Charged to 3.9v.. how to discharge

[beowulf]

Hey all, somehow my little crappy 10a/30v variable bench supply got rotated to 3.9v instead of 3.5 that I was set on. So now I have my LEV60 3.5v max charged up to 3.92v. I took charger off its down to 3.88 so far.. but is there a way to discharge it faster, safely? I don't have any sort of 3v or so device.. 12v amazon sticks, USB chargers, etc.. all of them are like 12v or more. I don't even have a resistor or led laying around that I could use. I was trying to find a 2 cell light but even that might be too low at 3v and 4.5v might be too high.

I read that a typicaly LifePo4 can go to a 4.2v max. I am a little confused now why is 3.65 considered 100% if it can go all the way up to 4.2? But since my LEV60 shows a max of 3.5.. I am assuming it has a max of about 4v. Hopefully that it's discharging a bit it wont be a problem. It was packed in with 16 cells so it didnt have room to expand. It doesnt feel warm at all either.

Do I need to discharge it or will it go back down in a day or so to 3.5 max?

Also.. as I am charging 16 of these to 3.5 (most have gone to 3.52 and lowered down to 3.48 to 3.5 now).. is this cell going to now be damaged and not work well with the rest? Do I need to replace it?

This is in the Jehu 16 cell 48v pack, I am trying to top balance all the cells right now before I put the pack together.

 

[sunshine_eggo]

I would personally discharge it.

12V halogen/incandescent light bulb

4.2V is where the electrolyte starts to break down.

You get about 10% more capacity between 3.65 and 4.20V, but the cycle life is notably worse @ 4.20V.

 

[beowulf]

OK.. friend who has built a few batteries said to connect a lower voltage cell (same cell) in parallel to help discharge higher to lower. Had another one at 3.2v so figured 3.8 to 3.2 will equalize to about 3.5 each. I'll keep an eye on it make sure it stays around 3.5. Goal is to top charge all these so I can put together the 16cell battery finally and get the BMS and inverter/charger to kick in.

 

[sunshine_eggo]

OK.. friend who has built a few batteries said to connect a lower voltage cell (same cell) in parallel to help discharge higher to lower. Had another one at 3.2v so figured 3.8 to 3.2 will equalize to about 3.5 each. I'll keep an eye on it make sure it stays around 3.5. Goal is to top charge all these so I can put together the 16cell battery finally and get the BMS and inverter/charger to kick in.

The issue with that approach is:

1) it's very possible that you may damage both cells due to a very large surge of current between them.
2) Cell voltages don't average like that.
3) They will both end up under 3.5V, and you will need to charge them to 3.5V again.

 

[beowulf]

Hmm.. so I tried that approach.. I put the battery that was at 3.8v to the 3.2v in parallel. I checked it and what I noticed is instantly the 3.8v was dropping down in voltage. The 3.2v didnt go up much. I then took it off last night before going to bed.. it only was connected for about 20 minutes. When I got up this morning, checked again, 3.8v and 3.25v. This time I connected the two directly.. last night I had used a make shift switch so I could have it off when connected, then flip it on. So today I connect them directly not seeing any spark in doing so. A few minutes later the 3.8v is down to 3.5v, and the 3.2v is about 3.3v. That was connected for about an hour. I came back, checked.. the 3.8v was now down to 3.4v and the 3.2v is still around 3.3. So.. I am a bit baffled where did the 3.8 down to 3.4 go if the 3.2 is only up to 3.3?

I have since let them sit for a few hours, just checked and both batteries are now at 3.3v.

So.. I am a bit confused. Did the 3.8 end up being a bad cell because it quickly dropped to 3.5 and now sits at 3.3 while the 3.2 may still be viable because it only went up? I had read a few times, saw Prowse video, etc about how the batteries will equalize in parallel where the higher charged one will "flow over" in to the lower one(s) in parallel and they will all equalize out. I am confused how I unplugged it at 3.48v and its now at 3.3v.

Not sure if this is now a bad cell, or should I charge it back up to 3.5 to match the rest and then assemble the battery?

 

[MisterSandals]

So.. I am a bit baffled where did the 3.8 down to 3.4 go if the 3.2 is only up to 3.3?

The voltage vs SoC curve is not linear.
Using rough numbers only to illustrate the concept...

Lets say you have a 100Ah battery at 3.2V. It may require 40Ah of charge to take it to 3.3V. This is in the middle of the flat charge curve.
At the extremes of SoC, below 3.0V and above 3.60V, it takes far fewer Ah of charge to change the voltage .1V. It might be just 10Ah (i picked 10Ah to match your numbers) to change the voltage .1V.

 

[sunshine_eggo]

So.. I am a bit confused. Did the 3.8 end up being a bad cell because it quickly dropped to 3.5 and now sits at 3.3 while the 3.2 may still be viable because it only went up? I had read a few times, saw Prowse video, etc about how the batteries will equalize in parallel where the higher charged one will "flow over" in to the lower one(s) in parallel and they will all equalize out. I am confused how I unplugged it at 3.48v and its now at 3.3v.

