This algorithm is pretty much another way of describing how HFI works. Instead of using sampling and injected square waves he modulates edge aligned PWM to inject frequency content and then filters the signals with some carefully designed analog filter stage. We achieve a similar result by injecting a frequency and then using digital filtering instead of analog.
To use ASS (hate this name) you need to have switchable filters on the phase voltage dividers. These will also give a better estimate of winding resistance measurements in some instances (probably) since it avoids the need for dead-time compensation to determine phase-voltage amplitude. Most vesc-based hardware currently doesn’t have these but they aren’t terribly difficult to implement in new hardware.
@glyphiks Now that Trampa is selling 4WD board, hopefully they will experience it and have it fixed, sad that this has to happen for something dangerous to get attention
And maybe it is fixed, the PPM rewrite by @Pimousse may have done it, by making the command sent over CAN bus a %, no a current value
This may also solve the problem I had in the past that if the master belt was broken and you didn’t have traction control enable, you had no power on the slave side
I didn’t address this issue in my rework. Sorry.
But it defo makes sense that TC can mess things up while braking.
Actually, I don’t like TC. Although it works great when everything’s fine, it has too many undesirable sides effects.
Is anyone able to update from android vesc tool?
Seems like it keeps uploading, restarting and showing the same version, and now since i updated from google play i don’t have the older apk anymore
I mean just by sending a % value, as I understood for some reason the current value sent to the slave was in part proportional to the current value of the master motor, this would cause the current to drop negatively (braking) on the slave loses traction
For me, even on 2WD TC is a must, or I eat sand due to one wheels spinning
