Hi folks,

 

I'm having difficulty diagnosing a problem with my MAC motor kit so I'm hoping someone here will have the knowledge and experience to tell me exactly what the problem is.

 

The problem started yesterday when I came to use the bike for my commute home from work. It had been absolutely fine on the ride in.

 

When I turn the throttle I get a short burst of power and then the motor stops. If I release the throttle and turn it again I get the same a short 1 or 2 second burst then nothing.

 

I've connected to a different battery and a different controller so I'm assuming the problem must be with the actual motor or hall wires. Does anyone have any ideas?

 

Thanks,

 

Justin

Probably a hall sensor problem. Did it get water in it? You can test it with a Ebike tester:

 

http://pages.ebay.com/link/?nav=item.view&id=181246430524&alt=web

Hi d8veh,

 

I don't think any water would have got into the actual motor. I replaced the cover on the motor recently as the bearing had gone and it looked pretty clean and dry in there despite virtually everyday use over the last 2 years rain or shine.

 

Also it was dry yesterday until the eve. When I had the motor out a few weeks ago I did put some more grease around the gears. Could this have got onto the hall sensors to cause the problem?

 

Some of the wires are bear where they thread into the shaft although they were like that when I first bought the kit and don't seem to effect its operation. I did cover them with a bit of insulation tape recently and I've just taken that off as I found moisture underneath it. But I've just dried the area with a hair dryer and that's not cured the problem.

 

I have got one of those testers and I've just tried using it. I'm not sure how the steer crocodile clips are connected or what that means even!! So I left those disconnected.

 

I'm not sure how to interpret what the lights mean. The motor doesn't run at all when the tester is connected. The green power light is lit and the red steer light is lit. When I turn the wheel backwards the hall signal a(yellow) and c(blue) alternate although are both on in between. The b(green) does not light up at all so I assume that is the problem. The coil lights don't come on at all and the circle of red lights come on in 2's in rotation when I turn the throttle. The central 5v light flashes though this seemed to go off when I jiggle the connector for the handlebar on/off switch, so possibly a loose connection there.

 

Any help is greatly appreciated!!

You connect only the 5 hall wires to the connector on the tester, then switch on the tester. When you rotate the motor backwards, you should see the three LEDs on the tester flashing in sequence. If one doesn't flash, that's your problem.

 

Then, you have to find out why. Barecwires running through the axle is definitely not good. Surely, you don't mean that!

Thanks d8veh. I guess it's the green sensor wire or the actual sensor then.

 

There is a tiny nick to the green wire where it enters the axle. The worst is the green phase wire though which is very bare. I've tried separating them out a bit to see what's going on. I cant see any bare wire actually touching the axle though. I'll upload a pic so you can see what I mean.

 

Could a short have killed the sensor? It seems odd that the motor runs for a few seconds each time.

Oh dear! You'll have to repair that or you'll be looking for a controller too. It's a problem to get the eight or nine core motor wire though unless you have a scrap motor to cut it off. If you have enough length, you can pull it through to lose just the bit between the damage and where they're presently soldered.

Thanks d8veh. I haven't really got any spare length but maybe I could move the controller forward a bit.

 

Would a short destroy the controller? I have a spare controller that I replaced a mosfet in last year. I've already tried connecting that up and get the same result.

 

Whilst you are clearly saying the chaffed wires need sorting out as I can't see any wires actually touching each other or shorting on the shaft (it's easier to see that looking at it than in the photo!) I'm assuming the problem will still be with the hall sensor.

Are there any continuity testing I should do to check for a short?

I would change the motor-cable if you have some welding eperience you can handle it. I always use extension cord, cut the side with no need away and fix the cable to the motor. It´s no so difficult but need time and patience

 

I didnt wrote a manual for the mac-motor how to do, I only wrote one for the CST-bafang. Main steps are the same open motor disamble motor old cable out new cable in. To put the cable trough the axle I use soapwater

 

I wrote a manual how to measure controller and Motor incl. Hallsensor but I still had no time to translate it in english but I guess by google translate it is easy to understand

have a look here: http://fatbike-motor.com/download.htm

 

Then you would have erased one obvious failure

 

regards

frank

Edited May 3, 201510 yr by mechaniker

Thanks Frank,

 

I did also wonder whether it wouldn't be better to use 2.5mm or 3mm flexible cable than the thick strand cable currently used.

hi justin,

 

it would be better but brings the price 2usd higher and the china factory only focus on the price never on quality I had this discussion for many years and tried to explain they should make the price higher for a better quality......

