Converting a hubmotor to a middrivemotor

bØb said:
I have never seen a hub motor in person, only the pictures on this forum, so it will probably get interesting. :roll:

I'll make sure we help you with more pics :D Feel free to ask anything and don't saw any axles before you are done, made this mistake once, now a have a "spare motor" without axle :(

mr.electric said:
The actual windings are ok.
nice to hear that, keep us posted!

file.php

The M4 screw broke when trying a wheelie in the highest gear - just as I thought. I'll use now three M4 screws, this will last.
 

I am not sure about this but I am wondering if the tension wheel shown right at the master link in the attachment here may be better used as a idler wheel (sorry not sure if my terms are correct here in my description) for taking up the slack on the portion of the assembly that would be more towards the left side of the photo. I am thinking of chain stretch down the road and if the idler wheel was used to take care of the slack like a rear derailier on a bike the idler wheel itself also would not be under so much stress. I may be missing something here but trying to picture the direction of power transfer etc through the chain and sprockets and thought I would put it out there for you to think about.

Following this with interest and wishing you good luck with the results.
 
waynebergman said:
file.php

I am not sure about this but I am wondering if the tension wheel shown right at the master link in the attachment here may be better used as a idler wheel (sorry not sure if my terms are correct here in my description) for taking up the slack on the portion of the assembly that would be more towards the left side of the photo. I am thinking of chain stretch down the road and if the idler wheel was used to take care of the slack like a rear derailier on a bike the idler wheel itself also would not be under so much stress. I may be missing something here but trying to picture the direction of power transfer etc through the chain and sprockets and thought I would put it out there for you to think about.

The cranks drive the motor shaft on this side, not the other way around. The small sprocket is an idler though. Chain pull is on the left. There is only one sprocket on the crank shaft. The motor drives the rearwheel on the other side.

This way gearing the motor right gets much simpler and the chain drive to the wheel gets more robust compared to pushing all the power through the crank shaft (=less chain pull). Also you do not need a freewheel crank (which fails frequently with these overpowered drives, I AM using my stock octalink BB and cranks) and you need one less stage for the motor and get a better drive train efficiency here. It's win-win situation :D Discussed various designs with miles. You can reread here: http://endless-sphere.com/forums/viewtopic.php?f=28&t=45392&start=60#p663471

file.php



Here is a pic how it would look like if one places the motor in front of the BB, not above as I did:
file.php
 
ImageUploadedByTapatalk1354235253.843249.jpg
ImageUploadedByTapatalk1354235271.496426.jpg

I just picked up my burnt phase wires BMC.
Am I right in thinking if you put a one way in the right place the motor would be gear reduction if operated in forward and direct drive if reversed.
 
Just curious. I'm not going to go adding one way clutches. I'm happy enough with a mid drive motor that takes a few hours to build not trying to bite off more than I can chew.
Miles said:
mr.electric said:
Am I right in thinking if you put a one way in the right place the motor would be gear reduction if operated in forward and direct drive if reversed.
You'd need two one-way clutches.
 
crossbreak said:
please take off the rotor and stator and make photos :) Like to see what's going on already
I haven't disassembled yet but there are good photos of the internals here:
http://www.ebikessf.com/disassembled-bmc-motor
 
crossbreak,

The laminations in the MAC 500/1000 Rear Hub I just received measure 0.5mm C to C by measuring the distance from sharp edge of one lamination to the sharp edge of the next lamination. This will include any air gap between laminations. measurements were made with a 50X microscope with measuring reticle with .02mm divisions.
 
bØb said:
The laminations in the MAC 500/1000 Rear Hub I just received measure 0.5mm C to C by measuring the distance from sharp edge of one lamination to the sharp edge of the next lamination.

Thanks for this info! So the 44/13 crank to MAC shaft overdrive is a bit too much, keeping the eddy current losses in mind. Maybe you MAC people should go for a 42/14 using the stock 14T 3/32" chain cyclone freewheel. The 10turn (torque version of the BMC) wind would be a good choice though for people who want to go for 12s / 44.4V lipo. Plz PM me your battery voltage and wind so we can discuss the best overdrive ratios for your builds.

