VESC FOC for a flywheel application....

[sniper1rfa]

I'm attempting to use a VESC to spin up a 5lb flywheel. Motor is a T-Motor MN4004 400Kv on a 6S battery.

Originally this was run using a castle creations Talon25, which worked well on a 3lb flywheel. Switching up to 5lb prevented the talon from successfully commutating, and prevented startup.

I've gone through the basic configuration of the VESC in FOC mode, and got mediocre results with no load. It appears to run and commutate, but it has a small amount of chatter.

With the flywheel loaded I don't get a successful start. The reported RPM is correct, but the motor is clearly not following along.

Not sure where to start, or even what information to provide to give a clue about the problem. I was not honestly expected FOC to have problems starting, since the CC Talon was so close to working properly.

Thanks.

[linsus]

It's a fairly common to have some problems with load startups in non-sensored FOC. Might want to consider trying sensored for a smoother operation in startup.

[sniper1rfa]

Easier said than done, unfortunately. OTS flat pancake style motors with 24V windings, sensors, and a kV above 400 are pretty rare and expensive (I think only maxon has one, non-stock.

I have some hall sensors and will set them up with a bracket if necessary, but it's not ideal.

I got one flywheel spun up yesterday, but only just barely. Switching shunts out to 5mohm to try to get a better read on the small motor.

[sniper1rfa]

update: switching to 0.005ohm shunts fixed it. Starts smooth as butter now.

[sniper1rfa]

another question:

The current plots are biased - is this normal or something to do with my crappy soldering?

[ThierryGTLTS]

Same problem for me and don't know where it comes from

[sniper1rfa]

Had another opportunity to play with this. Running with commanded duty cycle in FOC mode.

Startup is now reliable, and below about 60,000 ERPM the motor operates reliably, albeit with a small amount of hunting/wander. After approximately 60,000 erpm the motor cuts off and coasts to a halt.

I keep seeing mention of 60krpm being a limit, but this doesn't jive with the 100krpm specification. Am I running into some kind of RPM limitation that will be insurmountable, or is there something else I can change to get better high-rpm performance? Is this RPM cap going to be the same if I switch to BLDC mode with hall sensors?

The goal is to operate at 90,000 ERPM with minimal variance.

I'd love to get this working with a single speed controller, but if I have to I'll use the VESC for startup and then switch to a traditional BEMF controller once the RPM's are high enough.

[ThierryGTLTS]

VESC 6 seems to be reliable up to 100krpm.

Ready-to-use Trampa VESC6 is tested and reliable.

See also Stewii test here http://www.electric-skateboard.builders/t/hw6-4-based-esc-escape/37579/219

Have a Nice Day.

Thierry

[rew]

You can even run the latest firmware on older hardware to increase the achievable ERPM limit. (It's a software issue, so a software fix will fix even old hardware).

[arvidb]

rew, that's interesting: can you elaborate a bit on that? I.e. what has been fixed in newer firmware that enables higher ERPM? My belief was that it was the three shunts of VESC6 that enabled it to run at higher ERPM, but my understanding of this is fuzzy at best.