How to balance batteries?

[BarracudaBob]

This is a recurring subject. I had similar issues when I did the full 56.2/54V settings. One battery would go into protection on one cell and get out of balanced with the others.

EG4 batteries not charging to 100%

SOC meter is now reset. I would hold it there for a day or so.

diysolarforum.com

I recently tried a setting recommended by the link below.
3.45V x 16 = 55.2V absorption/bulk
3.35V x 16 = 53.6V Float

LifePower4 manual suggest 56.2 & 54.0

Other Posts match your numbers for being conservative.
BULK/ADSORB 3.45-3.52 (3.485 Ave) = 55.8 (16 Cell) 55.2V to 56.32Vmax
FLOAT 3.35 = 53.6 (16 Cell)

SunGoldPower 10kw with EG4 LifePower4 batteries - Closed Loop

I wanted to build a software "Master BMS" which will operate between my six EG4 LifePower4 batteries and my SunGoldPower 10kw inverter. I have discovered that the "Master BMS" in closed loop does not completely control the charge profile used by the inverter. I have spent the last two weeks...

diysolarforum.com

Nothing is going into protection and they are staying within 3%. I can see doing a little maintenance every now and them bringing them up to a higher voltage. It is a safe experiment to give it a try.

 

[Gotshocked!]

This is a recurring subject. I had similar issues when I did the full 56.2/54V settings. One battery would go into protection on one cell and get out of balanced with the others.

EG4 batteries not charging to 100%

SOC meter is now reset. I would hold it there for a day or so.

diysolarforum.com

I recently tried a setting recommended by the link below.
3.45V x 16 = 55.2V absorption/bulk
3.35V x 16 = 53.6V Float

LifePower4 manual suggest 56.2 & 54.0

Other Posts match your numbers for being conservative.
BULK/ADSORB 3.45-3.52 (3.485 Ave) = 55.8 (16 Cell) 55.2V to 56.32Vmax
FLOAT 3.35 = 53.6 (16 Cell)

SunGoldPower 10kw with EG4 LifePower4 batteries - Closed Loop

I wanted to build a software "Master BMS" which will operate between my six EG4 LifePower4 batteries and my SunGoldPower 10kw inverter. I have discovered that the "Master BMS" in closed loop does not completely control the charge profile used by the inverter. I have spent the last two weeks...

diysolarforum.com

Nothing is going into protection and they are staying within 3%. I can see doing a little maintenance every now and them bringing them up to a higher voltage. It is a safe experiment to give it a try.

Thank you for the information Bob. I'm getting one of these battery dongles per page 1 of this thread and we'll see if that does the trick.
If not, I'll try what you've suggested though I don't know what inverter you have and I cannot seem to find absorption/bulk and float settings on my inverter.

 

[GXMnow]

What inverter do you have? Maybe we can help with finding the settings. Do you also have a DC charge controller?

 

[BarracudaBob]

Thank you for the information Bob. I'm getting one of these battery dongles per page 1 of this thread and we'll see if that does the trick.
If not, I'll try what you've suggested though I don't know what inverter you have and I cannot seem to find absorption/bulk and float settings on my inverter.

I did all the updates with a computer so everything I have is up to date. It did fix some issues but not the balancing.

 

[Gotshocked!]

What inverter do you have? Maybe we can help with finding the settings. Do you also have a DC charge controller?

EG4 18kpv.
I sent Sig Solar an email about this, so hopefully they'll be more specific.
No charge controller.

 

[MaikaiLife]

EG4 18kpv.
I sent Sig Solar an email about this, so hopefully they'll be more specific.
No charge controller.

There is a way to export your settings to a PDF from your 18K,

Here is an example of the settings PDF I exported from a unit I was working with. When the unit is in lithium mode for batteries, everything is configured for SOC versus voltages, so you might not be seeing the settings they're suggesting because you have your inverter in lithium mode.

I would suggest exporting your settings so it's easier to see how you have your settings set.

 

[DIYrich]

Charge to full all your batteries.
Take the one problem battery out of parallel with the other batteries. Turn off or disconnect the other batteries.
Connect the comms to the one problem battery.
Put the battery through a forced charge cycle, from the grid if you have to. Maybe drain off a little power first to get the EG4 to start a charge cycle.
Reconnect all batteries in parallel.

Hopefully that fully charges and rebalances the problem battery.

 

[LithiumCowboy]

So the batteries are in parallel. That should mean they are all at the same voltage, which means they should all be at the same state of charge, assuming all the internal cells are all still good and balanced. How is the SOC calculated and reported?

Just to correct you on the statement. With this LiFePo4 chemistry same voltage does NOT equal same SOC% for about 1%-99% SOC.

 

[Brucey]

All 6 are at 56.5v.
I can check amps when the suns out to charge at max current.

Your batteries are fine, but likely the SOC level is giving you garbage readings. If you don't have time for the nonsense with weeks and months of back and forth with support just disregard the soc and trust in your shunt for SOC, use them in lead acid mode

 

[MaikaiLife]

While I do like everything to report correctly, if you have for example 12 EG4 LifePower4 batteries, you have 1200 amps total, or 960 if you always leave a 20% floor. That said I haven't had problems with my batteries balancing, but I did install Victron Shunt to my system so that if I ever have those issues the shunt will show me the true amount of energy in the battery bank. It's not in the picture below, but right where the negative connects to the Lynx is where I inserted the shunt, you can also buy the nicer shunt from Victron called the Lynx Shunt and connect to the Lynx distributor.

This won't solve your problem of wanting to see all your batteries in perfect balance, but it will enable you to know the true SOC for your entire battery bank.

