removed.
Obviously no load applied to the motor.
Measuring motor RPM and current at full throttle.
39.5V
5959 RPM (so as the gearbox is 9.33:1, output should be 638 RPM)
2.7A
[youtube]g-rLAq_xjT8[/youtube]

http://www.youtube.com/watch?v=g-rLAq_xjT8

For a comparison I have some measurements taken with the motor and gearbox driving through the chainring to the back wheel. Again the only load was from the drive chain.
Full throttle
Where I observed:
615 RPM at the gearbox output
175 RPM at the pedals
And at the wheel
668 RPM
4.5A

 

[boostjuice]

Interesting....

-Kv = ~150
-No load power consumption without gearbox = ~107W @ 39.5V
-Non-motive load power consumption with gearbox + chain based drivetrain + wheel = ~178W @ 39.5V
-Gearbox + chain based drivetrain + wheel adds enough load to drop motor speed (5959rpm - (9.33 x 615rpm)) = ~221rpm @ 39.5V

Could you please measure the Volts/Amps/rpm with just the gearbox installed. Maybe at 3 different rpm points (slow/intermediate/fast) so we can see the relationship of how the gearbox efficiency might change at different speeds, and also get a ballpark estimate of gearbox efficiency with factory grease.

I would do this myself if my own motor didn't have the gearbox removed and the shaft teeth machined off to fit a #25 sprocket.

 

[Ebikeman]

OK boostjuice here are some more figures.

I made 3 measurements without the gearbox:

1022 RPM 0.49A 39.8V 19W
2980 RPM 1.39A 39.55V 55W
5912 RPM 2.71A 39.33V 106W

And with the gearbox factory grease, at the output shaft of the box:

109 RPM 0.53A 39.73V 21W
319 RPM 1.52A 39.52V 60W
633 RPM 3.13A 39.33V 123W

The absolute accuracy of the figures is reliant on the instruments I am using.

Turnigy 130A powermeter
Digital Tachometer DT-2234C+

But the readings do seem to be reproducible, so their relative values should be OK.
It looks like I am using another 50W with load of the drive chain and N360 hub and wheel.

 

[boostjuice]

Good work Ebikeman!

More data-points are needed to accurately characterize a trend line, but it's still some useful data....

 

[Ebikeman]

Hi boostjuice et al

I've made a few more measurements with and without the gearbox. I tried to take readings at comparative points but there may be a few RPM between them.
First set with gearbox so readings are at the box output shaft. I took some temperatures as well.
Gearbox temp started at 22C end point 28C. I left the motor running for ~ 1 minute at each setting to get a reasonable stable reading, but the battery voltage may have sagged 0.2-0.4V over the compete test. I will just post the power measurements at each test point. From my previous test I made a reading at ~100 RPM of 21W

RPM WATTS
50 11
100 21
150 28
200 37
250 46
300 52
350 65
400 74
450 82
500 91
550 102
600 112

I took the gearbox off and measured at the motor. Temperature at start 29C end 42C
Previous test gave a reading at ~ 1000 RPM of 19W

RPM WATTS
466 9
1000 19
1400 24
1800 30
2300 38
2800 46
3200 52
3700 60
4200 70
4600 77
5100 85
5600 95

 

[VSkudarnov]

Please have a look at this nice 9T freewheel sprocket:

http://www.boosty.ch/en/conversatio...age=flypage.tpl&product_id=216&category_id=17
It seems it is based on a one-way roller bearing.

Maybe it will fit the cyclone shaft?
Then the cadency can be lowered even for 48V 1200W Cyclone kit - without large 60T chainrings.

 

[EstebanUno]

Please have a look at this nice 9T freewheel sprocket:

Then the cadency can be lowered even for 48V 1200W Cyclone kit - without large 60T chainrings.

Great find! 8/9/10 teeth. I wonder who makes it?

 

[dumbass]

Please have a look at this nice 9T freewheel sprocket:
page=shop.product_details&flypage=flypage.tpl&product_id=216&category_id=17[/url]
It seems it is based on a one-way roller bearing.

Maybe it will fit the cyclone shaft?
Then the cadency can be lowered even for 48V 1200W Cyclone kit - without large 60T chainrings.

Nice to see a smaller procket then a 13t being made. That being said though I think your a lot safer just changing your chainring even if you have to go with a 60t.

Bob

 

[VSkudarnov]

Cyclone 1200W is typically put on FR/DH bikes where 40-44T is max... 60T will look funny...

