diy solar

diy solar

JK BMS Losing Capacity Calibration Over Night?

MWeiss

New Member
Joined
Apr 22, 2020
Messages
73
I've had my battery bank running three 16S 280Ah groups of cells with three JK BMS running in parallel to a common set of buss bars for about 6 weeks now with no issues once I ran the batteries through a full discharge cycle. It was always accurate after that.. until this morning. The batteries got down to 20% overnight and sometime this morning, the second BMS reset it's "full" state to 100% when the battery was really below 50% charge.
Since everything is paralleled, the load never causes more than 2-3% difference between battery banks.

Is it common to have a BMS lose calibration like this? Note the Ave Cell Volt readings. I've turned off charging on the 99% one at the time of this capture until I figured out it wasn't REALLY 99%.

The only event that I suspect may have caused the glitch was that charge current went over 100A for a short bit as we had some diffracting clouds boost my solar arrays output by 130% of normal for a few minutes and the BMS hit the charge current limit and shut down charging until I set the charge current to a higher value. Possibly that condition caused a glitch, but would like to know if others have encountered this loss of calibration on this BMS before. These are 200A BMS units.


1659128823982.png 1659128831073.png 1659128838819.png
 
That looks a bit odd, voltages definitely don't indicate 99% SOC. Looks to me that the BMS have had an error and restarted, loosing some data...

I have 2x48V/230Ah battery banks, each with 200A JK BMS in parallel and they indicate SOC quite differently, I may have 60% on one and 45% on the other and it rarely adds up to what the Victron Smartshunt 500 indicates. According to Hankzor, this is normal when discharging the batteries using a sinus inverter... https://diysolarforum.com/threads/jk-bms-200a-24c-design-flaw-in-soc.41900/

In my mobile home, I have 3x125Ah LiFePO4 batteries with built-in BMS (that kind that looks like ordinary car batteries) and have NEVER seen more than 1% difference in SOC between them for 3 years but they are mostly discharged by other loads than a sinus inverter.
 
I'm watching it closely now. I took the system off line today and let all the batteries charge to 100% and let the BMS hit the charge OVP threshold so I KNOW they are all full charge. Now they seem to be all in agreement. Will see how they track as I draw down the voltage overnight.
 
I had a similar happening yesterday. Both my batteries were out of use and SOC 100%. Suddenly one of them went down to 84%. After a few hours, I turned on the solar charger and within minutes, it were at 100% again. This would normally take hours... But cell voltages told a quite different story than SOC.
 
I wonder what causes these "errors" to happen? Random cosmic ray hitting a transistor inside the BMS? Or impending failure?
 
Had it happen again today. Also just noticed that the days running time on BMS #2 is 64 when the other two in the trio read 58.
In a span of ten minutes (I've been checking frequently) the charge state went from 60% to 100% on BMS #2.
 
The bms' will require a full cycle to recalibrate. If charge amps went over it would indeed cut and the numbers will be skewed. Sadly few SCC have VOC which handle momentary over amp/volt production. Midnite Solar being one such.

BTW. We did get hit by Solar EMPs recently, couple of biggies too.

Do be aware that if you are using the defaults for LFP, they're crap and some bad settings are in there. This is now corrected & reflected in the new user manual. Link below.

Good luck, hope it helps.

 
Back
Top