2 280ah cells overcharging

[Greybeard]

Of the 16 cells I bought from Alex at basen I may have 2 bad. I balanced them all by paralleling them about 2 weeks. Then after trying to charge them paralleled for a whole weekend the next weekend I put them into 2 sets of 24v and charged them.
When I attached my bms to each I have 3.4 ish on all cells except 2. One is 3.9 and the other 4.02

What exactly does this mean?
I'm currently lowering them 2 to match the others before I hook up my invertor and run a capacity test probably next weekend.

 

[DJSmiley]

Did you top balance them? If just letting them sit parallel for 2 weeks, not much happens. Current flow is very minimal so balancing themselves would take forever, if ever at all.

if you did charge them for a weekend, what kind of power supply did you use? 16 cells with 280Ah at 70% SOC requires 1350Ah before they are full.
And only a weekend... so assuming 48 hours, requires at least a 30A power supply.

Its likely the soc is even lower, so I think they aren’t properly top balanced, and that would explain your issues

 

[Bob142]

A couple things jump out at me:

1) Unless I'm not following you correctly, it doesn't sound like you actually ever top balanced the cells.

I balanced them all by paralleling them about 2 weeks.

Do you mean you hooked them up in parallel and just let them sit there for two weeks? If so, that wouldn't have accomplished anything.

Then after trying to charge them paralleled for a whole weekend the next weekend I put them into 2 sets of 24v and charged them.

Unless you charged them in parallel to 3.65V to top balance them it's probably premature to think you have two bad cells.

2) It seems like your BMS protection settings aren't configured correctly.

When I attached my bms to each I have 3.4 ish on all cells except 2. One is 3.9 and the other 4.02

Yikes. How did you get to 3.9V and 4.02V on two cells using a BMS? I checked some of your earlier posts and you were using the Overkill BMS. I've got the 4S version of that in use and will be deploying the 8S soon. You should be able to set your Cell Over Voltage setting to trigger at 3650mV so that none of your cells can go beyond that (on the 4S version, that was the default.)

*Edit: Uh oh, I think I misinterpreted what you said you did:

the next weekend I put them into 2 sets of 24v and charged them.

So you charged them in series BEFORE connecting the BMS? Bad idea. You're lucky those two only got up to where they did. Put a load on them to get them back down and don't ever charge them in series without the BMS protecting them.

 

[Greybeard]

I'm only home on weekends so when I unboxed them I paralleled them all and started charging at 3.6v 10a. That was one weekend and they only went from 3.295 to 3.301. I changed them 1 day one weekend and 3 days the next weekend to get to that level. So the 3rd weekend I put them into 2 24v packs and paraleled them so I had a 24v 560ah pack. I charged them without the bms hooked up to 28.8 that's how they got so high.

They were paraleled for a total of 3 weeks but only charged on the weekends. I assumed that balanced them then I could top Ballance them to 24v but what I'm taking from the comments is that I now need to take them back apart and paralel them all again and charge to 3.6

 

[Bob142]

I'm only home on weekends so when I unboxed them I paralleled them all and started charging at 3.6v 10a. That was one weekend and they only went from 3.295 to 3.301. I changed them 1 day one weekend and 3 days the next weekend to get to that level. So the 3rd weekend I put them into 2 24v packs and paraleled them so I had a 24v 560ah pack. I charged them without the bms hooked up to 28.8 that's how they got so high.

They were paraleled for a total of 3 weeks but only charged on the weekends. I assumed that balanced them then I could top Ballance them to 24v but what I'm taking from the comments is that I now need to take them back apart and paralel them all again and charge to 3.6

Don't take them apart. Since you now have the BMS connected, charge it up to 29.2V. THEN break them apart and finish the top balance in parallel.

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

 

[Greybeard]

Don't take them apart. Since you now have the BMS connected, charge it up to 29.2V. THEN break them apart and finish the top balance in parallel.

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

They were fully charged to 28.8 that's 3.6x8 so I figured that's where I should charge the pack to.

 

[Bob142]

They were fully charged to 28.8 that's 3.6x8 so I figured that's where I should charge the pack to.

That's fine too. You're just using the BMS to cut off when the first cell hits 3.65V. That'll get you mostly charged and then the parallel top balance will take a lot less time.

Reading your initial post again, it seems like that if you bring the two high cells down and then just put them in parallel and top balance to 3.65V you'll be good. It may not save a lot of time to bring the two high ones down and charge as a pack first.

 

[Steve_S]

A quicker simpler FIRST TEST... Take the busbars OFF the affected cells, check them for burrs or ridges (stamped ones all have a slight ridge), if needed, file/chamfer the holes. Also ensure you wipe the busbars & cell terminals clean with rubbing alcohol, sometimes there are residues of wax or greases that can cause issues. If using noalox or similar, just a very very light dab is needed, and never on the threads, only on the faces.

 

[Greybeard]

I made my bus bars out of 1/8" copper. I filed them all flat then sanded them. But I'll check them. Alcohol isn't a bad idea I used lock tight and maybe I smudged some on them. Maybe hit the faces with an emory cloth too
I'm actually wishing I had not used the lock tight now. But that's another topic
I just Got done disassembling the first pack and putting them back to parallel. Still dropping this two cells in the other pack but I'll get it before bed. Thing they're at 3.72 right now.

