Chargery BMS8T Observations

[Nigel Snook]

I have just assembled and tested 2 off 280Ah LiFePO4 batteries each with a BMS8T (V4.04), 300A DCC (Discharge) and 100A DCC (Charge) configured as separate port on the positive rails with ISO board in line. I am powering from the battery directly (no Schottky diode). I would appreciate some thoughts on my observations, particularly on the differences between 2 identical set ups:

1. At switch on, Battery A BMS reads between 0.5 and 0.7A discharge. At first I assumed this was the draw from the BMS, however spec is 133mA. Batt B however is a rock solid 0A.

2. Current Monitor - when charging, current readings spot on in both. Discharge however awful - at circa 100A, current reading between 60 - 160.

3. When discharging individual cell voltages on Battery A, the sample to sample (2Hz ish) cell delta varies significantly e.g. 5mv - 70mv with extreme cells changing each time. I am convinced that this is not real and each cell is discharging proportionately. Battery B is far steadier.

4. The 300A DCC voltage drop spec is 80mV @ 100A. I am seeing 170mV on both.

In summary - Battery B seems a little better than A. I have remade sample cables and there are no hot spots anywhere so pretty confident that all well with connections. Both batteries are assembled the same although they are a mirror image of each other for installation reasons.

Has anyone experienced anything similar or have any insight to offer?

 

[Cal]

You're probably using an inverter to get a 100A discharge. Inverters have a huge ripple current. Chargery fails to filter this ripple. Hence you're seeing swings of 60 - 160A. Same issue measuring 300A DCC voltage drop. Is the 170 mV drop constant or does it vary with current readings?

Not clear what you mean by "I am powering from the battery directly (no Schottky diode)."

 

[Bob B]

If one seems to be performing better than the other .... switch them and see what happens.

 

[Nigel Snook]

You're probably using an inverter to get a 100A discharge. Inverters have a huge ripple current. Chargery fails to filter this ripple. Hence you're seeing swings of 60 - 160A. Same issue measuring 300A DCC voltage drop. Is the 170 mV drop constant or does it vary with current readings?

Not clear what you mean by "I am powering from the battery directly (no Schottky diode)."

Thanks Cal, you called it spot on - I guess that is somewhat reassuring......

I'm afraid I haven't made note of the voltage drop of DCC at other currents.

Some recommend powering the BMS via the external power in (via Schottky) rather than the cell sense cables - I believe even chargery suggest this.

 

[Nigel Snook]

If one seems to be performing better than the other .... switch them and see what happens.

I was thinking of doing that but kind of ruled it out as I have taken each battery through a full charge and discharge cycle so SOC pretty accurate. I have subsequently let one discharge lower than the other. Charger and inverter reinstalled on boat so not easy to play any longer in shed. Short answer - didn't want to mess with SOC calculation.

 

[Bob B]

I was thinking of doing that but kind of ruled it out as I have taken each battery through a full charge and discharge cycle so SOC pretty accurate. I have subsequently let one discharge lower than the other. Charger and inverter reinstalled on boat so not easy to play any longer in shed. Short answer - didn't want to mess with SOC calculation.

I definitely understand where you are coming from with that .... curious what software revision you have?

 

[Nigel Snook]

I definitely understand where you are coming from with that .... curious what software revision you have?

V4.04 - Unlike some, SOC does seem to be accurately stored by both my units.

 

[Bob B]

It does work pretty good with the fixes in the 4.04 version.

It is supposed to keep the SOC if powered down with the fixes in 4.03 .... but I have still had problems with that not working.

You could probably fully charge both packs and then reverse the BMS and still be OK .... but I won't guarantee that.

 

[grizzzman]

"I am powering from the battery directly (no Schottky diode" Though the BMS leads or from the 12 volt port?
My experience with the BMS8T was that cell #1 voltage was off by 30mv (low) and cell #4 10mv (low) this caused the cells to over charge. Powering externally solved the voltages to within a few mv. Unfortunately a few days later the ARM possessor in the BMS8T turned into a heater. After that l decided to go a different direction. Good luck with your project.