Problems with Batteries Low Cell Voltage - happened to four different batteries.

[pulper11]

That's amazing! I actually own that exact one. Well, it's a variation of it but I can't tell the difference. This is what I own:

https://www.amazon.com/dp/B0B7739SMN/ref=dp_iou_view_item?ie=UTF8&psc=1

I've been using this to bring up the cell that goes very low up to 2.8V and then reconnecting the battery back to the PV. I take the top off, set it to 3.6V, and using the alligator clips, attach positive to the low cell and negative to the cell before it. I don't adjust amps.

Are you suggesting similarly here - going cell by cell until it reaches 3.4V each? Or is there another way? If there is, I'd appreciate some help on it as far as what the settings would be on the power supply and where to connect the leads.

Thanks so much!

 

[sunshine_eggo]

That's amazing! I actually own that exact one. Well, it's a variation of it but I can't tell the difference. This is what I own:

https://www.amazon.com/dp/B0B7739SMN/ref=dp_iou_view_item?ie=UTF8&psc=1

I've been using this to bring up the cell that goes very low up to 2.8V and then reconnecting the battery back to the PV. I take the top off, set it to 3.6V, and using the alligator clips, attach positive to the low cell and negative to the cell before it. I don't adjust amps.

Are you suggesting similarly here - going cell by cell until it reaches 3.4V each? Or is there another way? If there is, I'd appreciate some help on it as far as what the settings would be on the power supply and where to connect the leads.

Thanks so much!

You asked for a charger. I assumed you meant a whole battery charger.

The alligator clips likely introduce significant resistance, so you may not consistently get 5A at individual cell voltages. It's important to take a reading AT the cell terminals rather than rely on the power supply voltage reading as the cell will always be lower.

I'm proposing you use the 60V/5A charger to charge the whole battery to at least 55.2V and hold it there until the cells all show 3.45V.

Since you're already familiar with individual cell charging, you could charge the lowest cell in the example you provided (where I estimated cell states of charge). Apply 5A to the low cell for exactly two hours (5A * 2h = 10Ah) - this should add approximately 10% to the lowest cell and get you closer to the average. The 2 hours of single-cell charging could save days of BMS balancing.

Additionally, Once the whole battery is charged to 55.2V, it's also a very effective method to then record the cell #s and voltages, stop the charge and hit each cell with 5A to 3.6V starting with the lowest voltage cell and progressing from lowest to highest followed by another sweep back through all 16 cells. This is essentially an in-place balance. Summary:

Charge to 55.2V
Record all cell voltages
Stop charge
Start with lowest voltage cell and charge cell to 3.6V @ 5A.
Repeat 3.6V @ 5A charge on each cell from lowest to highest voltage.
Once all 16 cells have been charged to 3.6V, repeat charge for all 16 cells - should only take a few minutes per cell

If you wish a minimum of touch time, simply hold the battery at 55.2V until all cells read 3.45V.

Concerning your tone, no apology is necessary. I tend to be blunt, and I'm not being personal, so I assume responses aren't personal either. Everybody has varying levels of shit they have to deal with, and I hope yours gets sorted out in short order.

 

[pulper11]

I do want a whole battery charger. I just don't know how to do this using what I have and you suggested (the DC power charger). I meant to say I only have used it the way I described. If there is a better way of charging the whole battery (especially if I don't have to take the battery top off), that would be great.

The alligator clips is what the charger came with. I'm not sure what else to use but am open to suggestions.

I didn't even think of this before but could I use those clips (or something else) and attach them to the battery bus bar? That way I wouldn't need to take the cover off. It would be more convenient. Perhaps that's the way you're suggesting originally but I'm too lacking in knowledge to realize it.

If that's the case, what settings on the charger would you suggest for the whole battery? I get I'd charge to 55.2V until all cells read 3.45V, but I'm not sure what I need to input for V and A on the charger itself.

Thanks.

 

[sunshine_eggo]

I do want a whole battery charger. I just don't know how to do this using what I have and you suggested (the DC power charger). I meant to say I only have used it the way I described. If there is a better way of charging the whole battery (especially if I don't have to take the battery top off), that would be great.

The alligator clips is what the charger came with. I'm not sure what else to use but am open to suggestions.

This document linked here contains recommendations for fabricating leads for single cell or parallel cell charging @ 3.6V.

