Hello.
I purchased a Phaserunner / Gmac 10T / CA3 setup from Grin back in March. I had them lace the motor into a 26 inch wheel for me. I had my own 14s LiIon battery.
I got everything setup and running on my Kona MTB.
During acceleration everything is fine. No faults.
However during strong regen braking, the PR was tripping Instantaneous Phase Over Current, blink code 2-2. This normally occurs near the 'end' of regen braking, for example, braking from 40 km/h down to 20 km/h is ok, then the over current trips.
I contacted Grin about it at the time, but in over 2 months, the issue still remains. I'm wondering if anybody else has any ideas?
Troubleshooting steps I've taken myself, or as directed by Grin:
Reset PR to defaults, and loaded gmac 10t preset.
Tried changing regen torque engage ramp from 200ms to 500ms.
I tried lowering the current regulator P from 0.75 to 0.3. This made it much worse. Over current tripped within 1/2 second of regen.
Raised P incrementally up to around 3. No change. About 3 seconds after applying regen, over current would trip.
Tried raising I from 300 to 1,000 with no change. Also tried higher P and I simultaneously.
Changed current regulator bandwidth to 1,000 rads instead of 0. I understand this stops using P and I and does things more 'on the fly'. Still tripped over current.
Raised PLL bandwidth from 800 to 1,000. Also tried raising PLL damping from 2 to 5.
Lowered regen phase current from 90A to 62A and it still trips.
I tried lowering battery regen current from 30A to 15A with no help. Didn't expect that to help as the fault normally doesn't occur at high speed where you'd be more likely to be battery current limited, but instead occurs at lower speed where you're probably phase current limited.
I also tried a different rider, as I am around 250 lbs. I had a 160 lb rider try, and they were able to reproduce the fault as well.
Checked stator play by opening up the motor at the suggestion of Grin, but it seems to be within their spec.
I do not believe the cycle analyst has anything to do with this, as I have a separate throttle control wired direct to the PR to command braking, thus bypassing the CA. The CA is still aware of the regen battery current of course, but it's not the one requesting it. Throttle / brake voltage mapping was setup in the PR correctly. Appx 1v -> 4v.
I finally tried lower regen phase current all the way to 40A, and this did help. It no longer triggers phase over current during braking. However this of course has drastically reduced the amount of braking that's possible at medium to low speeds. High speed regen is still battery current limited of course, but low speed braking is rather poor now. I definitely have to use the friction brakes a lot more now, but at least I can rely on the regen not to cutout and the subsequent need to pull over and power cycle the PR to clear the fault.
Aside from making it worse by lowering current regulator P, the fault never occurs 'instantly' or 'quickly'. It was always 2-3 seconds after it had been successfully doing strong regen. If I could see a debug graph of the phase current, or even better, a graph of the P and I terms it would probably help a great deal. I'm somewhat familiar with PID controllers from flying FPV drones, but there you could always look at debug graphs to help determine where the problem was. For example, I'm not sure if the current control loop is too slow to respond, and the error (and therefore phase current) gets too high, or if the control loop is too fast to respond and it takes 2-3 seconds of quick oscillation for it to finally spike too high.
Anybody have anything else I can try changing to get regen phase current back up without over current faults?
Thanks.
I purchased a Phaserunner / Gmac 10T / CA3 setup from Grin back in March. I had them lace the motor into a 26 inch wheel for me. I had my own 14s LiIon battery.
I got everything setup and running on my Kona MTB.
During acceleration everything is fine. No faults.
However during strong regen braking, the PR was tripping Instantaneous Phase Over Current, blink code 2-2. This normally occurs near the 'end' of regen braking, for example, braking from 40 km/h down to 20 km/h is ok, then the over current trips.
I contacted Grin about it at the time, but in over 2 months, the issue still remains. I'm wondering if anybody else has any ideas?
Troubleshooting steps I've taken myself, or as directed by Grin:
Reset PR to defaults, and loaded gmac 10t preset.
Tried changing regen torque engage ramp from 200ms to 500ms.
I tried lowering the current regulator P from 0.75 to 0.3. This made it much worse. Over current tripped within 1/2 second of regen.
Raised P incrementally up to around 3. No change. About 3 seconds after applying regen, over current would trip.
Tried raising I from 300 to 1,000 with no change. Also tried higher P and I simultaneously.
Changed current regulator bandwidth to 1,000 rads instead of 0. I understand this stops using P and I and does things more 'on the fly'. Still tripped over current.
Raised PLL bandwidth from 800 to 1,000. Also tried raising PLL damping from 2 to 5.
Lowered regen phase current from 90A to 62A and it still trips.
I tried lowering battery regen current from 30A to 15A with no help. Didn't expect that to help as the fault normally doesn't occur at high speed where you'd be more likely to be battery current limited, but instead occurs at lower speed where you're probably phase current limited.
I also tried a different rider, as I am around 250 lbs. I had a 160 lb rider try, and they were able to reproduce the fault as well.
Checked stator play by opening up the motor at the suggestion of Grin, but it seems to be within their spec.
I do not believe the cycle analyst has anything to do with this, as I have a separate throttle control wired direct to the PR to command braking, thus bypassing the CA. The CA is still aware of the regen battery current of course, but it's not the one requesting it. Throttle / brake voltage mapping was setup in the PR correctly. Appx 1v -> 4v.
I finally tried lower regen phase current all the way to 40A, and this did help. It no longer triggers phase over current during braking. However this of course has drastically reduced the amount of braking that's possible at medium to low speeds. High speed regen is still battery current limited of course, but low speed braking is rather poor now. I definitely have to use the friction brakes a lot more now, but at least I can rely on the regen not to cutout and the subsequent need to pull over and power cycle the PR to clear the fault.
Aside from making it worse by lowering current regulator P, the fault never occurs 'instantly' or 'quickly'. It was always 2-3 seconds after it had been successfully doing strong regen. If I could see a debug graph of the phase current, or even better, a graph of the P and I terms it would probably help a great deal. I'm somewhat familiar with PID controllers from flying FPV drones, but there you could always look at debug graphs to help determine where the problem was. For example, I'm not sure if the current control loop is too slow to respond, and the error (and therefore phase current) gets too high, or if the control loop is too fast to respond and it takes 2-3 seconds of quick oscillation for it to finally spike too high.
Anybody have anything else I can try changing to get regen phase current back up without over current faults?
Thanks.