 

[BarracudaBob]

12:58
Now playing


Watch later

Add to queue

Balance Start Voltage for LiFePO4 explained. Why you should not start too early​

 

[Flyview]

Just to correct you on the statement. With this LiFePo4 chemistry same voltage does NOT equal same SOC% for about 1%-99% SOC.

I know same voltage isn't the same state of charge for 2 separate batteries, but if you connect two in parallel, wouldn't they (very slowly) equalize in state of charge to each other? If you connect battery A at 70% SOC and battery B at 50% state of charge, they would remain at the same state of charge?

 

[BarracudaBob]

I know same voltage isn't the same state of charge for 2 separate batteries, but if you connect two in parallel, wouldn't they (very slowly) equalize in state of charge to each other? If you connect battery A at 70% SOC and battery B at 50% state of charge, they would remain at the same state of charge?

It doesn't appear to be that simple with LPF batteries.

 

[Brucey]

I know same voltage isn't the same state of charge for 2 separate batteries, but if you connect two in parallel, wouldn't they (very slowly) equalize in state of charge to each other? If you connect battery A at 70% SOC and battery B at 50% state of charge, they would remain at the same state of charge?

They will want to achieve equilibrium in voltage (which may or may not reflect on the BMS indicated soc).

 

[time2roll]

I know same voltage isn't the same state of charge for 2 separate batteries, but if you connect two in parallel, wouldn't they (very slowly) equalize in state of charge to each other? If you connect battery A at 70% SOC and battery B at 50% state of charge, they would remain at the same state of charge?

Yes they would equalize somewhat. However due to the very flat voltage curve they may never reach equilibrium just sitting there connected. Same voltage but different level of charge. Need to cycle the batteries for them to get in sync.

 

[GXMnow]

Correct, LFP cells are a bit tricky to balance up.
Once they are connected in parallel at the same voltage, slow charge them together until they hit the upper voltage knee.
At that point, the pack with the higher SoC will vastly drop it's current and the lower SoC pack will then be taking most of the charge current until the true SoC catches up to the other pack. Get all the cells to the same voltage up at 3.55 volts per cell, and then the SoC will be top balanced.

 

[BarracudaBob]

Correct, LFP cells are a bit tricky to balance up.
Once they are connected in parallel at the same voltage, slow charge them together until they hit the upper voltage knee.
At that point, the pack with the higher SoC will vastly drop it's current and the lower SoC pack will then be taking most of the charge current until the true SoC catches up to the other pack. Get all the cells to the same voltage up at 3.55 volts per cell, and then the SoC will be top balanced.

Problem I am having is BMS is going into alarm and disconnecting the charging at 3.5V with other cells at 3.4V, data in previous post. I did some step charging and discharging and did get one battery to sync with the others. I have new one that is doing it now. I'll discharge it and bring it back up. Taking a lot more time and effort that I expected. I can see this being a per month maintenance issue.

 

[time2roll]

Problem I am having is BMS is going into alarm and disconnecting the charging at 3.5V with other cells at 3.4V, data in previous post. I did some step charging and discharging and did get one battery to sync with the others. I have new one that is doing it now. I'll discharge it and bring it back up. Taking a lot more time and effort that I expected. I can see this being a per month maintenance issue.

Consider reducing the charging voltage to 3.400 Vpc until all is synchronized better. If planning to charge to 3.500 Vpc the alarm should be increased to 3.600 to 3.650 Vpc. Never going to be so perfectly balanced to charge at the alarm level.

 

[GXMnow]

I can see this being a per month maintenance issue.

Once the cells become well top balanced, they should become stable and the small balance current from the BMS should be enough to keep them in check. The main problem with the initial balance is the BMS only has about 0.1 amp of balance current. So for it to truly be able to balance, the charge current can only be that 0.1 amp as well. Then the BMS can literally stop the high cell from charging and allow the other cells to catch up. Once they achieve a good balance, the slight differences in charge rate can typically be managed with the small balance current.

 

[BarracudaBob]

Consider reducing the charging voltage to 3.400 Vpc until all is synchronized better. If planning to charge to 3.500 Vpc the alarm should be increased to 3.600 to 3.650 Vpc. Never going to be so perfectly balanced to charge at the alarm level.

I was aiming for 3.45 but the imbalance would make one cell go over 3.5 while the other was a 3.3 and change. My one success story was to run it down to 80% let it sit and then take it up to 55.2 and hold it. I'll get more creative next time around. Start work week back up.

Once the cells become well top balanced, they should become stable and the small balance current from the BMS should be enough to keep them in check. The main problem with the initial balance is the BMS only has about 0.1 amp of balance current. So for it to truly be able to balance, the charge current can only be that 0.1 amp as well. Then the BMS can literally stop the high cell from charging and allow the other cells to catch up. Once they achieve a good balance, the slight differences in charge rate can typically be managed with the small balance current.

Once the cells become well top balanced, they should become stable and the small balance current from the BMS should be enough to keep them in check. The main problem with the initial balance is the BMS only has about 0.1 amp of balance current. So for it to truly be able to balance, the charge current can only be that 0.1 amp as well. Then the BMS can literally stop the high cell from charging and allow the other cells to catch up. Once they achieve a good balance, the slight differences in charge rate can typically be managed with the small balance current.

So the problem child has ben tamed. Now having fun with the two new rack batteries. I ran them down, charged them up, and still had an imbalance. So what seems to be the fastest method is go run the battery 1/2 way down with a 1kw mini-split so I don't feel like I am wasting any energy. As they get close to each other I bring it up with 10 amps but only 1 volt at a time until the current goes to zero and they balance out. I kept walking them up until I got to 56.2 and held it there until current and differential voltage went away. On the last battery and looks good so far. Mid-Level staircase method to coin a term?