Does anyone who wants to do this conversion need a 20 to 12mm brass adapter from me to fit the freewheel on the shaft? I could make them on my lathe tomorrow and send them to you. I will post a HowTo for the output sprocket on sunday as well.

mr.electric said:
... there are good photos of the internals here: http://www.ebikessf.com/disassembled-bmc-motor
Thanks, that exactly what I was searching for.
 
HowTo convert a MAC hubmotor to make the axle spin
this is an alpha version, i do not own the MAC so enjoy this with care, dont just start this but think about it and tell us before you do something you might regret.

Tools needed:
-3-arm puller
-snap ring pliers
-bench vise
-drill set (at least 3mm, 4mm, 5mm)
-M5 screw tap and iron
-file
-sandpaper #200 for use with steel

-disassemble, like shown here: http://www.ebikessf.com/disassembled-bmc-motor, make sure you wear good leather gloves when removing the rotor. The strong magnets can easily hurt your skin if something goes wrong.

- unscrew this plate remove it, remove also the snap ring MAC_1.png

EDIT: You dont have to get off the rotor from the shaft if you use a eZee motor. It is hard to do that and may damage the rotor with eZee motors as StudEbiker sadly had to experience :( Instead pull out the shaft. I'm a little confused how exactly this procedure differs between MAC,BMC an Ezee motors. How this is done with these different motors will be posted here soon.

This picture show how forces have to be applied to get the axle out:
file.php


StudEbiker said:
You need to support the stator through the openings in the rotor somehow and then pound out the axle from the side opposite the sun gear. There's really not even a need to remove the gears for this conversion as they will come off with the rotor.

Or you can make a puller like Mr. Electric did.

http://endless-sphere.com/forums/viewtopic.php?f=28&t=45245&start=225#p678899

mrelectric made a special puller for this purpose from a steering wheel puller by reshaping it on a grinder:
file.php

file.php


- remove this key MAC_2.png

- use #200 sandpaper (should be purposed for use with steel) and grind the piece of the shaft where the stator is placed on (where the key was) a tiny bit down. You can stop this if the shaft can be turned in the stator carrier easily by hand. Lubricate this surface.

-dremel off at least 1mm off this advance on the stator carrier, maybe you must place two shim washers (high precision washer) where it was. Dont know since I have not seen a detailed pic of the bare axle yet. With my bafang this was necessary. We have to discuss this point, maybe you first try assembly without grinding this off and look if the axle may spin. MAC_3.png

-make the heat bridge. It is a ring with 6 holes which replaces this part by a much thicker one that connects the hub side plate and the stator View attachment 3 Dont know how thick this it has to be, the one I made for my bafang is 6mm thick. Use the original part as a former plate. Drill 6 holes into the hub side place, again using the old part as a former plate by putting it on the side plate and center it perfectly using the axle, use a drill dia that perfectly fits the holes of the old part (maybe 3mm?) . It is really important that this is centered right, otherwise the shaft will slip and do friction, only tiny bits of misalignment can be balanced during assembly, maybe 0.1mm, but not more.

- bore out the six existing holes in the stator carrier with a 4mm drill, cut M5 threads into it using a screw tap.MAC_4.png. Make sure you move it back and forth, and use lubricant to prevent the screw tap from breaking. Drill out the sideplate and the heat-bridge using a 5mm drill. Make sure all holes keep being centered perfectly.

- assemble, frequently try if the axle may turn during assembly. You should leave the planetary gears and clutch out for the first tests. The first stept in assembly is the axle, then screw on the heat-bridge and hub-side-plate using M5 screws.
 
