Bafang-BPM and MAC comparison for a mid-drive

speedmd said:
I could easily remove about 1/4 of its weight.
Click to expand...

Agree, 1/3 is possible in most mass produced items without too much structural loss. That would bring it down to around 3- 3.5 KG? I don't have one to look over and only going by photos but could aluminum replace the axle possibly as it is used mainly as a stator spacer / bearing supports and also thinking the planet gear support can be made more swiss cheese style.
Click to expand...
I may be misinterpreting spinningmagnets statement, but I took it to mean that the rotor bowl could be reduced by 1/4. Not that the whole motor could be reduced by 1/4.

My point was that while there is some scope for lightening, it is not all that much. The major masses are the stator core, the stator copper, the magnets, and the magnet backing. These can't really be lightened without compromising function or fundamentally changing the design of the motor. The other parts, the axle, the shell, the clutch, the stator support are either aluminum or relatively small. Sure the axle is heavier that it needs to be, and some weight could be turned off the shell, especially if they do not have to hold a bike up, but it just isn't that much compared to the copper, laminations, magnets, or backing iron.

I'd be surprised if 1/4 of the weight of the BPM could be eliminated and completely flummoxed if 1/3 could, even for a mid drive conversion. I'd love to see it worked out, so if you have specific ideas on how to do it, please identify what parts are modified and how to save 1.5 kg from a 4.6 kg BPM.
 
I apologize for not being more clear. I meant that I could reduce the weight of just the magnet bowl on the MAC to 3/4ths of its original weight.

Pic of the solid STEEL magnet-bowl on the MAC:
file.php


And here's the BPM, which has already been skeletonized:
file.php
 
Thanks for the clarification. Not having one in hand and no experience with a hub motor makes for some bad assumptions on my part. These look like they will make good reliable mid drives, but I am starting to rule them out due to the mass/cooling issues for a ultralight build.

For weight savings, the axle is not needed in steel unless I am missing something. It can also be as short as needed to just hold in the supports. The case can most likely be thinned greatly as there is not spoke/weight loads on it. The ring gear can be swiss cheesed. On and on..

When I was building steel frames, it was not easy, but I could loose more than 40% of the weight of the stock shimano forged drop outs. The same on for most fork crowns. Even destroyed a few to make sure the adjacent tubes would be first to fail. Certainly not something worth doing on a commuter, but when world championships or olympic hopefuls / medals are at stake the trade offs are worth considering. I would weigh each and every part, and put the work into the parts that would save the most weight first.

If these cant handle significant power they may not be a good choice given all the extra work needed to take off a few grams here and there.
 
could aluminum replace the axle
Click to expand...

That's an intriguing idea, speedmd. Since the axle in the shaft-drive conversion is not hollow, and performing jumps with the bike will no longer stress the motor, it may be possible. Aluminum would be much easier to machine than steel. 17mm diameter is pretty beefy.
 
Hard to tell what can be done without one in hand. Some assumptions I have are that the axle shaft is not needed at all on the non output side. All support structures (rotor and stator) can come from a new turned housing. Stator center has tons of extra material (both extra steel and aluminum) in it just to reach the axle. Clutch looks like it can be greatly reduced also. Looking like a complete redesign to maximize weigh loss. Not enough tweekers out there to make a new high power motor for mid drives unfortunately. Bosch and bafang kits seem just too conservative.
 
So what is the max RPM the BPM and MAC motors can handle before eddy current losses get too high ? And what voltage does that correspond to ?
 
The RPM's of the motor depend on many factors, you can go to a higher voltage to get more RPM's, but then you can select a slower kV of winding to keep the RPMs down at the lower RPM's even after you begin using the higher volts.

One of the desirable features of the MAC and the Bafang-BPM is that they are available in many kVs. The kV is how many RPMs you get per volt that you apply to the motor. The popular vendor cell_man stocks MACs in 5 different kVs from a 6-turn to a 12-turn (6,7,8,10,12). The Bafang-BPM is often stocked by their dealers with just two kV's available, but the factory makes them with nine selections (8-turn to 16-turn), and it should be possible to order any kV that is made.

I don't have the exact figures, but Miles and crossbreak said the BPM can run to 2,000-RPMs, and the old MAC can run to 1,000-RPM's but the new ones have thinner laminations, so the new MACs "should" be able to run to 2,000-RPMs also.

A 26-inch wheel running at 26-MPH will be spinning about 336-RPM's. But, since geared motors have the motor spinning about 5 times faster than the wheel, the motor would be spinning 1680-RPMs.

This is also one of the great things about a non-hub build. You can change the sprocket tooth-counts between the motor and the wheel, so...you can adjust the motors top efficient RPM to coincide with the top speed that the bike is designed to maintain. You can then change the voltage if you want, and also adjust the sprocket tooth-counts to still keep the motor-RPMs efficient while the bikes top-speed stays where you want it.
 
according to the ebikes.ca simulator, the eZee is still around 5% less efficient than the BPM. Still, If it has the thin lams a conversion cold be quite satisfying. 2700W_in /2000W_out continuous should be not a problem.

I have a hard time converting the BPM since the rotor bearings spin on the axle, not on the stator like (i think) it the MAC is designed.

some weight measurements of the BPM2... ~100g can be saved by removing the spoke flange, but I can't see what else could be removed. Already designed for saving weight http://endless-sphere.com/forums/viewtopic.php?f=28&t=45245&start=600#p769318
 
crossbreak said:
according to the ebikes.ca simulator, the eZee is still around 5% less efficient than the BPM. Still, If it has the thin lams a conversion cold be quite satisfying. 2700W_in /2000W_out continuous should be not a problem.

