Steel-geared hub motor?

[Lebowski]

pitty I don't have a second controller lying around (like I put on the recumbent, 100 to 120V, about 30 to 40A rms)

the 1000uH is really crazy, how much erpm does it do for 400rpm wheel speed ?

There are 8 magnetic sections here... even if it's a 7:1 reduction, consider that a 16 pole 5:1 MAC motor will spin up to 560rpm at the wheel, at least, on an EB3 infineon-like controller. That's a pretty insane eRPM..

How do i know? i hit 45mph on a MAC 8T on 72v.

So I don't think that eRPM is a limiting factor here..

Well.. 400rpm at the wheel with a 7:1 means 2800 rpm on the actual motor. With the 32 magnets the e-rpms is 2800 * 16, which comes down to 750 Hz.

At 750 Hz the impedance of a 1000uH inductor is 4.7 Ohm. So with 15A the voltage drop across the inductor is 70V.... hardly any room left for backemf.

The erpms are not the problem, the large inductance is.

 

[wil]

Where have you gotten your figure of 32 magnets from, Is eRPM not motor pole pairs * rpm, so 12, /2 for pairs., = 6
2800*6=16800 rpm
16800/60=280 rps OR hz

280 hz to rad/s = 1760 rad/s
Then inductance = 1760 * 1000 * 10^-6
=1.76 Ohm impedance, giving a more resonable voltage drop at 15 amp of
=26.4 volts

Or is my understanding of eRPM / the motor wrong?

 

[Lebowski]

I counted 32 magnets in the picture from the previous page, and I assumed the polatiry alternates...

I went back to the first page and there see 8 magnets, which makes the voltage drop 17.5V @ 15A

Anyway, point remains, there's a huge voltage drop across the inductance.

 

[crossbreak]

tried to ride it today.. it wont go.. controller says hall sensor error. It spins fine with little amps (0.9A the same as sensorless) and in the right direction.. as soon as i apply heavy load it stops and says said error...i have no clue what i do wrong. Battery current is set to 20A, phase to 30A, voltage is 45V.. can't be too much

Strange thing: it spins with both 60° and 120° hall setting setup. I have no clue how to figure out which one it has. When I turn the motor by hand, it switches the halls like A , A+B, B , B+C, C , C+A, A and so on.

from what i see here http://endless-sphere.com/w/index.php/Determining_the_Wiring_for_a_Brushless_Motor it should be 120° since it never occurs that all 3 are HIGH or LOW at once

 

[crossbreak]

tested the Kelly with the Bafang CST 500W. Works flawlessly on the road. Next i will try is a sensorless X8m06c, if that doesn't work either i will try an S06 Sine controller

 

[crossbreak]

the sensorless S06 says: Error 06: Motor error

Finally a brake through: It runs fine with x8m06c controllers in sensorless.. the last one i tested.. that's Murphy's law

It even starts from a dead stop at a 10% grad hill perfectly... all sensorless. I'm impressed. Noise is quite low either. Only at very low speeds <2kph there is some resonance.. but hey, i'm fine with that.

Now i want 74V/15A. but there is no x8m06c with such specs? I found no 6-Fet that is ok with up to 90V. Smallest I found is 12 Fet from Greentime

Maybe I can solder some IRFB4410 and 100V caps to a KU63 / KU65+ change the resistors of the LM317? Or even put a resistor in series?

 

[circuit]

Maybe I can solder some IRFB4410 and 100V caps to a KU63 / KU65+ change the resistors of the LM317? Or even put a resistor in series?

Or you can do it this way:
http://translate.google.com/translate?sl=lt&tl=en&js=y&prev=_t&hl=en&ie=UTF-8&u=http%3A%2F%2Fe-motion.lt%2F2013%2F09%2F02%2Fsudedamo-dviracio-tiuningas-22-kmh-i-40-kmh%2F&edit-text=

 

[crossbreak]

yep.. maybe worth a try.. but i have no clue how to do it right.. have to really read the article. Guess it's much harder than a simple delta/wye mod

 

[circuit]

It is quite simple actually, you just have to find where middle of each winding is.