 

[boostjuice]

Please have a look at this nice 9T freewheel sprocket:

http://www.boosty.ch/en/conversatio...age=flypage.tpl&product_id=216&category_id=17
It seems it is based on a one-way roller bearing.

Maybe it will fit the cyclone shaft?
Then the cadence can be lowered even for 48V 1200W Cyclone kit - without large 60T chainrings.

The Cyclone planetary gearbox output shaft is 20mm OD. I am certain that the ID of this one-way needle bearing is smaller than 20mm just by eyeballing it, so its too small to slide straight on. Also, as the vast majority of one-way needle bearings grip the shaft directly, they require a smooth precision-ground shaft surface around it's entire circumference. With the exception of the 'double freewheel' motor gearbox combo sold by Cyclone, there is a 6mm key-way machined into the gearbox output shaft in order to transfer torque to the freewheel adapter. This means that the key-way slot makes the standard shaft unusable for even an appropriately sized 20mm ID one-way needle bearing for the more commonly purchased/owned single freewheel gearbox versions of these motors.

The only way i could see getting one of these 8/9/10T one-way needle bearing based freewheel sprockets onto a cyclone gearbox OP shaft is by turning down its OD on a lathe to a precision tolerance, and finishing the surface with with a fine polish. You would also need to confirm that the reduced diameter would be below the base of the keyway slot so that an entirely smooth circumference could be obtained. Unfortunately, all this is far beyond the capabilities of those looking for an easy mod, so fitting a larger chainring with the smallest possible ratchet freewheel (13T) is oh so much simpler.

Below are some photos by the photoholic ES member 'MrBill' showing how Cyclone mount a 'double freewheel' arrangement on a cyclone motor/gearbox combo that comes with a keyway free shaft (a factory option by 'Headline Electric' who manufacture these motors for 'Cyclone'). Unfortunately another ES member broke his 20mm ID one way needle bearing under even Cyclone prescribed power levels (See here). The one-way bearing within the newly discovered 8/9/10 tooth sprocket has an even smaller ID and is narrower. It would therefore feature an even lower torque rating, thus likely leading to quick failure even if it was painstakingly custom adapted to to the gearbox OP shaft.

And, Single freewheel vs Double freewheel options pictured together

 

[VSkudarnov]

It would therefore feature an even lower torque rating, thus likely leading to quick failure even if it was painstakingly custom adapted to to the gearbox OP shaft.

that's why they sell replacement bearings separately

 

[darkone040]

so will changing to a 13t motor freewheel affect top speed?

im trying to somehow match my pedal cadence to my 960w 36v cyclone setup
was just going to throw a 60t chainwheel on then realized that it would have a 33mm radius bigger that the 44t one that is on there
so i went out with the trusty ruler and i think ive got at most 15mm radius to play with which should get me a 50t
for that i would have to shift the engine mount back near the tyre and put a 3mm spacer to drop the engine mount down
i could possibly elongate the engine mounts holes to drop it down a little bit more but i would like to avoid that..

 

[EstebanUno]

so will changing to a 13t motor freewheel affect top speed?

Either way you approach the reduction, at the motor or at the crank, lowering the rpm reduces the top speed, all other things being equal. But you can compensate by increasing the chainring size that connects to the rear cassette, or by using a smaller tooth-count top gear cog in the cassette, if need be.

Since you want to match pedal cadence you have to reduce the rpm at the crank to a comfortable cadence. So in a sense this value is fixed, based on the pedal cadence you are seeking. You can use a bicycle gear calculator to find the combination that yields the speed you want based on that crank rpm. Example: by replacing the motor freewheel and resizing the outer chainring you calculate 80 rpm at crank. If middle chainring is 48 teeth, and smallest cog is 12, the wheel will turn (48/12) or 4 times faster than the crank, or 320 rpm. For a 26" wheel the mph calculates to around 25mph. 50/11 yields about 28 mph.

im trying to somehow match my pedal cadence to my 960w 36v cyclone setup
was just going to throw a 60t chainwheel on then realized that it would have a 33mm radius bigger that the 44t one that is on there
so i went out with the trusty ruler and i think ive got at most 15mm radius to play with which should get me a 50t
for that i would have to shift the engine mount back near the tyre and put a 3mm spacer to drop the engine mount down
i could possibly elongate the engine mounts holes to drop it down a little bit more but i would like to avoid that..