Can you top balance an 8s connection with an active balancer and inverter charger?

[RossA]

I really wish I had an experienced LiFePO4 battery builder able to help me out with this, I'm very confused.

What I have:

8 304 AH EVE cells from Docanpower
JK 8s BMS
LV2424 Hybrid inverter charger
Multimeter

What I noticed when I received the cells:
They looked great, no bloating or scratches.
QR codes in tact. Looked brand new.

What I did initially.
Connected them all in series with the bus bars that came with.
Installed the Balance leads for the BMS
Connected the BMS and jump started it.
Hooked up the inverter charger to the battery and AC power.
Charged the cells at 30A
Set bulk charge to 28.8v and float to 28.6v

What I noticed during first charge:
As they charged up, slowly I noticed some slight bloating, but it was wavy. As I felt the sides of the cells, the middle was concave while the left and right side were slightly bloated.
It took several hours to charge the cells at 30A (expected this)

I charged them up this way, and my BMS was reporting about 0.16v cell differential at some points, and sometimes the BMS would cutoff charging due to over voltage.

Then the BMS would attempt to balance the cells.

once cell voltage differential dropped, it would continue charging.

Eventually all cells read the same voltage again and the inverter stopped charging.

Voltage readings from the inverter and my multimeter seemed to match up with the total battery bank voltage so I figured they were essentially top balanced and charged.

I then disconnected the inverter charger from AC power and let the batteries run the inverter.

Load according to the BMS was between 12-24 watts. It ran for 5 or 6 days as I monitored voltage.

Voltage started dropping very quickly day 4 and 5. BMS was claiming I had used about 80 AH and should be at 74% or something. But I was around 26v and every chart I read stated that I was about 30% capacity. I continued to run it to see what it would do. Eventually I dropped down to 24.8v after using roughly 100 AH (according to my BMS at least). This places it about 12% SOC while my BMS was still claiming something close to a 70% SOC.

At the end, when I turned the inverter off, the voltage increased a little bit, but still pretty low. I still notice the cell walls being a little wavy with the middle being lower than the sides despite low SOC (according to voltage charts anyways).

When I do the math, this large of a bank should be able to supply 20w continuous to run the inverter for at least double the number of days that it ran it for.

It almost seems like I'm only getting 120-130 AH out of the full system.

I can't think of anything that I may have done to hurt the cells, but that capacity has me seriously concerned that there is something very wrong with the cells.

I now have a compression box on them that is lightly compressing them in hopes to prevent too much further bloating.

I am taking a trip to attempt to finish my bus conversion this next week so I wanted a compression box for safe shipping of the cells and wanted them around 30% SOC for the trip just in case. I connected the inverter back into AC power to again charge the cells at 30A. I disconnected the AC and turned off the inverter when the overall battery voltage while charging hit 27v. It has now leveled to about 26.4v.

I noticed it only took about 1.5 hours of charging at 30A to bring the batteries up from 24.9v to 26.4v resting... Which has me concerned as well since it should have taken longer to bring it from 10% SOC to 70% SOC (all SOC's are based off of a 24v voltage chart I found on here).

After writing all of this out, I'm even more concerned about the cells than I was before... It all seems to be pointing to something being very wrong. I really wanted to avoid needing to spend an additional $100 on a variable voltage power supply for top balancing and anything to run a legitimate capacity test and thought that my method would work just fine. After all I wasn't charging any of the cells above 3.6v so they shouldn't be damaged in any way.

Can I please get some input from you guys about what I'm seeing? What are some things I can do with what I have to figure out the condition of the cells and solve these problems? I'm hoping that its just the BMS was drastically underestimating the power that the battery was supplying to the inverter to run it, but then the charging being so quick at 30A makes me still believe that the capacity is literally less than half of what it should be.

 

[sunshine_eggo]

What I did initially.
Connected them all in series with the bus bars that came with.
Installed the Balance leads for the BMS
Connected the BMS and jump started it.
Hooked up the inverter charger to the battery and AC power.
Charged the cells at 30A
Set bulk charge to 28.8v and float to 28.6v

Cool.

What I noticed during first charge:
As they charged up, slowly I noticed some slight bloating, but it was wavy. As I felt the sides of the cells, the middle was concave while the left and right side were slightly bloated.

EXPANSION is normal. BLOAT is not. Expansion is listed in the cell datasheets.

It took several hours to charge the cells at 30A (expected this)

Good.

I charged them up this way, and my BMS was reporting about 0.16v cell differential at some points, and sometimes the BMS would cutoff charging due to over voltage.

Then the BMS would attempt to balance the cells.

once cell voltage differential dropped, it would continue charging.

Eventually all cells read the same voltage again and the inverter stopped charging.

Probably good enough, BUT I would set absorption and float to 3.65V/cell and leave them there until the current drops to near 0.

Voltage readings from the inverter and my multimeter seemed to match up with the total battery bank voltage so I figured they were essentially top balanced and charged.

I then disconnected the inverter charger from AC power and let the batteries run the inverter.

Load according to the BMS was between 12-24 watts. It ran for 5 or 6 days as I monitored voltage.

Voltage started dropping very quickly day 4 and 5. BMS was claiming I had used about 80 AH and should be at 74% or something. But I was around 26v and every chart I read stated that I was about 30% capacity. I continued to run it to see what it would do. Eventually I dropped down to 24.8v after using roughly 100 AH (according to my BMS at least). This places it about 12% SOC while my BMS was still claiming something close to a 70% SOC.

Have you confirmed the BMS is measuring current properly?

Is the BMS configured for the correct capacity?

At the end, when I turned the inverter off, the voltage increased a little bit, but still pretty low. I still notice the cell walls being a little wavy with the middle being lower than the sides despite low SOC (according to voltage charts anyways).