The MAC is a little different when you take it apart. The cover that the wires pass under is the only thing that keys and anchors the stator to the axle.
file.php


When you remove the six screws the stator ad rotor come loose as one unit tied together by the sun gear, hollow shaft, large ball bearing and a large snap ring.
file.php


The little cover can be slid on axle and placed in the bearing inside the freewheel side cover. You can than use the little cover for a drill template to drill the motor freewheel side cover. A heat bridge/header can be made from a piece of aluminum round stock that is drilled for six screws to pass through the cover the header and into the stator. Also an over sized hole for the shaft. I am also considering adding a heat sink outside the freewheel side cover and adding oil to the motor. I will try to start a thread on converting a MAC soon. Right now I am taking pictures as I go. I think the MAC will actually be easier to convert than your Bafang. You have been the inspiration for me to start this project.

"Don't build an airplane because you want to fly. Build an airplane because you want to build an airplane".......Jack McWhorter
 

Attachments

  • PB300434.JPG
    PB300434.JPG
    83.6 KB · Views: 3,503
  • PC010438.JPG
    PC010438.JPG
    81.3 KB · Views: 3,503
I have found through trial and error that the heat bridge/header needs to be approx 0.1" /2.5mm longer than the cover that anchors the stator, but I have not made this piece yet. The freewheel side cover is flat and thin, so I don't think dimension is going to be critical. It is nice that the rotor and stator are tied together. The length of the header will only effect gear mesh depth of the sun and planet gears.
 
Hi Bob,
Nice to see the progress, great that you managed getting things apart!

1.Do I see it right? The rotors bearings are fit on the the axle? Or what is this bearing good for?
MAC_bearing.png


You can than use the little cover for a drill template to drill the motor freewheel side cover.

2. what would be the use of those wholes :? ?

I am also considering adding a heat sink outside the freewheel side cover and adding oil to the motor.

3. the heat-bridge can only be on the hall-sensor side. there is no way to guide heat through the gear side, i'm confused. Guiding heat completely around the motor though the hub shell will not have any effect, since heat resistance is much to big.

4. How tight is the fit off the stator carrier to the axle? can it be turned by hand if you remove the stock "little cover"?

The length of the header will only effect gear mesh depth of the sun and planet gears.

5. Guess with "header" you mean the advance of the stator carrier that axially restricts the carrier on the shaft. Are you sure the shaft wont rub against this "header"? That can not be true, otherwise the stator would normally float on the axle not being axially fixed ? Or is the stator carrier axially fixed on the axle by the "little cover" also?

6. What wind is your MAC? What battery voltage will you use?

7. Oil cooling will not work good, it adds fluid friction and your efficiency will drop. A tiny little bit of oil in the hub can increase lubrication, but it may not be more than a few milliliters. Letting the freewheel spin a few minutes floating in oil does reduce freewheel friction a lot with my bafangs btw, maybe it's good for your MAC freewheel, too ;)

8. hope you dont loose any parts in the lawn :D
 
crossbreak,

1. Correct. The O.D. of the bearing is a press fit into the Stator Carrier. The I.D. of the bearing is a slip fit over a hollow shaft that has the sun gear hobbed into it. The rotor is attached to the hollow shaft by unknown method ( I didn't pull the rotor all the way off the stator, only far enough to measure laminations ). The hollow shaft has more than 1mm clearance around the 16mm shaft. As a hub motor the stator was supported on one end by the little keyed cover and the rotor was supported by the sun/planet gear interface. The bearing ties everything together. If you hooked up the wires again to the controller and held the motor by the hollow shaft, you could run it :) All the motor clearances are maintained by the bearing in the stator. A very well thought out assembly :D

2. The holes (wholes?) are to run long 5mm bolts through the freewheel (right side) side cover, through the heat header/bridge and into the tapped holes in the stator carrier. This anchors the stator to the housing.

Got a call for supper :lol: I will answer the other points in about an hour.
 
Ok, i got he first tow points.

1. Looks like the construction is even better suited for a conversion as the small bafangs.

2. the holes on the freewheel side: You mean the sprocket freewheel not the planet carrier as I thought (it contains also a freewheel :D)

i'm waiting keen for your answers!
 
I'm back.