 

[crossbreak]

ok some update about the s06 controller: After correctly setting the pole number of the motor, it now works on the bench. This needs some further testing but i am confident to get it work wíth this sensorless one made for 72V: http://www.bmsbattery.com/controller/551-s06-250w-imitation-torque-square-wave-controller.html

A simple shunt mod (cut one of the two) can reduce the amps from 28 to 14A.. BUT these controllers seem to have slow throttle response.. just tried it

 

[circuit]

By the way, do you have a picture of other side of the stator, seen here? It would be interesting to see how it is all connected inside.

 

[crossbreak]

no..pressing out the stator was impossibe for me since i could not loosen all 6 bolts that fasten the axle on the back side. One of them was just too tight.. i damaged it's head it so much I would have to drill it.

 

[circuit]

Ah, I just use a dremel on such screws.. Just cut a groove in to accept flat screwdriver. Once that was not enough, so additional heat helped.

 

[crossbreak]

it tried both heatgun and dremel..no chance... thanks for the tip anyway I think i'll drill it as soon as i open it again... ATM i do not indent doing so

 

[wil]

Any update on running at 72v?

 

[crossbreak]

nope. I tried a 15Fet sensorless with cut shunt at 74V, still does 24A which seems to be too much IMO. But runs fine. Just a very big box for this little motor. I'll try a sensored infineon as soon as i have time for that. ATM it runs fine with a KU93 controller at 30amp 48V. Just slow, only ~41 kph. The advertised 45 kph are not true for a 26" rim, maybe for 28".

 

[neptronix]

Mind posting a video of the acceleration?

Does anyone have an exact figure for the weight of the hub as well?

 

[crossbreak]

sry too much to do with other stuff ATM. acceleration isnt much different from an SWXH bafang at this power level.. the SWXH just overheats much faster and wear it's gears

I measrued the weight with a kitchen scale... it was slightly over 3200g which is quite accurate. Can't tell you more since it is in a wheel now

 

[crossbreak]

about the s06 sensorless controllers: It does not show error06 (motor error) but sometimes failes (squeezes) during acceleration. Commutation error..as always So far I got no motor that runs 100% with this controller at 100% if the usage. in 99.99% they are fine... but what is that worth if it fails from time to time. I tested 5 units so i can exclude a broken one.

 

[neptronix]

Hmm... so you mean to say is that it's less powerful than the swxh?
That's sad if so..

The outrider motor seems fairly impressive, per the ebikes.ca simulator anyway.

 

[crossbreak]

interesting..is it helical steel geared too?

i didn't say the swxh feels stronger. I think they are about equal. SWXH will just overheat faster

 

[neptronix]

Oh, i misread your english. Sorry.

 

[wil]

Looking at the ebikes outrider its a smaller diameter, 118mm, compared to the ~135 that this motor is.
It is also 100mm wide, so it has obviously been designed for a front hub.
Being significantly smaller volume wise it probably is less able to go above the rated 250watt output, also lighter so less thermal mass = overheat quicker

Probably best to stick with either a SWXH or this zelenavozilla one unless your trying to go ultimate stealth

I might order one of these within the next week, depending on postage costs etc!

 

[neptronix]

Well, the difference between a front motor and rear in geared motor land is moreso just the axle width and accommodation for the freewheel threads. Don't base a geared motor's power on the axle diameter.

The outrider motor is rated for lower power AFAIK; not as tall either. Can't really compare the two. But one motor that compares to this is the Q128H ( 800W rated and the spec sheet indicates that realistically it's good for 700-750W continuous near peak efficiency - but it may be a stronger motor than this, and a bit lighter as well. )

 

[crossbreak]

i had the Q128 front motor. It is really weak compared to an SWXH. sold it. I recently built a bike with Q100-CST, which feels equal to the Q128, but it is smaller+more lightweight.

If I would want a front motor and could disclaim a disc brake, i would not buy any other motor than the Q75 anymore. it's tiny and has awesome torque. Plus it pulls less than half in amp no load at 48V...which tells me it must have thin lamination/awesome design. With 48V it is at ~35.000 e-rpm, still does not pull excessive current with no load. Sadly it is sensorless and a bit bitchy. It wont run well with any controller other than KU6x, they all suffer.. even the Infineons. Sadly the KU6x controllers dont really have a fast throttle response.