Interesting... I think most of the problems can be traced back to heat as the root cause. Heating the windings and sensors, heating the gears (and excessive torque), heating the epoxy holding the magnets on the rotor enough for them to separate at high rpm. Motor simulator doesn't help the cause with maximum temperature 150C. The original word from Grin was 700W, now reduced to 500W.
BTW, the trip simulator with manual slope input is a more useful tool than the motor simulator for estimating temperature behaviour.
They're a lightweight low power motor to fit EU EN 15194. That's 250W (about 300eW) & 6-25kph pedelec. At that speed limit the motor output is 160-200W. Used within or closer to design it **should** have much fewer problems.
My 311 is kept below 80C, usually running at 60-65C which means pedalling the damned thing (typically average 160-200W, 250-300W uphill, measured with Powrtap power meter - more than average FF), and below 30Nm and 320rpm. Hard limit is 350rpm in a 20". Temperature rollback at 80C, hard limit at 100C although I can reduce that to 90C without noticing a difference, I set the adjustable speed cut off just below the speed I maintain on human power alone which is typical for an EU pedelec. It has about 7000km. Already more than some will be ridden in their entire lifetime, but it is only a few months old and I may yet eat my words. I bought a spare motor just in case.
Controller is a stand alone baserunner. The motor is not silent on the bench. I can't hear it on the road at normal power, but it is about 2m behind me. I could hear it accelerating at 400eW during a few hundred kilometres of testing with logging. With that power it heats up fairly quickly.
I don't run it that high now. The reason I switched from 9C DD to the 311 was that the more efficient and temperature resistant DD drew too much queer eye attention from police and I had to defend it too often. Not the US, eBikes and cyclists in general are soft targets for police here. Give them an excuse and they'll seize it out of spite. Safety is their excuse, but they have an appalling, nearly non existent close pass enforcement record. Go figure... soft targets.
500W should be okay with effective thermal management. The motor needs to be modified for a temperature sensor.
If a rider is only going to use the pedals as foot rests or at best soft pedal and expect the motor to drag them along at 40kph and last more than five minutes then I'd suggest a stronger motor.