Top Balancing LiFePo4 Cells using a low cost benchtop power supply.

To get the paper, click on the orange at the top of this page. This is a tutorial with detailed steps on how I top-balance my LiFePO4 Cells using a low-cost benchtop power supply. To get the tutorial, click on the orange "Download" button...

diysolarforum.com

At full battery voltage, the alligator clips may be fine as the much higher voltage reduces the impact of the voltage drop.

I didn't even think of this before but could I use those clips (or something else) and attach them to the battery bus bar? That way I wouldn't need to take the cover off. It would be more convenient. Perhaps that's the way you're suggesting originally but I'm too lacking in knowledge to realize it.

Yes. That's exactly what I intended.

If that's the case, what settings on the charger would you suggest for the whole battery? I get I'd charge to 55.2V until all cells read 3.45V, but I'm not sure what I need to input for V and A on the charger itself.

With the unit disconnected from the battery, set voltage to 55.2V and crank the amps all the way to the right assuming that will get you to 5A (most don't display amperage adjustment when the current is not flowing). Connect the correct polarity to the bus bars. if the current is lower than 5A, crank it up to 5A.

If the battery is below 55.2V, the power supply will be pulled down to battery voltage and will continue to supply the set current until 55.2V is reached. it will then hold 55.2 and taper current until it gets to 0A supplying the tiniest trickle of current needed to hold at 55.2V while the BMS balances the cells.

 

[kts78]

Sorry I'm not on all the time. Yes my problem was my inverter was staying on float voltage and not charging high enough to balance cells. I had plenty of solar, but it would start limiting production saying batteries were fully charged when they were not. If I raised the bulk charge voltage it would then overcharge because it didn't have enough time to slow down solar production and batteries would kick off to protect themselves due to overvoltage. Inverter is on float probably 95% of the time and if the float voltage was set at 55v it would never balance. Dan had told me some inverters needed a higher float, so I kept raising it til I could see the balancing. I am not sure all of my setting are perfect, I need more time to see how it reacts to different situations. System is running an offgrid school in a different country I can only monitor it while I'm home.

 

[pulper11]

So I'm charging one of the batteries as described above. Got the lowest cell up to about 3.15V. Then Connected the charger to the busbar and set it to 55.2V with 5A. It charged slowly but the AH was moving up.

I've checked it every 1/2 hour or so.

About 3 hours later, I checked it and it was at 53.81V (started at 52.64), with the max cell being 3.701V (previous max about 1/2 hour previous was 3.397V), the min cell being 3.279V, and 13.9AH. I knew the max was too high so I kept watching it.

It went up to 3.752V very quickly (about one minute) and the alarm blinked red for about 30 seconds before it dropped down under 3.7V. No more alarm. The battery indicates it is at 100% now (about two minutes after being at 13.9AH), the voltages are dropping, and the charger doesn't show any amps.

I've kept it attached to the charger just in case, once the max cell drops more, it resumes charging. When I left to type this, it was at 3.626V max, 3.263 min, and 53.5V.

Any suggestions as to how to proceed would be greatly appreciated! Thank you.

 

[400bird]

It went up to 3.752V very quickly (about one minute) and the alarm blinked red for about 30 seconds before it dropped down under 3.7V. No more alarm. The battery indicates it is at 100% now (about two minutes after being at 13.9AH), the voltages are dropping, and the charger doesn't show any amps.

I've kept it attached to the charger just in case, once the max cell drops more, it resumes charging. When I left to type this, it was at 3.626V max, 3.263 min, and 53.5V.

Ouch that's wildly out of balance

Any suggestions as to how to proceed would be greatly appreciated! Thank you.

Let it keep going on this same cycle (charge until full, balance, restart charge) until all the cells are over 3.5 volts.

 

[sunshine_eggo]

Ouch that's wildly out of balance

Let it keep going on this same cycle (charge until full, balance, restart charge) until all the cells are over 3.5 volts.

^This except @ 55.2V, all cells will be at 3.45V. That's the initial target.

I've kept it attached to the charger just in case, once the max cell drops more, it resumes charging. When I left to type this, it was at 3.626V max, 3.263 min

Hopefully the 3.263 cell is the one that's 2.5V when empty.