 

[Steve_S]

I made my bus bars out of 1/8" copper. I filed them all flat then sanded them. But I'll check them. Alcohol isn't a bad idea I used lock tight and maybe I smudged some on them. Maybe hit the faces with an emory cloth too
I'm actually wishing I had not used the lock tight now. But that's another topic
I just Got done disassembling the first pack and putting them back to parallel. Still dropping this two cells in the other pack but I'll get it before bed. Thing they're at 3.72 right now.

Some types of loctite can cause problems with electrical conductivity, while others do not. I forget which is which, unfortunately.

 

[Greybeard]

I think all except the special electrical loctite causes problems. The thinking that the contact with the flat pad is where you're getting your electrical connection. But still I think any extra is better. Plus you can't check voltages off the end of the studs with your meter.

 

[Greybeard]

Finally got all these top balanced. All within .004 anyways. But I did have one swell up pretty good. I don't know how much of this is normal or not. I had a heck of a time getting the buss bar off it
I'm going to put a load on them tomorrow and check what they do.

 

[cinergi]

Regarding loctite ( @Steve_S & @Greybeard ), current through the stud is unimportant and unnecessary. It doesn't matter if the stud ends up being electrically isolated from the cell terminal. I know this from experience. Also (as far as testing voltage via the stud), the stud will remain energized from the TOP of the bus bar via the nut even if it has no electrical connection from inside the terminal.

That being said, I won't do loctite again for two reasons:

1. The loctite requires active metal to form a bond. Stainless steel is barely active and aluminum isn't active at all. It's a weak bond (I can remove the stud with an allen wrench) and not all of the liquid hardens (leading to #2 below)
2. Unbonded loctite will ooze/erupt out of the terminal and go underneath the bus bar, eventually causing a high resistance connection. Again, from experience.

It is ridiculously easy to have high resistance busbar connections ... I just had it happen again to me when I removed and re-added a busbar. I had to clean it up again and reinstall it a third time because it was getting warm and the BMS actually called it out as a high-resistance connection.

As for the overcharging .. 4 is definitely rough but may not be a big deal depending on how long it sat there. The electrolyte probably doesn't start breaking down until 4.3+ volts from what I've recently learned.

 

[Greybeard]

I agree that's why I won't use it again. Plus it has a habit of making a bead around the top of the threads as you screw it in and it's super easy to get a film of it over the terminal from that.

 

[Greybeard]

I hit the bottom of my invertor setting about 4:30am. That's 14 hours. 8hours of the ac in constant and another 6hr of the ac and a space heater on high. (I wanted to know how long it would run just my ac but my bedroom was too cold to sleep?,)
I had a few cells stay higher then the majority and I believe this is from resistance from the locktite as I stated before. I need to get some scotch brite and clean themBtw my bank 1 started out slightly lower then bank 2. I don't know why that's how they were after top balancing then sitting for a week paraleled.
Also the cell that swelled up so bad is cell 8 on the second screen shot

 

[Gazoo]

I had a few cells stay higher then the majority and I believe this is from resistance from the locktite as I stated before.

If you have loctite residue on the terminals it's super easy to remove. I used acetone after the studs had set for at least 24 hours to get it off. I will be rearranging my cells next week and will check to see if any more loctite oozed out when I remove the busbars.

The voltages of your screenshots look fine to me. You have parallel top balanced. So expect the difference of the voltages to be more when you are out of the lower part of the knee. When you charge and reach the upper part of the knee the difference of the voltages should be less....because you parallel top balanced.

 

[Greybeard]

It's official that cell is bad. It's the one that registered so high the first time I charged them and now I just recharged the pack. I can't get it fully charged. Once they get about 3.35 then that one swollen cell just jumps up to 3.7 and kicks out the bms. Then it drops back down. Try it again and the same.

Now let's see if Mr Alex at basen is going to stand behind his cells

 

[Greybeard]

 

[johntaves]

I charged them without the bms hooked up to 28.8 that's how they got so high.

Now let's see if Mr Alex at basen is going to stand behind his cells

I don't see why these are bad cells from Basen. Charging a series pack without a BMS is exactly how you wreck a cell or two.

Balancing the cells only squeezes more capacity out of the pack, it does not somehow relieve you of the need for a BMS. The problem is that the cell voltages go exponential at the top of the SoC, and that means even a slight difference in balance or chemistry or temp or mood, one cell will go exponential shortly before the others, and thus will spend some time getting damaged before the others catch up and trip off due to whole pack voltage.

The swelling is the cell getting physically wrecked. I suspect we want no swelling. I will be doing some pressure measurements soon (see Taves BMS thread) and I hope to learn something from that.

 

[Marinepower]

It's official that cell is bad. It's the one that registered so high the first time I charged them and now I just recharged the pack. I can't get it fully charged. Once they get about 3.35 then that one swollen cell just jumps up to 3.7 and kicks out the bms. Then it drops back down. Try it again and the same.

Now let's see if Mr Alex at basen is going to stand behind his cells

I think "Mr. Alex" is actually a Ms. Alex.