I have a hard time converting the BPM since the rotor bearings spin on the axle, not on the stator like (i think) it the MAC is designed.

some weight measurements of the BPM2... ~100g can be saved by removing the spoke flange, but I can't see what else could be removed. Already designed for saving weight http://endless-sphere.com/forums/viewtopic.php?f=28&t=45245&start=600#p769318
Click to expand...
What are the internal differences from the BPM to the BPM2? Has anyone had a look at both. I'm just curious why they introduced the BPM2 since it does not appear to offer any advantages over the BPM. Cost cutting maybe? Or what?
 
and the old MAC can run to 1,000-RPM's but the new ones have thinner laminations, so the new MACs "should" be able to run to 2,000-RPMs also.
Click to expand...

Please get the lamination thickness verified by cell man or measure an actual "new improved stator" before inserting this as truth.

The high pole count of the Mac seems excellent for torque density but they suffer very high losses when the low turn count (6T) motors are run on high voltages (80v +). Brian was kind enough to let us pull his 6T Mac down and weigh and measure stuff. With the bare motor stripped of planetary gears and outer hub, with just the bare motor in the vice free revving in the air, at 80v the motor got to hot to touch, not doing any work, and temp still climbing (i should of measured every thing phase resistance, watts of loss etc. but i was slack and working on other stuff at the time so i can not add much helpful information). And that"s without gear friction heat and all sealed up in a hot hub. I think these are a good little motor but they can not be over volted and revved, they just cook themselves. With such a high pole count they are really screaming out for thinner laminations, like the over twice the price V2 BMC has.
So what is the lamination thickness of the "improved stator" please Cell man?
Zappy
 
I believe the BPM-2 is a freehub? EDIT: the CST is the Bafang freehub, the BPM-2 is not (thanks -dg).

I thought I had posted it here (must have been somewhere else), Paul PMed me and verified the new MACs have 0.30mm thick lams, instead of the 0.50mm ones used previously. I have yet to read about anyone here who has purchased a new one and measured/counted the lams.
 
spinningmagnets said:
I believe the BPM-2 is a freehub? (instead of a multi-speed freewheel, like the BPM-1)
Click to expand...
BPM has freewheel and sidecover fastened with screws. BPM2 has freewheel and threaded sidecover that requires the special large face wrench. CST has a freehub (for cassette) and sidecover held with screws. Supposedly the CST has internal differences too.
 
Has anyone suffered from broken axle with the bpm?

It only has 12mm diameter and I've read member having broken axle few times. That's why I'm hesitating with the motor
 
If you are taking jumps in a hardtail frame, that would put the most stress on the rear axle. If you plan to use the BPM on a street commuter and use a full-suspension frame, then you should be fine. There are literally millions of these in use in China, and quite a few in North America.

Sorry for the late reply arcticfly, but crossbreak is documenting everything that has to be done to make the BPM a shaft-drive. However, Whiplash has built a successful mid-drive by using a hub in its conventional configuration as a shell-drive.
 
It'll be on a full suspension frame for a dual bpm 328rpm at 36V with the 2 new 35A 72V controller from bmsbattery:
http://www.bmsbattery.com/controller/607-s06-250w-imitation-torque-square-wave-controller.html

No risks of failure spoted?
 
do you have an idea of the actual KV of the MAC? I posted the loaded rpm @ 36V in the wiki... but this is useless number. http://endless-sphere.com/w/index.php/Puma/MAC/BMC

The KV of one motor would be fine... the rest could be extrapolated
 
OK since I got no answer I did some guess work: The 8T MAC seems to behave like a BMC-V2-Speed, so this should be 8T as well. This motor can be found at in the ebikes.ca simulator. From this simulator i read the KV of this motor: 10.4 rpm/V geared or 52 KV for the motor itself. From this we can extrapolate the data for the other winds:

6T: 69.3 KV
8T: 52 KV
10T: 41.6 KV
12T: 34.7 KV

iceman told me that the no load consumption below 2000rpm seems to be acceptable. Taking this into account, the most powerful setup for 12s Lipo that can be geared to human cadence with a normal 44T chainwheel / 13T freewheel is a 6T MAC. It should spin at 360rpm loaded at it's maximum output power of 2440W ( 3350W input, 910W loss) at 80A battery current.

The same case for 20s Lipo: 10T MAC, 48A current, Same power.

BUT now the whole thing gets confusing. Seems like the "upgraded" MAC has a thicker stator stack. So the KV should be lower for the same turns. Asked ecrazyman, hope to get a reply soon to clarify this
 
You must log in or register to reply here.

Similar threads

R
    • Like
A critique of mid-drive riding recommendations (and hub drive criticisms)
Replies
15
Views
5,226
Bullfrog
B
R
    • Like
GMAC/MAC hub motor servicing + freewheel modification
Replies
1
Views
906
tom355
tom355
EVGator
EEB Mid Drive Bike - Complete or Part out
Replies
3
Views
1,021
skeetab5780
skeetab5780
R
    • Like
Really good article comparing Bafang vs. Tongsheng vs. CYC
Replies
12
Views
10,774
999zip999
999zip999
A
best Bafang Mid Drive option for off-road trike? No human power input
Replies
8
Views
2,813
Bullfrog
B
Back
Top