Is the 3.626V cell the same one that's 2.5V when empty?

What are the other 14 cells reading?

 

[400bird]

^This except @ 55.2V, all cells will be at 3.45V. That's the initial target.

Pfffff, you and your "math" ?
I should have done the math, not just gone by general practice.

 

[pulper11]

I unfortunately had to turn off the charger for now. I'll turn it back on tomorrow and hopefully it will charge again.

The battery was down to 53.35V with max being 3.582V and min being 3.256V. However, there wasn't any charging going on still. I think it was about 15 minutes before I turned it off that I wrote down the cell numbers:

1. 3347
2. 3314
3. 3308
4. 3351
5. 3344
6. 3263
7. 3633
8. 3350
9. 3356
10. 3350
11. 3326
12. 3314
13. 3317
14. 3314
15. 3312
16. 3314

This battery is the first battery listed in post #3 (where the individual cells were typed out).

Unfortunately (based on your post) the cell that goes low on this one is cell 6. Cell 7 is the cell that went really high this time.

 

[400bird]

I unfortunately had to turn off the charger for now. I'll turn it back on tomorrow and hopefully it will charge again.

The battery was down to 53.35V with max being 3.582V and min being 3.256V. However, there wasn't any charging going on still. I think it was about 15 minutes before I turned it off that I wrote down the cell numbers:

1. 3347
2. 3314
3. 3308
4. 3351
5. 3344
6. 3263
7. 3633
8. 3350
9. 3356
10. 3350
11. 3326
12. 3314
13. 3317
14. 3314
15. 3312
16. 3314

Because it's just that one high cell, I'd recommend a slightly different direction. Measure the cells with a volt meter to verify the BMS readings, then discharge that single cell.
If you have something to use for a resistive load, I'd use that. Otherwise, you can buy single cell dischargers on Amazon or eBay.

This battery is the first battery listed in post #3 (where the individual cells were typed out).

Unfortunately (based on your post) the cell that goes low on this one is cell 6. Cell 7 is the cell that went really high this time.

This is really good news. If it was the same cell that was both lowest and low SOC and highest at high SOC, that would mean that cell has less capacity.
This is just a terribly out of balance pack.

 

[pulper11]

I just did some quick research. Very quick.
I came across this post:

One cell level too high - LiFePO4

According to my BMS, I have one cell of of an 8S LiFePO4 battery that is significantly higher then the other seven (3.6 vs 3.35V at full charge). Other than doing a discharge then top balance all 8 cells, is there another method that would get this cell back in line? Would a discharge of just...

diysolarforum.com

would using a regular light bulb do this?

I also know I have some resistors but I think they're pretty small for tinkering with electronics.

Would you suggest bringing the volts down on the high cell below the other cells before charging again?

Thanks for your help!

 

[sunshine_eggo]

I unfortunately had to turn off the charger for now. I'll turn it back on tomorrow and hopefully it will charge again.

The battery was down to 53.35V with max being 3.582V and min being 3.256V. However, there wasn't any charging going on still. I think it was about 15 minutes before I turned it off that I wrote down the cell numbers:

6. 3263
7. 3633

This battery is the first battery listed in post #3 (where the individual cells were typed out).

Unfortunately (based on your post) the cell that goes low on this one is cell 6. Cell 7 is the cell that went really high this time.

If the same cell his low and high first with clearly reduced capacity, it would be a bad cell for certain.

would using a regular light bulb do this?

Only a 12V one.

Would you suggest bringing the volts down on the high cell below the other cells before charging again?

No. Give the battery a chance to do some balancing. It already looks better.

 

[sunsurfer]

Perhaps I over-purchased on the batteries

No such thing as to much battery.....

 

[pulper11]

No. Give the battery a chance to do some balancing. It already looks better.

I actually forgot that I own the item below from back when I was doing drone building:

https://www.amazon.com/gp/product/B00L5YBJI2/ref=ppx_yo_dt_b_search_asin_title?ie=UTF8&psc=1

I could use this if I can find it. If I can, maybe bring the high voltage cell down to around 3315V (around the other cells) or just try charging again?

Thanks again,
Paul

 

[sunshine_eggo]

Limited to 5W or 10W, and you'll have voltage drop too.

No. Give the battery a chance to do some balancing. It already looks better.

Just try charging.

