Converting a hubmotor to a middrivemotor

I wanted to have a 2:1 reduction ratio before the crank, just to be sure I had enough torque.
Besides, if I run the chain directly from motor to the rear wheel, when I'm on the lowest gear the chain will rub the tire.

I bought the motor 3 months ago but I think I didn't get the latest version. At least, I have the blue composite gears but I learned from EM3EV that the new version has 2 smaller bearings supporting the rotor, while the older version had one large bearing doing the job.
And my motor winding is made of 3 smaller gages of wire instead of a single larger gage. But that doesn't necessarily indicates it's newer or older.
I ordered a spare stator that I'll wind as a 9t I think. I also have a set of new hall sensors that I'm going to put in, hence my presence here to gather info on exactly where I place them.
 
I think that the Adaptto limits the max RPM of the motor, maybe because it measures increased inefficiency past some RPM.
Early on, when I began tuning the controller, there was times during auto-detect where the RPM went to twice the normal maximum I get now.
But I never saw that again. The motor speed is still reasonable now, it's just that low RPM behavior that drives me nuts.
 
Oh and another reason why I have 2 chains. I needed a jackshaft to put the freewheel for the pedals. So now I can interchange the pedals or the footpegs easily.

Changing the gear ratio is also very easy, I machine my own sprockets for the primary chain, it's a #219. With the adjustable length motor mount, I do not need a chain tensioner for the primary chain.
 
I'm wondering why this MAC motor is so difficult to drive for a controller.
Could it be because it has concentrated windings?
Do any of the other popular hub motors also have concentrated windings, or never?

The freewheel is in for something anyway, it greatly reduces the inertia in one direction only, maybe that confuses the controller.

Would there be still a better placement for the hall sensors? Maybe they get adversely affected by the changing magnetic fields there right on the edge of the stator. Bury them lower on the face of the stator?

Slot spacing for the sensors should be in multiples of 3, right? But what slot to use as the #1? (Assuming that we're gonna place the sensors into the slots)
If we refer to the winding scheme:
MAC winding.jpg

Placing the sensor between poles #36 and #1 will have varying magnetic fields constantly affecting the sensors.
But if the sensor is placed between poles #1 and #2, its electrical environment will be much more quiet because, due to the concentrated windings, the pole #1 is always of the opposite polarity from the pole #2.

If we're gonna place the sensors on the face of the pole, then pole #2 has the advantage because it is flanked by two poles of opposite polarity, always.
I didn't think about this during the time that my motor was dissassembled but it would be interesting to know where the sensors were placed exactly, by the manufacturer. Anyone has a MAC on his bench?

I don't know much about motor design but I wanted to get this out, just in case...
 
@altair: i dont know if you followed all the thread, but if you look at post #905 you can clearly see where i moved the halls. this was a picture of the new stator.
@crossbreak: good to know that you tried the kelly kls with the mac now as well. i was waiting desperately for your result. good (bad?) that you confirmed my findings: it's impossible to go above 2krpm. at least not sustained. when i release the throttle at wot it revs up to 2.3k for the blink of an eye. also if i test it from a stand still with the tire lifted. going wot makes the motor rev up to 2.3k and fall back to 2k almost instantly.
i manged to find some settings for the kelly that works nicely (still to be tuned). check my last post of the third link in my signature.
 
Altair said:
At least, I have the blue composite gears but I learned from EM3EV that the new version has 2 smaller bearings supporting the rotor, while the older version had one large bearing doing the job.
correct. and the color of the gears says little about the motor itself. :(
 
@izeman, yes I have read most of the thread, and I was aware of the change that you made to the placement of your halls.
I am very happy to learn that it solved your problem. I will probably do this mod to my motor.
Can you tell me if you placed your sensors in the center tooth of each group of 3, or if you placed them in either the first or the third one?
 
OK izeman.
But are they still standing up, or you laid them flat on top of the laminations?

I wonder what the magnetic fields look like there on the top edge of the lams. Who has the software to simulate it here on ES?

I also wonder how it would work if they were placed further outwards, right over the magnet line, in a quieter magnetic environment. They could be supported by a PCB that is centered on the shaft of the motor for easy advance adjustment.

EDIT: Now that I have read the thread referred to by Fechter earlier, many of my questions have been answered. Thanks Fechter.
 
It is quite easy. Take a heat gun and soften the epoxy. Then lift the hall's board with halls still attached to it. Heat the solder and move the halls further to the board (about 4-5mm). Move the board some degrees CW and glue it to the windings. That's how i did it. So the halls moved 5mm up and 4mm to the side. They were NOT bent. This would not be good as they imho sense the magnetic field on the small flat side normally facing to the magnets.
 
StudEbiker said:
What is the advantage of the BPM over the MAC?? I thought the whole point of the past couple of pages of this thread was trying to figure out a way to use the MAC motor because of how well suited the 10 and 12t versions would be for this application. :?

