diy solar

diy solar

XiaoxiangBMS iOS Strange Values & Charge Calibration

MadMacks

Off Gridder
Joined
Sep 19, 2020
Messages
56
Location
Virginia City area of Nevada USA
I’ve been running the XiaoxiangBMS software for about a year and today, for the first time, I had to run my generator to charge my battery bank (LiFePO4, 2P4S, 560ah using Eve LF280 cells). I’m using a Shenzhen E-Fire LFP-412120-COMM BMS (4S 120A), and Lilly Chan, their support person said the XiaoxiangBMS software supports that BMS. There is apparently a WinDoze desktop app that works with USB as well, but I haven’t used it.

Anyway, when I started my generator to power my Aims inverter/charger, rated at 75amp +-5, I watched the charge current run up to about 60 amps going into the battery, and then the charger would trip and the current would go back to zero. This happened several times until I throttled back the Aims to only output about 50 amps (the unit has a dial to limit the amp output if needed). I wasn’t sure why I needed to do that as the BMS is rated at 120 amps (and showed no overcurrent fault), but after going thru the Aims manual I found that it will shut off if the input AC current drops out of a very specific current and hertz range. Then I realized I was using a small 2000/1600 watt (start/run) generator at 6,500 foot elevation and you lose about 3% power for every 1000 ft in elevation, meaning that I was really only supplying about 1,280 watts of power to the inverter and when it tried to go above about 50 amps the generator was dropping out of the allowable voltage and cycles range, so the Aims would shut down. Anyway, once I figured all that out I started charging the battery at about a 50 amp rate.

All that is preamble… Here’s the strange part. As I watched the BMS ramp up the charge of the battery from about 70% SOC to close to 100% SOC I couldn’t figure out why the software was reporting the SOC at 100% but showing 557ah out of 560ah, and yet I was still pushing 40 or so amps into the battery and the 557ah number wasn’t increasing. I know these BMS’ report data that the software interprets into a SOC calculation so I figured no big deal, and that the battery would continue to accept a charge until it hit high voltage cutoff. I’d been charging for a couple of hours prior to hitting the 557ah/560ah 100% SOC so I started working on other things. After about half an hour the charger cycled off so I check the BMS and it was still at 557ah. Anyway, the charger had cycled off because the generator was out of gas. No big deal…

So, I was watching the BMS as it started reporting current draw and it showed about 2.5amp current coming out of the battery. I also have a digital meter and shunt connected independent of the BMS and it was showing about 6.5 amps coming out of the battery, so I started looking for a way to calibrate the BMS thinking the in/out current calls were way off. The XiaoxiangBMS software does have a setting page to calibrate the BMS so I tried to adjust by about 4,000 mAmps but apparently it won’t let you save the calibration adjustment if you’re drawing current, so that failed.

BUT, when I went back to the main display page the software was now reporting a SOC of about 71% and 398ah out of 560ah. Damn!

So, I am thinking something is funky with the software and/or the stored values on the BMS. Right now the software shows a current draw of 1.62 amps but my meter/shunt shows a draw of 5.61 amps. So, does anyone know of a better software package I might be able to use to recalibrate the BMS? I’m connected with Bluetooth via iOS, but am wondering if I should switch over to a UART usb connection and try a WinDoze based software package that may work better with the BMS. A 4 amp difference between the software reported value and the meter/shunt is way to much. I live off grid with this battery so disconnecting it to reset the BMS would be a real PITA as I’d lose power to my refrigerators and other items in the cabin.

But somehow I need to give the BMS a wack to get it reset and tracking the current flow more accurately, or I really don’t have a clue as to my SOC because LiFePO4 batteries don’t tell you much about SOC using voltage alone.
 
So I found the Overkill BMS calibration instructions and will give them a shot. The XiaoxiangBMS software reports the BMS as manufactured by DGJBD and the JBD-SP04S020-L4S-120A-B-U model even thought the sticker says Shenzhen E-Fire LFP-412120-COMM. I assume the firmware on the BMS is correct so that’s essentially the same as the model sold by Overkill, and I’m hoping the instructions will resolve the issue. But, I’m now looking to see how to re-zero the overall capacity and “current” state of charge recognized by the BMS. The battery was top balanced originally, but I’d rather not go thru that process again if I can help it, although I’m not sure I’ll be able to avoid it. Sucks, cause like I said, the cabin runs on this battery and I don’t want to be “off line” to long or I’ll lose the food in the refrigerators! Ugh.
 
Not sure what happened there. I did try to make a comment which was -
It seems like you're having the same issue as me with the BMS changing the state of charge, estimating it from the cell voltages, at unwanted times.
 
Thread 'jbd / overkill bms - state of charge'
 
The android app allows you to reset the state of charge apparently. The iOs app does not aa far as i can tell.

In the ios pro version app you can change all the cell voltage thresholds to 3000 mV to force it to 100%. It doesn't solve the underlying problem though.

The current calibration is separate issue that you can hopefully resolve.
 
What do you mean by the “iOS pro version”? Do you mean the paid version? I have that. But I don’t know about a “pro” version?

I’m running “Xiaoxiang BMS Monitor” version 1.2.24 Build 5496”, which as far as I know is the latest version.
 
I found some more info - check it in my thread, see if it makes any sense to you
 
Back
Top