These are a common PACE or similar BMS like PBMS, I’ve worked with them a lot. The LCD and button layout is a tell-tale. There is also another similar cheap BMS that comes in batteries like these, i have the software for both if you can’t get it elsewhere.
Don’t confuse SOH with SOC.
You will find the SOH (state of heath, not SOC, state of charge) will drop in this BMS randomly, and too fast.
The idea behind SOH is that you decrement the SOH counter when certain events occur such as 100% SOC, 0% SOC, high current, low temp, etc etc. If you decrement the counter correctly, this would ideally track the capacity as it diminishes over time, and prevent you trying to put say 120Ah into an old battery that can now only take say 90Ah.
As SOH decreases, the available capacity (or at least the apparent available capacity) decreases, so with a SOH of 50%, your SOC will drop from 100% to 0% after only 50% of your amp-hours have been drained.
The issue is that this happens far too fast in some BMS brands. I’ve seen one battery (that looked exactly like your one) that was losing about 0.75% per day from the SOH and therefore was apparently full to empty in minutes. The cells were actually in good condition, just the BMS was useless.
You mention both SOH and SOC concepts, so i’m not sure what you are saying about your battery, but if it doesn’t charge to 100% SOC (the number reported in the LCD) by the time the voltage is ~56v, you should take it back and demand that MM calibrates the shunt. If the vendor won’t take action you will have to threaten using the CGA.
These batteries generally arrive not calibrated, as the BMS is slapped in them and shipped, and often just set to 50% when the cells might have been delivered from the cell manufacturer at 65%, so they are not accurate or precise from the get-go.
To get them to show the actual SOC, you need to set the SOC in the software - message me if you need the software. This gives you accuracy at one point, but not over time or charge. For this, you must calibrate the shunt.
FYI: you might know this but for completeness; The shunt measures how much current is flowing in/out of the battery (amps), and by keeping track of this over time (hours), you can know amp-hours (ie capacity, aka SOC)
You can set the calibration on the shunt in software, for both charge and discharge. I can almost guarantee that MM don’t update the firmware, or calibrate the shunts.
There are a number of issues with the BMS. They are underpowered from a CPU point of view, and we have seen situations in stacks of 3 to 6 batteries where the master BMS (in your case the top one) requests updates from the other BMS units (by number, as determined by the red DIP switch settings), but some BMS units, some of the time, are busy processing messages on the bus and don’t respond. This leads to all kinds of issues, such as the pack SOC (sent from the master BMS to the Victron/GX/Growatt/Etc) jumping up/down - for example if the batteries are at 50, 55, 45 and 60%, the master should send “53” to the GX (210/4), but if the 60% bms doesn’t respond, the master sends 38, then next round (about 20 seconds later) gets a reply from that BMS and goes back to sending 53.
This also happens with current. Each BMS ‘requests’ a certain current, say 40A, so the master should say “160A” to the inverter/GX, but it will sometimes jump down to 120A if one BMS fails to report back.
You can get horrible crashes with these BMS - especially when the BMS counts down too slowly under discharge (due to the lack of accurate calibration of the shunt) so that the Ah counter still says maybe 40Ah, when in fact you are at 10Ah left, and a crash is not far away.
The opposite issue is not so bad … if the counter is counting up too fast under charge, and reaches say 120Ah (for a 120Ah battery), it will be reporting 100% even though the battery is not full and the charge is still flowing. However, this depends on your mppt - if you have a Growatt and are using comms, the battery master BMS will tell the growatt to stop charging (“Give me zero amps - i’m full”), and your battery is now at say 80%, but reporting 100%, and asking for zero amps - and will never get fully charged.
If your mppt is working on voltage, it will keep charging and might eventually get to a full charge.
Another situation is that your battery is full, but still saying less than 100% (ie charge counter is counting too low) - here your mppt is still ramming the charge in (the BMS is saying “I’m at 80%, give me all you have (but not more than 100a)”) and your cells are at 3.65. The over-cell voltage alarm will go, and the BMS will panic and send a high voltage back to the mppt, forcing it off. The BMS then resets and asks for charge again, the mppt charges, and again the alarm is raised. After the third loop, the SOC is force-reset to 100%, the SOH is decremented (because we just shortened the life of the cells by putting them over-voltage 3 times) and the situation calms down.
Another issue with some of these BMS units is that with a particular firmware, instead of reporting the average SOC of the pack (back to the inverter/GX/etc) they report the lowest SOC in the pack on discharge, and the highest on charge, so your pack is charging, gets to 100%, the mppt backs off, the pack starts to discharge a tiny bit, and because the pack is not balanced (ie one is at 100%, but one is at 90%) the master BMS now reports 90% to the inverter/GX and suddenly charging starts again.
In short, these BMS units need very careful setup and tuning, and even then, they don’t compensate very well for temperature, the Peukert exponent, or any of the tens of minor factors that distort the actual amount of charge that is stored in the battery. But … they are cheap, and make for nice margins, and combined with B grade cells, the weakness of the BMS is somewhat masked by the performance of the cells.
If you don’t already have the software, ask MM for it - i’m sure he will refuse, as this ensures you wont be able to see the cell difference (which is a good tell for poorly matched cells, or B grade cells). Seeing what is happening in the BMS from time to time is critical, otherwise you only have 1 metric to go on - is the battery delivering about the right number of kwh from full to empty. With the software, you can see the difference in SOH and SOC between batteries, the difference in cell voltage across the 16 cells, and things like alarm logs.
You can also reset your SOH to 100%, and reset all your SOC’s to the same value (which they should be if the shunts are calibrated, and the batteries have similar internal resistances, and the batteries have been in a pack for a couple of cycles)
Also, not wanting to nit pick if your setup is only temporary, but heat shrink is cheap, and having red heat shrink on the end of negative cables, and black on the positives, is asking for trouble when a “helper” puts the wrong lug in the wrong place and gets showered in molten copper.
It looks like you have the data cable for the RS232 port, but if not i’ve got those too, and USB to serial adapters.
I would very very very much doubt these are anything like a pylontech.
@Marc_HD Its all just serial, so unless they use a terrible chip, you generally can’t break them.
RP>
