Odd discharging, not sure if this is normal or not

[buttabean]

So I've been trying to figure out the behavior of these lifepo4 batteries. 1-2-3 were purchased 3 months apart from 4-5-6. I've tried peak charging them to top the cells off but that had zero effect on the discharge. They can at times show a 12% soc variation between the top 3 and bottom 3 packs with different discharge wattages. I read that this is normal behavior but to me it seems odd that the packs that were purchased at the same time act the same?

I'm guessing it has something to do with internal resistances between the manufactured cells? I updated the firmware on all 6 packs to be the same. no luck. I even set it up to be closed loop with the sol ark 12k and they behaved the same.

Also when the packs drop below an average of 25% soc, they start to cut back on discharge rates. This is while on grid with work schedule active. Is this the sol ark doing this or the battery bms? I temporarily disabled the close loop and connected the batteries to solar assistant to get more info. Seems odd that the top 3 have pretty much stopped discharging. Could this be due to cells not being balanced? I don't normally discharge below 50% on any normal day with tou. I wanted to discharge them to see if that would have any effect on the balancing in question

picture 2 red = grid power, blue = load
picture 3 top arrow = curve which battery cutts power, bottom arrow = 25% soc

I saw will had a video on the eg4 racks and he mounted the negative to the bottom of the rack to help eliminate "current sharing problems" Could this behavior be that?

 

[rodrick]

Wiring can cause that problem also connections being loose or dirty
all wires should be the same size and length and on a rack like will’s diagonal wire connections help
are any of you connections warmer than the others that would mean resistance at that terminal

 

[buttabean]

Wiring can cause that problem also connections being loose or dirty
all wires should be the same size and length and on a rack like will’s diagonal wire connections help
are any of you connections warmer than the others that would mean resistance at that terminal

All wires are the same length. Hmm, I did use a lower gauge wire that came with the batteries to connect them to the bus bars. I don't believe I've checked the warmth of the wires. I'll keep that in mind next discharge. thx

 

[buttabean]

So i keep monitoring the battery behaviors between the two pack purchases. I don't think it's a wire issue. I decided to turned off the breakers on 2-5 packs and left 1 and 6 on. I see 4-5 resting voltage drop more than 2-3. This screenshot is after 3 hours. Would this indicate I need to top balance 4-6? Or i should lower the absorption/bulk which is currently at 54.8V

 

[buttabean]

after 6 hours

 

[DougfromdaUP]

I don't think you have a problem.

As the batteries charge and discharge their internal characteristics are changing constantly. With a nice, low impedance connection to your AIO a very small difference between batteries will make a big difference in their current draw. It looks like they start and end up at pretty much the right spot, and diverge a little in the middle. Just what one would expect. If they're all delivering around 100 Ah what more could you want?

Also, the indicated SOC may not be right on, either.

 

[buttabean]

I don't think you have a problem.

As the batteries charge and discharge their internal characteristics are changing constantly. With a nice, low impedance connection to your AIO a very small difference between batteries will make a big difference in their current draw. It looks like they start and end up at pretty much the right spot, and diverge a little in the middle. Just what one would expect. If they're all delivering around 100 Ah what more could you want?

Also, the indicated SOC may not be right on, either.

In the earlier post I was concerned with how the batteries start to reduce output at 25% but I suspect that's the sol ark. I'll have to try going off grid to see what happens.

I'm going to leave them to rest over night and see where they settle.

Thanks for the feedback. I appreciate it.

 

[DougfromdaUP]

I have a SolArk 12k and 5 EG4LifePower4 batteries. and I've been watching them for ten months. The batteries have never limited their discharge, so it must have been the inverter. The time-of-use settings can be tricky, but I think I have them figured out now.


Here's some more info about my batteries:

https://diysolarforum.com/threads/w...ting-as-its-state-of-health.55178/post-718290

 

[buttabean]

I have a SolArk 12k and 5 EG4LifePower4 batteries. and I've been watching them for ten months. The batteries have never limited their discharge, so it must have been the inverter. The time-of-use settings can be tricky, but I think I have them figured out now.


Here's some more info about my batteries:

https://diysolarforum.com/threads/w...ting-as-its-state-of-health.55178/post-718290

hmm, i'll have to give your float bulk settings a try. thank you

*edit* the two batteries i left connected to the sol ark seem to have balanced out with each other. i may try connecting only 2-5 tomorrow and run those two for a day or so and see if pack 6 drops off like 4-5 did

 

[RCinFLA]

To understand batteries, you have to have some understanding of overpotential voltage, also called polarization voltage.

It is most of the voltage slump for discharge current and bump up in cell voltage during charging and is the overhead required to drive the kinetic processes in battery. For lithium-ion it is mostly the energy to push lithium-ion migration from one electrode to the other.

Overpotential voltage slump has an exponential time decay, typically 1-3 minutes to level out at equilibrium for LFP cells. It is the difference from rested open circuit cell voltage to loaded cell voltage at given load current after 1-3 minutes. For LFP cells, you need load to be in the 0.2 to 0.4 C(A) current range to get a good number for overpotential slump. It gets greater for greater load current.

As LFP cell gets older it requires more overpotential overhead to migrate lithium-ions to create the externally demanded cell current. Matched cells have similar overpotential voltage slump for similar load current on cells.

Check the open circuit rested voltage to loaded current voltage slump delta after 1-3 minutes of load current. It is affected by temperature so make measurements within 20-30 degs C. A poor cell will have greater voltage slump. For LFP cell, the slump for given current level is fairly constant from about 25% state of charge to 85% state of charge. It gets a bit greater outside this range of SoC and much greater at colder temperatures below 10 degs C. Open circuit, rested cell voltage to perform load test should not be above 3.35v or less than 3.25v.

 

[TomC4306]

absorption/bulk which is currently at 54.8V

I would raise this to at least 55.2