3. Right the heat bridge is only on the Hall sensor side. The stator is connected to the stator carrier, which is connected to the heat bridge, which is connected to the cover that the free wheel would have been screwed on. At this point you have helped some, but not much. The heat bridge added mass, which will delay overheating but it must conduct heat to somewhere else to be effective. The cover is not sufficient because it has neither mass, conductivity or surface area to radiate heat from. It is also insulated by power coating. So on the outside of the cover, mount a heat sink with fins to radiate heat to the air. The heat sink could even be part of the mount. Clean the powder coat off the cover and mount the heat sink to the cover.

4. The stator carrier does not touch the axle. It only touches the O.D. of the large motor bearing and the little cover in the stock motor.If you remove the cover that the bicycle free wheel would screw to (right side) Here is a picture of the motor out of the housing:
file.php

In the picture the motor would run if connected to the controller. It's being held up by the axle in a vise. If you cut the wire (you must to remove axle) and removed the cir clips and clutch from the other side. The motor is then only held by the six screws in the little cover. If you remove the screws, the motor will literally fall off the axle :oops: When I removed it I still had the stator hanging by the wires. This allowed me to get a gear puller under the little cover and pull it off the shaft. It comes off easy. I was trying to avoid disconnecting the wires, but it easier to cut them to start with. Then you could pull the little cover off by hand.

5. The header is what you call the heat bridge. The term header comes from transistors or other semiconductors that are heat sinked. The header is between the die and the case of the device. It determines the the maximum possible dissipation for the device connected to an infinite heat sink. Same thing, different languages. That is the reason I tried to use the term header/heat bridge to get the message across to more people. Wonder what the French would call it :?:

6. 12 turns. I am going to try 36 and 48 volts. Have not committed to a battery yet. When I do, it will be LiFePO4. Right now testing will be with Pb + H2SO4.

7. The oil cooling will be minimal. Only for light lube and cooling the magnets in the rotor. Probably 0W-5wt synthetic and just enought to cover the lower edge of the rotor. The magnets are subject to inductive heating under load and have a very poor heat path. Maybe we can make the magnets and adhesive last longer :wink: bØb
 
crossbreak said:
Ok, i got he first tow points.

1. Looks like the construction is even better suited for a conversion as the small bafangs.

2. the holes on the freewheel side: You mean the sprocket freewheel not the planet carrier as I thought (it contains also a freewheel :D)

i'm waiting keen for your answers!

1. I'm very pleased thus far.

2. Right On.

Oh yes. the oil can reach both sides of the motor. There are holes in the rotor cup and clearance between the case and the rotor.

It's hard to define the sides by right and left, because the motor will have to turned around in the frame. The rotation of the motor as a shaft motor will be in the opposite direction. Have to find common ground in terminology, my German is only as good as Bable Fish :mrgreen: bØb
 
StudEbiker said:
This thread is getting very exciting! :D I'm starting to run down a list of things I can sell to buy a MAC motor. :lol:

Got an extra SOUL you're not using? :twisted:
 
StudEbiker said:
This thread is getting very exciting! :D I'm starting to run down a list of things I can sell to buy a MAC motor. :lol:

buy a bafang BPM instead! It has less poles, so you can go higher with both speed and power.. and no one has tried it yet ;)
 
StudEbiker said:
Oops, I was typing while you were editing. :oops:
that's ok, i'm always editing :lol: now back to topic... the bafang has the same lamination thickness, so the speed limit is about double, since it only has half the poles of the MAC. Since they are almost equally sized you may get almost double the power. BUT you have to go for a fast wind AND have to use a high primary overdrive (from the cranks), which also has drawbacks. SO if you are willing to do this convert, you should wait, tell me your battery voltage and see how things go on till next sunday. This sunday will already get interesting cause I will first try how much benefit in heat rejection we will get. THEN we might know if a MAC can do 2hp and if a bafang may do 3hp continuous.

Or in better words: We can calculate the continuous power at least about 20% accurate for the bafang swxb.
 
Back
Top