The BMS has no way to do anything other than complete shutoff of power when a limit is reached. Only the controlller has the electronics to reduce power output when limits are approached or reached.
I don't see any cells dropping too low. But I don't see any current varying normally. It doesn't vary, it just jumps to the few values it ever displays, meaning the BMS isn't sampling it very often or else it isn't bothering to send that to the display app, or the display app isn't bothering to update the screen very often with the data it does get. Either way, it's pretty tough to tell what's really going on like that.
The only non-error current readings are -49.5A, at 56.3v, -213.3A at 53.1v, and -242.2A, and voltage reads 53.2v at that moment.
Except for those, everytime it shows any current reading, it shows an overcurrent error: once for negative 327.2A, voltage drop only to 50.3v, (when I see it shut off the charge and discharge FETS the first time) and once for 326.3Amps, and voltage sags to 49.7 volts at that moment, giving a Sum Volt Low Level 2, which is the total pack voltage below the limit (presumably 50v?), when it shuts off again.
Note that amps should read positive, but they always seem to read negative, so either there is a problem with noise in the BMS's current sensor or software, causing it to think it's negative when it's actually positive, or there's just a plain bug in the display app doing that. If the limits set in the BMS for current are positive numbers, maybe when it reads out a negative current it's shutting down because it considers that too far out of limit, for instance if it's set for say, 350A positive current as a limit, but it reads -326A, the difference betwee those is over 600A, and it may shutdown because even though it didn't reach 350A, the software still sees it as almost 700A difference and thinks that's almost double it's limit. Just a thought. If there is a way to set negative numbers for current, maybe that will make a difference?
It's possible there is just a plain fault with the BMS causing the shutdown problem, most likely a bug in the software inside the BMS itself if this is the issue.