Shenzhen (ecrazyman) Controller Information

[oddjones]

Bafang/crazyman - the continuing saga....

Your throttle seems to be connected correctly, if the color matches... and the wheel spin. At this point, I would try to check if the motor is defective... First try to measure resistance between every combination of the phase wires: it should be the same. Then, power up the rig, and disconnect the three signal wires from the hall connector. leave the red and black wires connected. Measure the voltage between red and black, ther shoul be at last 5V. then put the black wire of the tester on the black wire of the hall. with the red wire from the tester, measure, one at a time, the three hall wires (thin Yellow, green, blue). They should alternate from 5V to 0V when you slowly rotate the wheel.
Does anyone have some other suggestion?

OK I've been talking with Knuckles on the email and it all got a bit confusing so I dedcided to run all the tests again - There's a new spreadsheet of (intriguingly different!) results available here http://www.oddjones.com/docs/PHASE-HALL-ODDJONES-2.xls This time I made sure the motor was in the right way round and the right way up, and I wired a 5a fuse between the battery and the controller (well, ten of them so far - they kept blowing!)

I'm conscious that it's really difficult to get across exactly the problems in text sometimes so I've posted the following videos to hopefully get to the bottom of the situation...

1) - This is what I mean by "nudging the wheel backwards"

[youtube]N17gcS_M_GA[/youtube]

I also took the time to run gip_mad's suggested tests on the motor again - with different results...

2) - Phase wire resistance - last time I realised I was measuring the resistance across the HALL WIRES not the Phase wires...
[youtube]NzbYk40luKI[/youtube]

3) - Hall Sensor test -
[youtube]aYllXwQI0p4[/youtube]

So - what do people think? Have I fried my motor? -

 

[gip_mad]

You have a weird hall output.... Anyway, the green phase wire is interrupted. This is bad. It could have happened that you "cut" it internally by bending it too much, or simetimes it gets ripped off from the hub (internally) if you power the motor without fixing it to the bike. Or maybe it is a factory defect... If you didn't exercise any force on the phase cables, you should return the motor under warranty.
I'm sorry that you are having such a bad experience with your ebike, I hope you get up and running soon.

 

[oddjones]

gip_mad you're right - this really isn't doing much for my spirits. - the motor has been treated with kid gloves - I spent 15 hours lacing the thing into a rim then it was mounted directly and it's only been run in the bike - I can't see how I could possibly have broken a wire.

you should return the motor under warranty

Hmm, Yeah - slight problem with that - I bought the motor from Bafang direct - the motor was worth £25 - postage and tax was £50 - sending this one back and having a replacement sent out would cost me the equivalent of 4 new motors... can anybody think of a way around that?

Is anybody on here (preferably in the UK) dealing with Bafang so I could lump mine in on a shipment that's already going back to them?

Or maybe supply me with a new one and earn my undying loyalty and respect as a happy customer?

I'm not a quitter and I will buy another motor if I have to but this is no longer seeming the economical alternative to my Volvo that it did at first!!!

 

[docnjoj]

Follow Links replacement guide to wire replacement. The Bafang opens only 1 way with just 6 screws! Or perhaps Knuckles could walk U thru it! Its not rocket science to repair Halls and wires. If the phase wires are made to work, consider dropping the Halls for the converter from Knuckles/Keywin. Then U could avoid replacing the Halls. The motor runs great when set up and properly connected or repaired.
otherDoc

 

[oddjones]

Follow Links replacement guide to wire replacement. The Bafang opens only 1 way with just 6 screws! Or perhaps Knuckles could walk U thru it! Its not rocket science to repair Halls and wires. If the phase wires are made to work, consider dropping the Halls for the converter from Knuckles/Keywin. Then U could avoid replacing the Halls. The motor runs great when set up and properly connected or repaired.
otherDoc

Now that's my kind of solution... But before I start I really need to be convinced that the motor isn't toast... Given that it does spin (as you can see in my first video) can I assume that if I sort out the wires and Halls that it will work? Like I said before I'm fairly certain I haven't done anything that would break a wire - but I did ride the thing for half a mile with the wires set up all wrong - is it possible that the interrupted green phase signal could be down to something more serious? (Like a blown winding - excuse me if that's a stupid question, but I am new to this and don't want to go to the bother of unlacing the rim, opening the motor, rewiring it, replacing the Hall sensors only to find it still doesn't work - at that point I would really be close to quitting the project!!!)