Is the mac 8t made differently inside apart form the winding?

thinking of doing the conversion over the winter , right now i have the hub spinning mid-frame witch is not ideal , i wish ive seen this wonderful thread before

anyone had issues with their setups so far or its proven to be reliable ?
 
as far as i know, there are only few people who still use this setup. the cyclone 3kw kit works the same as a converted geared hub, yet it is wider.

the MAC motor that you can buy to date (V4??) is different from the one shown in this thread and may or be not convertable. Izeman had bad experience with it as far as i can remember
 
yes. those are different, similar to the bmps i guess. harder to convert but still possible.
i would do thr conversion only if you have the skills and tools (lathe,mill, etc...). it's important to work very precisely to avoid rubbing and unwanted noise.


Gesendet von iPad mit Tapatalk Pro
 
:shock: oh! well , i dont have machining tools (yet) thanks for the info , the setup works great already , but always striving for best :p
the fact the cyclone is wide is what made me choose a hub to use as middrive/midmounted , with the loop frame i can have 10'' diameter motor , but not much more than 3-4'' wide
 
Hi Crossbreak,

after avidly devouring your thread on this subject i started looking into doing one for my self and through all the reading, maybe I missed it and you can correct me if I'm wrong the stator laminates of the bpm are said to be 0.5mm.

I may have found another version of the bpm that claims 0.35mm lams and curved magnets on alibaba unless they are lying here is the link anyway Im currently communicating with the seller about getting one sent to australia for conversion on my FS bike i was looking at an rc drive but this modification of a hub is so much simpler.

here is the alibaba link https://www.alibaba.com/product-detail/Greenpedel-48V-500w-BPM-brushless-geared_60716275952.html?spm=a2700.7724838.2017115.2.396c049j33pPo

another thought i had is about the failure of the nylon gears, i suspect this is likely from high power levels being applied when the nylon is too hot. have you played with the replacement steel gears that are available ?

and I love your idea of using one of these for an epicyclic variable ratio transmission that you mention in another thread
current ride is a 1000w rc motor friction drive using a vesc.
 
This version has the 6 screw casing available with front wheel configuration or rear with screw thread freewheel mount or casset and comes in 4 different winds the available specs are:

9x 11 turn 260rpm @36v
10x 10 turn 280rpm @36v
11x 9 turn 305rpm @36v
12x 8 turn 336rpm @36v

I presume the first spec number ??x designates number of strands its wound with, all using multiple strands of the same wire.

I'm going to order one once I work out which winding spec to go for, it claims to have the 0.35mm lams as well .
 
when the bbshd appeared, this thread went out of my sight. the really interesting thing here was the dual jackshaft thing though. Bosch uses this dual drive stuff today even with a planetary drive within the bottom bracket, but not with a powerfull motor so its somehow meaningless. Still its a nice concept with an outrunner. Im sure it will find its niche
 
Im interested in building an epicyclic variable speed drive as you discussed in one of your other threads as I think this would yield the ultimate in efficiency and usability for a left ide drive solution that bypasses the existing gear set allowing a strong 219h or belt drive to the rear brake mounting, so far I'm envisioning a large drone motor they are very short along the axis with a concentric small drone motor rotor with a custom ironless stator as the generator/ ratio adjustment motor coupled to a BPM planetary gearset, with output to belt/chain this allows a single speed reduction to the rear wheel and variable ratio with the best torque from stand still and would allow a sensorless controller to take off from a dead stop easily with wheelie levels of torque. this could also be built in to a hub motor for those diehard hub lovers
 
Hi Crossbreak,

maybe this should go in a new thread as it apears to be the holey grail of motors for single stage reduction to the rear wheel but could be easily built in to a hub as well, the idea is not new, using two motors in an epicyclic configuration to produce a variable ratio drive unit without the need for any changing of gears, there is a nice paper on the subject at the link below for those with interest.

https://www.podbike.com/downloads/Full-10.pdf

It mirrors your earlier thoughts of using a geared hub motor in combination with another larger external motor to produce a variable ratio constant motor speed drive using two motors, one small motor driving the sun gear and a larger one driving the ring gear that can produce a truly variable speed drive output from the planet gear carrier.
 
I just ordered a code 8 bpm for a conversion, they lied about the 0.35mm lams on their site though, has 0.5mm lams but the price is good $86 usd and 56 freight to Australia, I have lathe access and an old engineer friend who is an excellent driver of the lathe, he taught me all the machining skills I know from age 8 onwards. So a new shaft, stator to rotor bearing holder and sproket holder should be fairly easy, I notice that Staton inc has copied your freewheel and nexus 3 notch sproket holder and now sell one in steel for 30 somthing usd. add freight and its cheaper to make it here as i also have access to a very goodie filled scrap pile, I'll put up a build thread when I get into it . will use a vesc to feed it. The guy from the chinese factory was extremely helpful and dug out all the specs I asked for about the motor, I prefer novel projects and a bbshd is too expensive for me. im thinking of a fluid cooling system that has a convection only flow loop inside a new stator holder and an external radiator for the nasty australian summers that can get above 40 degrees c regularly. cpu radiators are cheap these days.
 
Back
Top