diy solar

diy solar

Chargery BMS16T 4.05 issues

scpanish

New Member
Joined
Jul 29, 2021
Messages
26
Location
NH, USA
I'm wondering if anyone has insight into the issues I've found before I plague Jason. I've just put together two paralleled 2P8S batteries using Eve 280AH LiFe cells, each with a separate port Chargery 16T v4.05 BMS and two DCCs. This is my first LiFe build, and it will first replace my 24V FLA battery bank and later be reconfigured as two 16S batteries, paralleled, when I replace my 22 year old Trace 4024 inverter/charger with a 48V inverter. The Eve cells came from Amy at Shenzen Luyuan and from the specs (and measured voltages) were extremely well matched, so I didn't top balance. The bank is only connected to a 10A power supply. The wiring is as shown in the DCC manual's diagram for separate port configuration with two 8S batteries.

The first oddity is that (even) with the power supply off, both BMS displays show charging current. Battery 1 shows 4.6A and battery 2 shows 3.5A. With the power supply on at 10 amps that goes up to 9.2 and 8.7 respectively. So battery 2 is taking a bit more current. The resistances should be identical since the batteries are constructed identically except for some minor differences in cable lengths going to the bus bars. All connections are tight although I didn't use a torque wrench. One side effect of this issue is that I can't put the BMSs into storage/sleep mode. I have to turn them off to save power.

Second, both display units show all cell resistances to be 0. So that feature just doesn't seem to be working. Bug, or related to my configuration? I know the paralleled pairs of cells are seen as single cells by the BMSs. So I would think it would still yield a resistance measurement, but I'm pretty ignorant about how that works.

Perhaps related, battery one began alarming after one day of charging due to differential voltage exceeding the 30mv default threshold. It turned off the discharge DCC. Typically when charging, the delta ran about 39mv and one cell pair was clearly the outlier, consistently high. Tomorrow I'll measure the voltages of each cell separately during charging. With charging off, both batteries have differentials around 10mv. I raised the threshold to 40mv to stop the alarming.

This may warrant a separate thread, but does anyone know if it is OK to use the charging function of the inverter/charger with a separate port system? I hadn't considered this when I ordered, and had not seen the diagram in the DCC manual showing the charge DCC being common port. Seems to me that the inverter/charger will charge the batteries but can't be shut down by the BMS since current will be feeding in through the discharge bus.

Thanks to Steve S for the great job on the manual! That must have been quite a chore.

Steve P
 
Correction - Duh..no issue with the differential cell voltages and alarming. I hadn't tightened down one nut on an interconnect on the offending cell. Dumb.

But the other issues remain. I hope some similarly silly explanation exists.

Steve
 
Jason's advice was to recalibrate the shunts after making sure the connections are correct (they are). Apparently calibration was not done correctly by Chargery. Annoying, but it should be doable. I hope the procedure is clear in the manual. The IR is only measured when current is (really) flowing so that is why it is reported as 0.
 
I've got a similar setup to you. Cells from Amy, and a Chargery BMS. Even though Amy's cells are well matched, they were no-where near balanced. It took my dedicated 4amp active balancer about 2 days to top balance the cells.
Internal resistance measurement doensn't work until you've cycled the cells a bit.
Sounds like your packs have different resistances, hence you're getting different current between them. Also make sure they are at the same state of charge before connecting them. The packs will have different resistances at different states of charge.
Make sure that all cable lengths are as similar as possible. Other than that, I wouldn't worry about it as they will work together. When one becomes more drained, then more current will be drawn from the other pack.
 
Interesting, thanks for the input. I'm surprised you needed the top balance, but I'm new to LiFe. I'm in no rush to put the LiFe battery bank into service since the FLA bank is still working OK if a bit tired. And longer days are on the way. It will take a long time to charge the new batteries given I can only do it on surplus power (2.1 KW arrays) and my power supply is 10A. When they are charged up I may reconfigure and do a top balance. My plan was to do that at the point I get a 48V inverter and reconfigure the batteries for it. Given world events, upgrading the entire system may get put off.
 
Shunt calibration is a PIA. I did it, and it is basically working, but only the 0 value ended up really precise. So at least it goes into storage mode now when charge or discharge are absent. Charge and discharge are off by small but not insignificant amounts. I'm not satisfied.

You need precise charge and load current. My digital power supply may not be working at spec and I didn't put an ammeter in line (DMM) to check, which is what I'll try next. That really isn't a lab grade device either. To simulate a load, you reverse the sense wires on the shunt. This apparently must be done before entering the "set" phase of the discharge function, or the BMS alarms when you disconnect the sense wire, and you have to shut down the BMS...and lose your other settings. It's a bit awkward to do, and those sense wires need to be kept under control or they whizz off and can go across the shunt.

Also, for a two parallel battery system like mine, I suspect it is necessary for accurate results to isolate each BMS/battery by only having the battery and BMS under test present on the busses. Not really sure about that but it may be part of the inaccuracy issue.

I'll tackle it again tomorrow to see if I can get it more accurate. I really wish Chargery had done it as they were supposed to.
 
Back
Top