LFP battery internal resistance

[iRondo]

I have a 2P4S battery bank built from Sinopoly cells, circa 2018. It's been great but I notice lately that the internal resistances have changed. The spec is .4 mOhm and as of right now, under light load, no charge source, they're reading .15, .53, .24, .62. The pair I'm most concerned about is cell #2 and has been as high as 1.9 mOhm under heavy load.
The voltage delta is always very small, but my concern is how high is too high for the IR?

Also, is replacement of a single pair of cells okay?

TIA

 

[sunshine_eggo]

I have a 2P4S battery bank built from Sinopoly cells, circa 2018. It's been great but I notice lately that the internal resistances have changed. The spec is .4 mOhm and as of right now, under light load, no charge source, they're reading .15, .53, .24, .62.

Measured by?

Have you broken down the battery, cleaned all contacts, etc.?

 

[iRondo]

Measured by?

Have you broken down the battery, cleaned all contacts, etc.?

As measured by the BMS (Orion Jr II). At present, I have only checked tightness of connections. I am presently trying to find a decent/reasonably priced tester before breaking down the battery. Until I get one I'm monitoring the situation.

 

[sunshine_eggo]

I have use a YR-1030 and YR-1035 on a comparative basis for several years.

 

[Marinepower]

I have 16 new 100ah cells I am currently trying to match them for resistance to make a 4p4s 12 volt battery.

I have been using a YR-1035+ , which is a great, affordable tool and seems very consistent and accurate.

To double check the IR, I have also been subjecting the cells (when at the same SOC and temperature ) to a 0.2 load and recording the voltage-sag after 6 min ( it takes a while for the voltage to settle). I have found that the 1KHz AC test is not that consistent with voltage-sag recorded when the cells are loaded to 0.2c.

My current thinking is in the voltage sag test is more relevant than the 1Khz AC IR test?

MP

 

[sunshine_eggo]

Far more valuable as a comparative tool than an absolute measurement. It's good to have because cell datasheets report IR based on that same test.

DC IR measurements are usually higher than 1kHz AC impedance measurements, AND resting voltage to loaded voltage calculations are never accurate.

Run A-B-A (start with condition A, change to condition B, return to condition A) test between two currents. Start with 0.1C, wait until voltage stabilizes, record voltage, increase to 0.2C, wait until voltage stabilizes, record voltage, decrease to 0.1C, wait until voltage stabilizes, record voltage

Now you have V1, V2 and V3.

IR1 = (V1 - V2) / 0.1C
IR2 = (V3 - V2) / 0.1C

Should be closer to the impedance value, but probably still higher.