Did you ever get a reply about the safety issue from Elaine at Li Time? I am not able to get the bluetooth discharge switches to work when my batteries are wired in parallel, so I was not able to turn one off then on again to see if both batteries would finally take the load.
I have the same experience as you with parallel "Battery System" set up in the app which is essentially lipstick on a pig. It is only a layout representation and has nothing to do with the actual battery conditions as shown by my DC current clamp meter.
The result is that I have 2 batteries that I am not able to use wired in parallel. I have no doubt that the issue is the Li Time BMS. I am trying to return the batteries and get a full refund. I have not heard back from Li Time yet about the refund. Hopefully I get a response by tommorow. If they are avoiding the issue I'm not sure what to do after that.
2 100ah Li Time batteries were $606 USD.
2 100ah Battleborn batteries are $1850 USD.
I did save $1244 over high quality Battleborn units by going with the cheap Chinese Li Time batteries. However I cant use the cheap Chinese Li Time batteries in parallel. Moral of the story, buy cheap, buy twice.
Unfortunately Li Time does not have a good track record backing up their warranty, so likely the tuition for my continuing education is $606.
I got a response from LiTime when I raised the safety issue and pointed them to this post. It was after they asked the same questions about the load at the time of the problem (was it over 1A? -- yes, 5A) and was I charging at the time (no). Here is what I got from Ricket:
-----
The reason why the batteries are discharged in sequence is that the voltages of the terminals of the two batteries are different. The battery with a larger voltage bears all the discharge current first. When the voltage of the battery with a larger voltage drops due to discharge, and the voltage drops to the same voltage as the battery next to it, the two batteries can be discharged together. This is the normal discharge phenomenon of batteries in series and parallel.
Possible questions: The battery APP monitors the battery cell voltage, not the battery port voltage. After the full charge protection occurs, the port voltage will drop by 0.4V~0.7V compared to the battery cell voltage.
After the overcharge protection, the voltage will be 0.5~0.7V lower than the normal full charge voltage;
It has no impact on the normal use of customers, but only avoids the oscillation caused by frequent charging;
-----
The response does not address how one battery ran down to 18% (12.8V -5.8A) while the other stayed at 100% and did not discharge. The say this is "normal"? I'm pretty sure a parallel battery setup free of control by a BMS would not do this.
I don't know if we can't understand each other or if they are obstinately hiding behind the language barrier.
I notice they use the term "full charge protection" and wonder if this is what "standby/full" means. The battery that does not discharge reports it is in this mode. The discharging battery starts in plain "standby" mode, when they are both fully charged, and reporting the same voltage (13.3).
On their point about the voltages: I know the reported voltages are not so accurate because they seem to use a high capacity shunt, but certainly while falling to 12.8V an initially higher voltage battery must match the lower batteries voltage at some point, to allow them to "discharge together", which never happens.
I will try to address this with LiTime. Argh.
With my method of turning off "discharge" on the discharging battery, which wakes up the lax one, I do seem to seem a brief period of one battery charging the other with the app. But as you know, it is only possible to look at one battery at a time, and it takes a while to switch which one is being read. I haven't gotten a meter under there.
The fuse idea is a good one, but hard for me.