When I do the math, this large of a bank should be able to supply 20w continuous to run the inverter for at least double the number of days that it ran it for.

It almost seems like I'm only getting 120-130 AH out of the full system.

I can't think of anything that I may have done to hurt the cells, but that capacity has me seriously concerned that there is something very wrong with the cells.

I now have a compression box on them that is lightly compressing them in hopes to prevent too much further bloating.

Compression is not necessary. It may cycle improve life, but they can operate uncompressed without issue. I tested 9 280Ah Eve cells completely uncompressed, and they were 95%+ rated capacity.

I am taking a trip to attempt to finish my bus conversion this next week so I wanted a compression box for safe shipping of the cells and wanted them around 30% SOC for the trip just in case. I connected the inverter back into AC power to again charge the cells at 30A. I disconnected the AC and turned off the inverter when the overall battery voltage while charging hit 27v. It has now leveled to about 26.4v.

I noticed it only took about 1.5 hours of charging at 30A to bring the batteries up from 24.9v to 26.4v resting... Which has me concerned as well since it should have taken longer to bring it from 10% SOC to 70% SOC (all SOC's are based off of a 24v voltage chart I found on here).

Voltage vs. SoC is pretty unreliable especially when current is flowing. Voltage vs. SoC is not linear. If you look at the charts, a very wide range of SoC is covered by a very narrow voltage range. Thus, the observed voltage increase after 45Ah input shouldn't concern you.

After writing all of this out, I'm even more concerned about the cells than I was before... It all seems to be pointing to something being very wrong. I really wanted to avoid needing to spend an additional $100 on a variable voltage power supply for top balancing and anything to run a legitimate capacity test and thought that my method would work just fine. After all I wasn't charging any of the cells above 3.6v so they shouldn't be damaged in any way.

Something is suspect. Triple check all connections and confirm nothing is loose.

Charge all cells to 3.65V and near 0A. Repeat discharge test. Confirm ALL voltage/current measurements with a separate meter.

 

[RossA]

Cool.



EXPANSION is normal. BLOAT is not. Expansion is listed in the cell datasheets.



Good.



Probably good enough, BUT I would set absorption and float to 3.65V/cell and leave them there until the current drops to near 0.



Have you confirmed the BMS is measuring current properly?

Is the BMS configured for the correct capacity?



Compression is not necessary. It may cycle improve life, but they can operate uncompressed without issue. I tested 9 280Ah Eve cells completely uncompressed, and they were 95%+ rated capacity.



Voltage vs. SoC is pretty unreliable especially when current is flowing. Voltage vs. SoC is not linear. If you look at the charts, a very wide range of SoC is covered by a very narrow voltage range. Thus, the observed voltage increase after 45Ah input shouldn't concern you.



Something is suspect. Triple check all connections and confirm nothing is loose.

Charge all cells to 3.65V and near 0A. Repeat discharge test. Confirm ALL voltage/current measurements with a separate meter.

This is all good help, thank you! What would you recommend for a good way to measure the current out of the batteries? I looked for a current clamp and they were rather expensive, I wonder if a separate shunt might be a better idea if in fact the BMS is measuring current incorrectly.

What would be an affordable shunt? I know that victrons shunt is really good, but its also a little pricy. I found a different brand shunt on Amazon for $40, but the reviews aren't great.

I will pay more attention to heat loss on bus bars or terminals. I will also charge the batteries all the way up again. They have been charging for another 2 hours and the voltage hasn't changed much so it seems that they have entered the flat part of the curve. They are currently showing about 27v while charging.

I set the battery capacity in the BMS to exactly 304 AH, but its hard to know if that is correct without knowing what the actual capacity of these cells are.

 

[sunshine_eggo]

A DC clamp ammeter is invaluable.

Will recommends these shunts:

Off-grid Solar Battery Monitors

Building a vehicle mounted solar power system? Let me help.

www.mobile-solarpower.com

 

[RCinFLA]

Under the assumption the batteries are fine, your issue is likely too much resistance in battery cabling and/or improper or in complete setup parameters in BMS.

If you have too small a gauge and/or long battery cable run you might be getting too much voltage drop. Also check all terminal connections. Check voltage drops across all connections while under the 30 amp charging current.

Set BMS to start balancing above 3.4v. Verify it is not balancing below 3.4v per cell voltage.

BMS will figure consumption based on setting you put for battery AH and its reset point of 'full' battery voltage. Make sure you have allowed charger to bring battery above this reset voltage to establish 100% 'full' reference. This establishes the starting point for the AH capacity you set up. This 100% reset voltage is normally set a little below absorb charger voltage.

With 30 amps of charging, you will get 50 to 75 millivolt voltage rise per cell from its rested no-current equilibrium voltage. With eight cells that is a net total eight cell pack over-potential voltage rise of about 0.4v to 0.6v due to 30 amps of current through cells. Give five minutes of resting no-current on cells to allow cell's voltage to reach rested equilibrium voltage.

Measure cell voltage directly on cell terminals, not on tops of bus bars to avoid any possible connector resistance issues. You can check bus bar connections by measuring mV's of drop from cell terminal to top of bus bar with significant current flowing through connection.

For example, with 30 amps of current flowing, 10 mV of voltage drop between directly on cell terminal and top of bus bar would indicate 0.010v / 30A = 0.3 milliohm resistance in bus bar to battery terminal connection. This would be reasonable maximum contact resistance. Obviously the lower the voltage drop the better. With 0.3 milliohm contact resistance, 100 amps of current would have 3 watts of heating to terminal connection which is about the maximum you want to allow.

If batteries are cold (below 10 degs C) their internal resistance rises which will also reduce the amount of capacity they can deliver.