I have done some investigations into temperature sensors and there is a sensor available with a linear dc output proportional to the temperature, driven from 5V so I assume it could use the 5V hall supply, but haven't tried it yet. I do intend to get some and start having a play with these. The new CA has an option to fit a pot onto the 5V supply that allows the speed or current limit to be adjusted on the fly. Can you see where this might be going....
Like I said I do now have a few of the double width stators available. I will try to arrange the parts to get a couple of prototype motors up and running. Maybe use a welded clutch as I think such a motor would likely be a bit too much for a standard clutch. Standard 500W stator is 21mm wide, these doubled up stators are 42mm. If you maintain the number of turns, a wider stator will be slower all things else being equal than a thinner stator, so I'm guessing this double width stator might be better used on higher voltage, or it will end up being a crazy small number of turns and therefore a hard load for the controller. Higher voltage will keep current levels more reasonable, so giving the phase wires an easier time as that is a bit of a bottleneck.
The Mac is a 16 pole motor as is the BMC and Ezee. Bafang and most other small geared hubs are 8 pole. there is a 5:1 gear reduction to figure in to. The high frequencies (80 cycles per wheel revolution) and the relatively small teeth in the stator means that the flux density in the stator can be quite high. This is why the Mac et all tend to have quite high no load current, it's also why they have good low down torque, you win some, you lose some. Stator losses cause heat in the stator and means that the motors are practically limited by stator losses to a maximum wheel rpm of about 550-600rpm. A higher grade stator material will have less stator losses, so should run cooler and be more efficient, at least at higher speeds. It can maybe also allow other aspects of the motor to be optimised as the stator is no longer the limiting factor, which could possibly further improve efficiency and therefore power handling. Anyways, as soon as these new stators arrive I will get some wound and see if theory follows through with hard figures produced on the dyno. I'm no motor expert, far from it, I have a lot I need to learn, but I have always liked to tinker and make the best of things. The Mac is already a nice little motor, but if it can be better even if it does add a little additional cost, it's all good IMO.
The EB3 I sell are all with sensors, but sensorless is available in the EB3. The Mac is a 16 pole motor, and I don't think sensorless controllers work very well with them, although there is no reason a very good sensorless couldn't work perfectly. Andy is I think the only person that I can think of that has burned the phase wires like this and it was only this that resulted in the damage to both halls and the controller. I've abused a Mac pretty bad to where the windings were very discoloured and smelly and the windings finally gave up, but the hall sensors still worked ok. Mac use the Honeywell ss41 sensors, they can take some abuse. I have had very few hall sensor issues with the Mac and most of those were not due to heat damage, almost always another cause.
Temperature sensors are cheap anyway. 
have to get my hands on one of these ...