2) Cell voltages don't average like that.

Hopefully, you figure it out via this plot:



3.2V is a very low state of charge.

3.8V is a very high state of charge.

when you take a high state of charge and parallel it with a low state of charge you get two mid-level states of charge...

so,

3) They will both end up under 3.5V, and you will need to charge them to 3.5V again.

Do you have me on ignore?

 

[beowulf]

Hi.. not sure why you asked on ignore? I don't fully grok all this so I apologize if you explain something and maybe I ask again.. not trying to be an ass.
I understand a little bit of the graph. What I dont understand is how my 3.8v went to 3.5v in minutes while the 3.2v didnt move. How did it drop .3 v in a few min while the other battery didnt budge. My guess is the graph as you show.. so 3.8v is high and drops fast but at 3.2 it wont rise very fast thus why I see 3.8 drop fast and 3.2 not move up fast? Is that what I am learning here?

So if that ist he case, then the 3.8 now down at 3.3.. that's OK and battery should still be good? It was only at the 3.9 to 3.8 for an hour or so, then 3.8 left over night till morning when I tried again to put in parallel with 3.2 cell, in which case it dropped fast to 3.5. So I am not sure if I should consider the battery bad now that it went over its 3.5 max (the LEV60 claim 3.5 max, not 3.65 like more LFP batteries state). It did NOT swell or get hot though, so I am assuming it is OK since it didnt swell or get hot.

Sorry again, still learning all this. Taking me a little longer guess I am not as smart as I thought I was with electronics.

 

[sunshine_eggo]

Hi.. not sure why you asked on ignore? I don't fully grok all this so I apologize if you explain something and maybe I ask again.. not trying to be an ass.
I understand a little bit of the graph. What I dont understand is how my 3.8v went to 3.5v in minutes while the 3.2v didnt move. How did it drop .3 v in a few min while the other battery didnt budge. My guess is the graph as you show.. so 3.8v is high and drops fast but at 3.2 it wont rise very fast thus why I see 3.8 drop fast and 3.2 not move up fast? Is that what I am learning here?

it's a zero sum game. If you pull 10% off the high, you add 10% to the low.

So if that ist he case, then the 3.8 now down at 3.3.. that's OK and battery should still be good? It was only at the 3.9 to 3.8 for an hour or so, then 3.8 left over night till morning when I tried again to put in parallel with 3.2 cell, in which case it dropped fast to 3.5. So I am not sure if I should consider the battery bad now that it went over its 3.5 max (the LEV60 claim 3.5 max, not 3.65 like more LFP batteries state). It did NOT swell or get hot though, so I am assuming it is OK since it didnt swell or get hot.

Sorry again, still learning all this. Taking me a little longer guess I am not as smart as I thought I was with electronics.

 

[beowulf]

Got it. So from 3.2 to about 3.55 or so its "slow". But once it nears that end curve it goes fast both charging and discharging. So charging I just have to keep testing the battery until I see it hit about 3.45 and then keep checking quick.

One thing I havent figured out. I charge to 3.5. Then take charger off and it seems to quickly settle to about 3.45 to 3.48. If I have to charge 16 cells to about 3.5 (individually right now).. by the time I get to the 3rd or 4th, the 1/2/3 or so are already settling to about .03 to .05v less. Is it going to be a big deal when I put all 16 cells together and they vary from 3.45 to 3.51 or so.. or will the BMS properly handle balancing them at that point once I put them on the inverter/charger.

My concern is that will the range difference affect the durability/longevity of some cells over others since the voltage cutoff would likely hit on the cell that is 3.5 and the ones at 3.45 wont charge any more since the charger stops?

 

[sunshine_eggo]

When you remove the current the voltage settles. This is normal. When you charge your lead-acid batteries to 14.4V, they don't stay there. They drop below 13V overnight. Same concept.

Don't confuse voltage with state of charge. Once you have hit 3.5V @ 0.05C (5% of capacity), you are fully charged. Just because the cells loses a little voltage doesn't mean it's losing a significant amount of charge.

Since you're doing this individually, just prior to build, you'll likely want to hit each cell a second time to 3.5V after you've finished the last. Shouldn't take more than 5-10 minutes EA

 

[beowulf]

Alright.. so I am charging the 3.3 (the one that was 3.8) back up to 3.5, and the other (3.2) up to 3.5). I think the rest are around 3.4 to 3.5. So I'll get the battery parts set up, then hit each right before I put the PCB on and bolt it down and connect it. Thanks for your help.

 

[MisterSandals]

They drop below 13V overnight.

14V?

 

[sunshine_eggo]

14V?

Lead acid will typically drop below 13V after sitting overnight (10+ hours).