If anything, use that to boost the low cell while you charge the whole pack.

 

[pulper11]

I just tried charging. It worked for two minutes. However, the max cell went from 3523MV to 3750V in that time. Then the charger shut off.

I can access the BMS of the battery. Is there any settings that I should check? On the BMS, I only changed settings previously that Dan had advised. He asked me to change the starting voltage where balancing begins from 3400V to 3200V. Unfortunately (sorry about this) I had forgotten about this when I originally posted this thread. Not sure if this is a big deal or not.

 

[sunshine_eggo]

I just tried charging. It worked for two minutes. However, the max cell went from 3523MV to 3750V in that time. Then the charger shut off.

Assume you mean the power supply set to 55.2V/5A? This is the expected behavior, and you should just let it run. A snapshot of all cells at cut-off is meaningful for comparison to evaluate the progress of balancing. This is likely to be a days long exercise.

I can access the BMS of the battery. Is there any settings that I should check? On the BMS, I only changed settings previously that Dan had advised. He asked me to change the starting voltage where balancing begins from 3400V to 3200V. Unfortunately (sorry about this) I had forgotten about this when I originally posted this thread. Not sure if this is a big deal or not.

It's mostly fine. I don't think Dan understands how ineffective it is at lower voltages, so he does that. I would recommend you set it to no LOWER than 3380 to ensure the BMS focuses on only the highest cells. Also, there should be a deviation value. It's typically 20mV, and that's fine.

Edit: fixed typo. originally said higher.

 

[400bird]

I would recommend you set it to no higher than 3380 to ensure the BMS focuses on only the highest cells. Also, there should be a deviation value. It's typically 20mV, and that's fine.

Seconded! Maybe even 3400

3.2 volts is too low for balancing. The cell voltage is nearly flat between 3.2 and 3.4 volts which is something like 20%-80% SOC

 

[pulper11]

I've had a few problems when following the suggestions here so hopefully someone can help.

The battery that I've been working on most is not connected to solar but only charging through the DC charger linked in posts above. That's the battery mentioned in the most recent few posts.

As previously mentioned, when charging at 55.2V and 5A, the one cell in the battery will go up to 3.75V and then the charger will shut off. The cell voltage will start coming down. However, it comes down EXTREMELY slowly. For example, after about 6 hours, it will be down to about 3.55V. And, while the DC charger remains connected, it will not resume charging. It will only resume charging if I reboot the battery. Then the cycle continues. However, the other cells are not getting higher in voltage over this time. Their voltages go up when using the DC charger, but by the time I have to reboot the battery, their voltages are actually about 2MV below where they started before the previous cycle of charge/discharge/reboot of the battery.

In the meantime, I also have two other batteries connected to solar through the LV6548. They've had access to a couple days of solid sunlight.

One of those (which has 100% SOH) has gotten back to 100% SOC. It will stay at 100% for sometime while the largest cell gets above 3.7V. Then the LV6548 will show no PV input (even though it is sunny - so obviously no longer charging) and the battery voltage in that max cell goes down much more rapidly compared to the other battery mentioned above. For example, it will decrease to 3.55V in about two minutes (as opposed to about 6 hours in the other battery). However, it will not resume charging (at least for that day) and the overall voltage of the battery stays at approx 54.3V. Please note that I had changed the float to 55.2V and the bulk to 57.6 for the LV6548 as suggested.

The other battery attached to the solar is one of the 80% SOH batteries. It actually went up to 56.67V when the max cell was at 3739MV (smallest cell was at 3441MV). It dropped down to 53.46V and 3365MV (max)/3336MV (min) after the OV alarm kicked in.

So my questions are:
What should I do about the first battery mentioned above? Do I need to look into finding a way to discharge that specific cell directly?
Any BMS settings I can change so that the first battery discharges faster or resumes charging while the max cell is still at a higher voltage (but not over voltage)?
Any settings/etc that I should change for the second battery mentioned above so that it pushes the voltage higher for the cells? And also so that it resumes charging at a higher voltage?
Considering how screwed up the BMS must have been based on how I've handled the batteries, is it possible that the 80% SOH on the second battery is actually not accurate and that it will go up to 100% Sorry if this is a stupid question .

If you need more data or settings info, please let me know and I'll see what I have written down from the past.

Thanks so much!