MAC 350w (front) and Infineon 20A compatibility

[toaroa]

Looks like I'm paying the price for purchasing motor (cell_man) and controller (cycle9) from different vendors (I so knew at the time that this would come back to haunt me!)

I've tried matching the colours for phase and hall wires and switcing green and yellow for both phase and halls. Both ways, I get the same result: very slow starting on no load, with ability to (smoothly) rev up to WOT to achieve only 12-15kph (whatever that is in RPM, it's low, as measured via cycle computer) at 6-7A peak (measured by turnigy watt meter before the controller; can't be much more after). When I try this with load from a pedalling start the assist is so negligible that it's barely there, but I can hear the motor.

I wonder if I may need to a) get a TTL-USB cable and reprogram the controller b) purchase another controller (v expensive from NZ) c) consider buying a cycle analyst to see if I can alter the current limiting that way or d) check the halls somehow.

???? ANY IDEAS WOULD BE WELCOME

First ebike was sensorless and I'm beginning to regret my search for greater power!

 

[johnrobholmes]

You may open it up and inspect the shunt(s). It may have a cold solder joint.

 

[toaroa]

Thanks for your reply. You mean in the hub motor itself?

 

[Burtie]

Is that 6 to 7amps with no load on the wheel ?

If so you have not got the right senor + phase wire combination. Trial and error will find it (go easy on the throttle when testing).
No load current should be around 2 amps when you have it right.

 

[toaroa]

I checked the no-load amps again (three repeats) using all colours matching. Peak amps were 2.41, 2.35, 2.5. I can't recall whether the earlier 6-7 readings were with the greens and yellows transposed but performance at the wheel seems to be the same either way. Perhaps I'll have to systematically test all 36 combinations to see if it's a wiring issue before considering something like re-programming?

 

[wanders]

Typical for MAC (Shanghai) hub motor and so-called Infineon controller, swap yellow and green on the phase wires and on the sensor wires, leave blue connected to blue, and you should have success.

I know, because I have the same setup.

wanders....

 

[toaroa]

Cheers Wanders. That's what I started with (as I saw that combo posted on ES a few times), but I got the exact same result. I'm beginning to wonder whether my hall connections aren't the best, but it could be so many different things it's hard to know where to start.

 

[dogman dan]

It's possible you have a bogus hall sensor. Hook 5-12v to the red wire pos, and black wire neg. Then a voltmeter red wire to the red wire , and the black to either red blue or yellow. Spin the wheel, and see if you see voltage, then no voltage. If one is always hot, or always cold, then you have a bad hall, or bad wire to a hall, or something like that. It doesn't take much to cause a problem, and its entirely possible the controller doesn't work right.

Another possible issue is that some motors are 60 degree spacing, and others 120 degree. But that's unlikely if cycle 9 sent you a controller intended for a gearmotor.

 

[johnrobholmes]

The shunt is in the controller. The other possibility is that your throttle may not be working properly. Do you have another throttle to test with?

 

[HighTekBikes]

If you search this forum, you will find detailed info on how to match up the motor wires. There is also info on testing the halls. The outputs are pulled up by resistors in the controller, so if they are not connected to the controller, you have to use a resistor to pull them up to 5V. It's easier to just conenct them to the controller and leave the power wires (to the motor) disconnected. Spin the motor backwards slowly and monitor the outputs.

The no load current should only be about 1 amp.

The speed is another issue. I don't know how you are measuring that but check the throttle output, it should be 4V at full throttle.

 

[toaroa]

Thanks heaps for the tips on testing - I didn't know how to test this without specialised equipment which will take weeks to have sent here (New Zealand) = great to know that I can test the halls with a simple voltmeter. I could check the throttle, but I'd be surprised if it was that because I bought the throttle from the same vendor as the controller and paid extra for compatible wiring.

I will certainly redo my checking of the 36 wiring combinations but this time I'll strip the wires bare and use alligator test clips for all the hall connections. I'm pretty sure of my anderson connector crimping abilities, but the JST connectors I've been using for the halls could be suspect. Finding a crimping tool that small has been tricky. That said, I've used all three cable combinations for mac+infineon that have been reported on ES, so I'm not expecting the problem to be mis-matched cables. I'm fully expecting to need to replace one or other of motor and controller, but I'll check all I can first.

I'm measuring speed via my cycle computer. Originally, I thought I had the thing working, but when I then put it on the bike I realised that the speed was way down on what it should be - the cycle computer is one of those with a magnet in the spokes. 15kmh is about 120 rpm (by my rough calculations and I reckon that's 1/4 to 1/3 of what it should be.

 

[John in CR]

When you say you can hear the motor, does it sound like it's working and spinning fast? I wonder if maybe you have the wheel on backwards, so the motor is just freewheeling and the friction in the freewheel is the only thing getting the wheel spinning when off the ground and provide no meaningful assist with you on the bike. Your no load current seems to indicate the wiring is correct for that direction. If you lift the wheel off the ground and spin it with the motor off, does it spin freely in the forward direction and you feel the motor drag turning it backward? That's what it should do when mounted in the correct direction.

If the wheel spins freely in both directions, then you have a freewheel issue inside the motor. If there's motor resistance only rolling backward then the problem is elsewhere. As posted earlier go easy on the throttle when trying wiring combos, because incorrect combos result in high currents. With each new combo only swap 2 wires at a time, either 2 phases or 2 halls, whichever is easier.

John

 

[toaroa]

It'd be a bit sad if, on my third ebike build, I was sufficiently dumb to put the wheel on backwards! (especially as it has a disk brake/rotor to tell me which way to go)

BUT THANK YOU ALL FOR YOUR HELP!!! The kindness and thoughtfulness of folk on ES never ceases to amase.

Bugger me - after all that fretting about the motor-controller hookup it was, indeed, the throttle. One wire had pulled slightly loose inside the base. Once that was put back in its place, things improved a little but not a lot, so I re-tried other hall and phase combos, including one which had partially worked the last time around. I'm not completely convinced of this one (as it only gave me 30km/hr no load and 26km/hr load on flat, and I had been hoping for 32 on the flat) and I may try others, but this seems very stable:

Controller: Motor

P H A S E
yellow: yellow
green: blue
blue: green

H A L L S
blue: blue
yellow: green
green: yellow
red: red
black: black

Weird combo, but if it works, it works I guess. I rode for about 3km, including a small hill, with peak amps of 12. As the controller can supposedly do 20, I may be able to eek out a little more speed with a few adjustments and programming.

THANKS AGAIN!