Oh dear! You'll have to repair that or you'll be looking for a controller too. It's a problem to get the eight or nine core motor wire though unless you have a scrap motor to cut it off. If you have enough length, you can pull it through to lose just the bit between the damage and where they're presently soldered.

 

Hi d8veh,

Thanks for your help so far.

I've been trying to pull the wires through as you suggested. I've squirted a bit of wd40 in where the wires enter the shaft and am using some thin nose pliers at the stator end to pull them through. But all I seem to be doing is stripping the wires.

It feels like the angle that the wires come out of the shaft is just too acute to pull the wires through in this direction. Have you succeeded in doing this before and do you have any tips?

 

Thanks,

 

Justin

You have to push from one end and pull from the other. Once you're round the corner, it's easier. Good luck!

Hi wonder if anyone can help with this.

 

Following the above I ended up making a new cable. I also replaced all 3 hall sensors and the pcb they're attached to (thanks Fordulike for sending those with the other MAC parts).

 

The lights on the tester now all come on and off in turn when the wheel is turned. There is an overlap when the next light comes on before the previous one goes out. I assume that's normal.

 

But when I run the motor I get the same (or at least very similar) problem. The motor cuts out after a short period. Although this is longer than it was when the problem first started probably more like 4/5 secs now instead of 2/3. Again dropping the throttle and turning again puts power to the motor immediately.

 

I know the problem is not the controller or battery connection as I've been using my girlfriends wheel and motor on my set up while trying to fix mine.

 

I feel it's got to be hall sensor position but I've glued the new hall sensors in exactly the same places as the old ones and there really aren't any other options.

 

So has anybody got any suggestions or ideas of what could be causing this?

 

Btw if anyone reads this who plans to make up a new phase/hall sensor cable, I used some 6 core burglar alarm cable which had all the right colours for the hall sensors and used 3mm thin wall automotive cable for the phase wires and some heat shrink cable sleeve to pull them all together.

 

Thanks again,

 

Justin

I'm not sure, but I think the middle one might have to be the other way up. Is there anywhere that mentions the orientation?

Hi again d8veh,

 

I saw posts about orientation on endless sphere but it wasn't the mac motor so I don't know. Looking down at the pcb all the positive feeds (red wire) go to the left terminal negative (black wire) in the middle and output (coloured wires) on the right, so I put them all in with the beveled edge facing out.

Hi again d8veh,

 

I saw posts about orientation on endless sphere but it wasn't the mac motor so I don't know. Looking down at the pcb all the positive feeds (red wire) go to the left terminal negative (black wire) in the middle and output (coloured wires) on the right, so I put them all in with the beveled edge facing out.

IIRC, the hall sensors can only be slotted into place one way round, due to the bevelled edge, so I don't think wrong orientation will be a problem.

Before going to bed last night I tried running the motor again and it didn't cut out. I hate intermittent problems!!!!

 

I will road test today and am hoping that I'd just not made a good enough connection with the controller when I first plugged it in which improved after retesting the sensors.

 

Come to think of it the japanese connector does feel loose when I plug it in.

 

So now I'm worrying, could a loose connection from controller to hall sensor cause the hall sensor to fail and be the original cause rather than a short from the frayed cable?

Oh dear, bad bad bad.

 

I test rode it and found that at about half throttle I got a juddering from the motor which then disappeared at full throttle/speed. The motor seemed to run at the same power level as before but sounds different somehow (i think lower pitched and slightly louder).

 

I assume this must be a hall sensor issue although on the tester all sensors are lighting up in turn.

 

I then went up a hill and I assume the part they call the clutch started slipping. It felt just like a car clutch slipping.

 

I had to take the shaft out to feed in the phase wires and wonder if I've damaged this part somehow, it did require a lot of levering to free it. I also put lots more grease on the gears so maybe they're slipping somehow(?).

 

I have a week off to do jobs and must get on with other things but ideally need to get this sorted by Sun eve.

 

Any ideas what I should do next?

What feels like slipping clutch could be just the controller losing sync unless you heard the motor whining at full speed with no drive.

What feels like slipping clutch could be just the controller losing sync unless you heard the motor whining at full speed with no drive.

