Issue with regen from old phaserunner to new phaserunner on bafang BBSHD

[thender]

I replaced a dead phaserunner MT that I used with a cycle analyst on my Bafang BBSHD with a newer L10 v6 phaserunner. I would've used a direct replacement but the old MT is not for sale anymore. This wiring harness worked just fine after replacing the anderson connectors for the motor phase wires with gardner bender spade crimps.

I can't get the controller to stop doing what seems like regen braking or limp-moding the motor when the throttle goes over 3.8v. As soon as my throttle goes over 3.8v on the cycle analyst, it feels like the controller is trying its best to stop the motor. If I step down below 3.9v on throttle it will then begin to go full speed again after a second(which is the regen braking torque ramp down time)

Regen braking is absolutely useless to me since this is a mid-drive and even if I did want regen I would never want the throttle being *HIGHER* to cause regen to happen.

I attached my parameter file. I made sure on addr 212 to uncheck engine braking & analogue braking enable. I set braking start/max voltage to 0.1v-0.3v so it would be completely out of band of my throttle to begin with since my cycle analyst has the throttle set to start at 0.99v and max out at 4.3v. I set regen braking current to 0.

I am out of ideas.

Looking at this file, do you have any insight into why the motor gimps itself when my throttle goes over 3.8v?

Thank you!

 

[amberwolf]

Just to be certain:

Is it literally just the throttle voltage going higher than 3.8v, or is it the motor spinning faster than a certain RPM or pulling more than a certain phase current, because of the throttle voltage? Or some other limit being reached?

If it's the latter, it could be tuned incorrectly for the motor (but if you are using the same motor parameters and PLL values, etc) as before that shouldn't be the case).


I'm sure there are differences in the base ASI firmware in the controllers between the MT version and the latest, but I can't imagine a reason they'd behave this differently with the same settings between them.

What specific settings (or advanced parameters in the dev screen) are different (if any) between what you used on the MT unit, and waht this new one is set to?

 

[thender]

Is it literally just the throttle voltage going higher than 3.8v, or is it the motor spinning faster than a certain RPM or pulling more than a certain phase current, because of the throttle voltage? Or some other limit being reached?

If it's the latter, it could be tuned incorrectly for the motor (but if you are using the same motor parameters and PLL values, etc) as before that shouldn't be the case).

I don't have enough to answer your first question unfortunately. I don't have a way to tell RPM while on the bike. I know I am below the RPMs I used before since I have my field weakening amperage set way lower than before, and I am way below the power I have used before since I used to use this on the MT with 96 phase amps and the same tuning(before they disabled you from doing that, because... it blows up the controller ) )

What specific settings (or advanced parameters in the dev screen) are different (if any) between what you used on the MT unit, and what this new one is set to?

I tried a firmware file directly from the old phaserunner mt so everything was the same, the only thing I had to adjust was the throttle input since this one ties throttle/brake to the same line if you are using a cycle analyst, and change the source/sensor input for the throttle line so it could see my throttle.

The old phaserunner, to my knowledge, was based on the BAC800. My new one is based on the BAC855. Inbetween these two revisions, they modified it so you cannot use 96 phase amps. This was causing them to blow all the time. It's limited to 90 phase amps now. That could be it. Some hard software limit that gimps it at 90 phase amps, or when it gets *CLOSE* to that so the controller doesn't get destroyed.

 

[amberwolf]

If you were actually using those few extra amps, then I could see how it could be limiting to not have them...but it still wouldn't actually *brake* the motor, which is what you describe happening.


To brake the motor it has to either reverse the direction of the currents it's sending (common FOC braking), or stop sending currents and repeatedly short the phase windings (common cheap braking).

To cause that, either a limit has to be being exceeded, whose result is to force the controller to try to slow or stop the motor, or a tuning issue with how the motor responds to controller drive signals (if those are different for the L10 vs MT).


Is it safe to assume you've already tried using the PR directly without the CA attached to it at all, and gotten the same results whenever throttle goes above 3.8v?

Regarding RPM while on the bike...could you setup the CA's speedo input to hook up to a motor hall instead of wheelspeed sensor? (the PR can stay hooked up to the WSS). Then you can set teh CA to the correct pole pairs for the motor (instead of 1 for a typical WSS),and see RPMs while riding (you'll see them as a road speed, but you can convert that back to RPM easily enough especially if you change the wheel size to an even decimal number so it's just another power-of-ten in the calculations).


When you say "firmware file" do you mean "settings file"? I would guess that any actual firmware is not compatible between the two hardwares and could cause all sorts of problems, if you could even get the hardware to accept it.

When you load the settings file from MT into L10, what does the PRSS program displayfor each value when you read those back (after a complete power cycle, and closing and reopening the PRSS program), compared to doing the same in the MT? I ask this because fi any of the MT version settings aren't compatible with the L10, it probably either discards it and uses a default value, regardless of what is saved in the settings file, or it uses the closest value it can to the saved value.

So comparing what the PR's are actually set to vs what the files used say they should be set to may be useful to find a "faulty" value / setting that might cause a problem (even if it isn't this one).

 

[thender]

That makes sense about the regenbraking. The reason I thought it had something to do with regen braking is because grin did something when advancing from the phaserunner MT to the L10 V6 where they tied the throttle and ebrake together. I wasn't getting any movement when I flashed my old firmware to the new phaserunner, which made no sense since the hall sensor test worked. I noticed in the dashboard the brake voltage was going up when I ran the throttle up and I had to switch them.

Changing the brake/throttle from one version to another is just plain evil - on par with future motion using the XT60 power connector in reverse!!!

I tried it a few more times. It looks like as soon as I go over 25-26 amps with the throttle around 3.0-3.6v, it dies. With the throttle over 3.8v, it goes down to 1 amp and slowly climbs up to 3 amps. If I keep it on for a long time with throttle over 3.8v, it'll make its way up to 9 amps. If I lower the throttle below 3.8v, it'll go back to wherever it was before it went limp, back to 25-26 amps. It's jarring. The best I can describe, it feels like field weakening in reverse - at the point I'd expect to get more RPMs as I enter field weakening, I get the opposite. I tried turning off field weakening, to 0 amps, just to see what that would do.

I haven't used the new phaserunner without the old CA attached to it because the new PR has a different throttle connector which I don't have an adapter for here

In terms of comparing it to the MT's reaction to the same settings, that doesn't boot anymore. It's just a pile of smoot. I'll see if any of the settings revert. Thank you for giving me some ideas. I sent an email to grin with my config file to see if they can tell me what I messed up.

 

[Papa]

I tried a firmware file directly from the old phaserunner mt so everything was the same

Curious... Did you try configuring and testing the V6 L10 prior to loading the MT firmware? Just wondering if the V6 L10 has regen configured by defalt.