Knuckles/Keywin - could you point me in the direction of a thread detailing your convertor and how I make it work?

 

[gip_mad]

Well, if that is your kind of solution, you can open up the motor! you can check for continuity between the two ends of the green phase wire. If the tester gives you 0 ohm, then the problem is in the winding and you can't do much to solve it, other than rewinding the motor.... A thing that I wouldn't recommend. Since the motor spins, I would assume that once you repair the phase it is going to work, but I can't be completely sure. The fact that hall sensors works only when you turn the wheel backwards, probably means that the motor has a freewheel. So, when you turn the wheel the correct way, the motor is "mechanically disconnected" and the sensor reading doesn't change. As a side note, this means that you can't do regen, I suppose.

So, now if you're brave enough you can open the motor and see if the problem is in the winding or in the cable (and hope that it is the cable) or get a warranty return. Be careful if you open the motor, there should be gears and things that can be misplaced/lost, be sure to remember where everithing goes.
You souldn't need to go sensorless, as the sensors seems to work correctly.

 

[docnjoj]

U dont have to unlace the rim! Just take the wheel off the bike so U can easily work on the it! Then remove the 6 screws and Shazam, U R inside. Be careful with the wires, as U can imagine, they are fragile Knuckles has some pictures of an open motor somewhere in the Bafang thread. Gip_mads post seems to say it all, but a picture is worth 1000 words. I think all the parts are clipped into place so U cant lose any gears. But dont work over a holey floor!
otherDoc

 

[oddjones]

The fact that hall sensors works only when you turn the wheel backwards, probably means that the motor has a freewheel. So, when you turn the wheel the correct way, the motor is "mechanically disconnected" and the sensor reading doesn't change. As a side note, this means that you can't do regen, I suppose.

The motor definitely has a freewheel - I've read about this, what I don't understand is what purpose the hall sensors have if they don't give a reading when the wheel is turning in the right direction?

If I'm reading this right, the way it's working currently is that the motor won't turn until it has a reading passed to the controller from a hall sensor - but the hall sensors only function in the reverse direction, which is why I need to turn the wheel a fraction in reverse in order to make it work.

Is the problem that the motor is actually meant to turn regardless of hall sensor input? If so, then what are the hall sensors for?

>>nooby nooby noo!<<

 

[docnjoj]

U gotta remember that the motor turns 5x as fast as the wheel and there is probably a little freedom in rotation befor the gears catch. This should be enuff to enable the Halls! They will of course allow rotation in either direction, but obviously the wheel wont turn with the motor in reverse!
otherDoc

 

[Knuckles]

Time out everybody!

That is a Bafang Motor? Direct from from Bafang? Dom ... How the heck did you manage that?

No worries. WE WILL SOLVE THIS MYSTERY!

Why? Because we so F'n CRAZY!

:x

 

[John in CR]

OddJones,

Did you try riding the bike with the "reverse nudge" to get going? If not, give it a try, and if it does run fine other than the start, then it's not the motor and you just fried something in the controller with your bad wiring combo torture test for hundreds of yards including a hill attempt. Until you do that, I don't think you can rule out the controller or damage to the motor with that first run, since it was dumping loads of current through the motor essentially fighting against itself due to an incorrect "firing" order.

If it doesn't run well, then take your choice on which one to open up first to look for trouble.

John

 

[fechter]

When testing the hall sensors you need to turn the wheel backward so the rotor moves.