 

It didn't lose all drive though it did lose power.

 

I've been searching the forum about losing sync, I don't really understand what you mean by it but I see from your posts on other threads that it can be hall sensor related which is the most likely in my case.

 

I think one of the hall sensors got positioned slightly lower than the other two. Could this be causing the problem?

When the controller fires it's pulses not in time with the motor rotation, you get juddering that feels like a serious mechanical problem. It's often caused by a faulty hall signal but can happen when you have a controller that's not suitable for your motor. If your controller was evr running OK with your motor, that's not very likely, so have a good look at your hall sensors and their connections.

Thanks d8veh, I think all is ok now. I'll put this post up in case it helps anyone else.

 

I checked the hall sensor connector and although it feels like a loose fit (or at least slides in very easily) I'm satisfied that it's making good contact. When connected to the tester all the hall indicators light up in turn and when connected to the throttle the wheel turns.

 

I was ready to replace the hall sensors thinking the sync problem must be to do with their position. It was a bit of a bodge when I put them in. I thought the best method would be to put the epoxy in all three positions and thread the hall sensors through the holes on the pcb so there's not a problem lining them up once glued in. The epoxy obscured the gaps where the sensors were going into and trying to line up all 3 got a bit messy and I think one of them got pushed down further than the other 2. My advice to any fellow newbie doing this for the first time would be to glue each sensor in separately before putting the pcb on.

 

So I took the motor out again. There was something else that had been bothering me in that there was a gap on the shaft between the circlip and the clutch/gears after I'd put them back on (I took the shaft out when replacing the phase wires). I had another look at this and decided that the gears were rubbing on the back of the magnet bowl that contains the stator. I decided this was the cause of the change in motor sound and needed fixing so levered it back up the shaft. This requires quite a bit of force on my motor, compared to the description by spinningmagnets on his MAC tear down thread on Endless-sphere. So having shifted it up a bit I decided to put it all back together and test it. I was really reluctant to start chipping out those hall sensors again!

 

Anyway I took the bike out for a test ride and got the same judder at half throttle. I took it up the hill where I got the clutch slipping feeling on the 2 previous road tests and that had gone. The motor also had the same higher pitched sound as it used to.

 

I suspect the difference in relative position of the hall sensors is the cause of the juddering (sync issue) at half throttle. But since this only lasts for a couple of seconds as accelerate through this speed I feel I can live with it. Please let me know if this is likely to damage the motor or controller.

 

Thanks again d8veh for all your guidance.

 

I'll just add one other thing for anyone reading through this. The wires passing through the shaft on the MAC motor are VERY tight. I tried pulling them through but it wasn't gonna happen so ended up cutting and replacing. The protective sheaf around the old wires didn't even budge once I'd pulled the wires out from the controller end so there is no way it would have pulled through round the 90 degree bend at the stator end. Spinningmagnets makes this point in his tear down thread http://endless-sphere.com/forums/viewtopic.php?f=16&t=51310 - trying to pull them through just tears the insulation off. In my case levering with the pliers also scratched the stator windings, fortunately not deep enough to cause a short and I've painted some electrical varnish on the scratches to be sure. It at least gave me a wake up call to be careful on how I handle the stator!

Edited May 30, 201510 yr by Justin.Clements

Justin, I admire your perseverance in replacing those hall sensors and getting it running again.

 

When a hall needed replacing on my old MAC motor, I attempted to dig out the circuit board, but gave up and ended up buying a new motor from em3ev.

 

But it was not all in vain. Paul sent me the spare circuit board and halls and we all know where they are now

 

BTW, if you ever decide to replace the motor for whatever reason, then just buy the motor unit from em3ev and not the motor built into a wheel. Fitting the motor to the old wheel and casing is so easy.

The delivery works out much cheaper for just the motor, due to shipping being based on size/weight.

Edited May 30, 201510 yr by Fordulike

Thanks Dave,

 

I must admit if I were to convert the cost of a new motor into an hourly rate for the time I've spent on this it would be even less than I earn at work at the moment lol. So I'll bear that in mind if there's a next time. That said it's very satisfying to fix things.

 

I find it eerie that just about all the parts you sent me, the replacement side cover, brakes, pcb and hall sensors even the gold bullet connectors and shrink sleeves have now been used. That is apart from the spare gear wheels/clutch. I hope I'm not tempting fate there.