Strange: Sinewave controller won't start my MXUS XF07.

[docw009]

I believe it's MXUS, but I bought it off ebay. I only used it once on a square wave controller, and it turned to be a 210 rpm model, not that good with my 20" wheels. So I put it aside.

Today, I had a flat tire, and this wheel was sitting there with a tire, ready to go, so I tried to put into service and got an error 03 (Hall sensors) on my KT controller. Connected it to a second controller and same problem. Then I remembered that I had changed out the first bike to a sine wave model.

So I found a square wave box and yes, the motor will run on any of my old square wave units.

Anyone ever hear of this?

 

[spinningmagnets]

Does the newer sine-wave controller have a program for entering the number of magnet-pairs, and stator-coils?

There is also a data point for the arrangement of hall sensors in degrees.

The combinations of blue/green/yellow motor phase wires is not always the same for every controller, as are the B/G/Y wires for the hall sensors. There might be nine possible wire arrangements to get it to run.

 

[docw009]

It's not the wire colors. Maybe I'm lucky, but I've never seen a mismatch with over 10 bike builds, plus I can plug in the 9 pin connectors from a Q100 and this MXUS and it runs with two different square wave controllers.

Maybe I'll try my KT settings. Could have set one of the C values off.

 

[MadRhino]

It's not the wire colors. Maybe I'm lucky, but I've never seen a mismatch with over 10 bike builds...

Funny. That is more than lucky IMO. I have seen very few controller/motor wiring color matches in the last 12 years and a good 3 dozen motors with 5 different controllers.

 

[amberwolf]

My first bet would be the hall-phase wiring combination.

I've often seen different wire combos between different controllers and motors. (I even have at least a couple of 9C 280x motors that both have different wiring orders!)

If its' sensorless, it doesn't matter.

But if it's sensored...the wire order depends on the order that that specific controller builder on that particular workshift happened to use for the hall wires, *and* the phase wires, each of which could be different, *and* the order that that specific motor builder on that particular workshift happened to use for the hall wires, and the phase wires, each of which could be different, all ending up matching.

I dont' know the math to figure out the statistical chances of that...but you have 3 wires for hall, and 3 for phase, with 3 possible orders for each, and that is in each of two devices that have to connect to gehter. I think that's 9 possible orders, on each end, which is 9 x 9, so a 1 in 81 chance of connecting together correctly at random? It's probably really a much higher chance as there are probably more manufacturers that intend for their stuff to be wired up in a particular order that happens to match other manufacturers intent, so there are probably some wiring color pattersn that happen a lot more often than others. And my math is probably wrong anwyay.


It's unlikely to be the old "60 vs 120 degree" problem, as almost every motor I've run across (even here on ES) is 120 degree, and so far all but one ancient controller (actually analog control chip, vs MCU!) that I can recall I have, has defaulted to 120 degree. (some have a jumper, some don't appear to have a way to work with 60 degree motors unless it is automatic detection).


It could also be as Spinningmagnets points out--the controller may need to know the number of pole pairs (# magnets / 2), if it is an FOC controller it might need more info than that. If it's just a lookup-table type sinewave controller (the common cheap kind) it probably doesn't need any of that info, and will just run, as long as the phase/hall combo is right.


There are also some basic sinewave controllers, like the later Grinfineons, that if you unplug the halls will work in trapezoidal sensorless mode. Noiser, but easy to make it work.