 

[gip_mad]

John, if you watch the videos you can clearly see that he has an interrupted phase wire.

The hall sensors purpose is just that: provide synch to start the motor even if it is stopped. If you measure the hall voltages with the motor stopped, you will see that they are not all the same: that is to tell the controller wich phase to energize first.

 

[John in CR]

John, if you watch the videos you can clearly see that he has an interrupted phase wire.

The hall sensors purpose is just that: provide synch to start the motor even if it is stopped. If you measure the hall voltages with the motor stopped, you will see that they are not all the same: that is to tell the controller wich phase to energize first.

I only watched the first one before. Will a motor spin up that well on just 2 phases? I guess his "ugly sound" that ended the initial bad test ride was the pop of a winding. The problem now is that the test ride may have destroyed any evidence if the motor was faulty to begin with. Time to open it up and see what things look like. I hope it was just a bad solder connection on a phase wire that finally popped loose and is easily repairable, and that the controller survived. Then he could be on the road in a jiffy....optimistic but possible.

John

 

[gip_mad]

The motor would spin without a phase, but it would be noisy and very inefficient, and almost unable to move under load. You shouldn't burn mosfets if you have an interrupted phase.

 

[oddjones]

bad wiring combo torture test

Heh heh - that's me, gonna have it put on a t-shirt

 

[oddjones]

The motor would spin without a phase, but it would be noisy and very inefficient, and almost unable to move under load. You shouldn't burn mosfets if you have an interrupted phase.

It doesn't sound noisy to me?

 

[gip_mad]

The motor would spin without a phase, but it would be noisy and very inefficient, and almost unable to move under load. You shouldn't burn mosfets if you have an interrupted phase.

It doesn't sound noisy to me?

On the spinning video... It should be almost silent. Or is it the auto gain adjustment on the microphone that amplify it too much?

 

[John in CR]

The motor would spin without a phase, but it would be noisy and very inefficient, and almost unable to move under load. You shouldn't burn mosfets if you have an interrupted phase.

It doesn't sound noisy to me?

It definitely sounds rough at lower speeds as it's accelerating. Like gip_mad said, that resistance of 1 between 2 of the phase wires tells you something is wrong with it. Another way to tell is with nothing connected, spin the wheel against the motor resistance. Then short 2 of the phase wires together and try again. It should be significantly harder to turn due to the electronic braking. Then short the other phase too. If doesn't get more difficult to turn, you've got a problem.

John

 

[fechter]

Another way to tell is with nothing connected, spin the wheel against the motor resistance. Then short 2 of the phase wires together and try again. It should be significantly harder to turn due to the electronic braking. Then short the other phase too. If doesn't get more difficult to turn, you've got a problem.

John

That is the best test. Measuring resistance (ohms) will only tell you if a wire broke off and is disconnected unless you have a really fancy meter.

If a winding or wire shorts, the turning resistance will always be present and not increase much when you short a certain combination of phase wires during testing.

 

[gip_mad]

Measuring resistance (ohms) will only tell you if a wire broke off and is disconnected

Exactly, it is disconnected....

 

[oddjones]

Another way to tell is with nothing connected, spin the wheel against the motor resistance. Then short 2 of the phase wires together and try again. It should be significantly harder to turn due to the electronic braking. Then short the other phase too. If doesn't get more difficult to turn, you've got a problem.

John

oh dear.

Yellow - Blue = +resistance
Blue - Green = nothing
Green - Yellow = nothing

I guess I just have to pull the motor to pieces and hope that green wire's hanging off?

Houston?

 

[docnjoj]

Knuckles will save U!!!!!!! But yeah! U will have to open the motor!
otherDoc

 

[oddjones]

Knuckles will save U!!!!!!! But yeah! U will have to open the motor!
otherDoc

Surely it can't be this simple?

Gonna solder it up now... so excited!

 

[docnjoj]

Cant see from the photo but sometimes U bite the bear! Did u find the offending phase wires?
otherDoc