battery ballooned

[Sabre36]

You connected the power supply to the battery and then set the voltage? Don't do that - set the voltage first, then connect. Always measure no load on the power supply! Also, 3.65V for days will kill the battery since it will overcharge. At low C rates, you can overcharge at low voltages; please read: https://files.ev-power.eu/inc/_auto88/_info/Doc/GWL-Power-Cell-Damage-OverCharge.pdf

We have seen folks destroy cells this way more times than I care to count.

An adjustable power supplies voltage is always set in to a 0.00A load, not when connected to the battery, and then you connect the battery. Also be sure the connection leads have minimal voltage drop.

 

[Deleted member 9967]

I am curious what did you learn? I think that's what SherylinRM was asking too.

I
WAS
NOT
ASKING.
I was replying, answering, giving help to
someone else that asked the forum group.

 

[Gazoo]

Not sure why you are telling me this. But ok.

If I interpreted this the wrong way then I apologize. It came across to me as though you were asking for an explanation.

 

[Deleted member 9967]

If I interpreted this the wrong way then I apologize. It came across to me as though you were asking for an explanation.

No I was explaining it to some one else and then another person assumed and jumped in trying to explain it to me what I had already said to someone else and then you assumed as well.
And here we are.
Apology accepted.

 

[Dzl]

From what I have read for lifepo4 it is best to not do the absorb or do a very short absorb like 5 minutes.
Sorry I don't have a quote on hand but I'll look for it.

edit: I found this info saying no absorb is better. Someone linked to a more concise article but I can't seem to find it.

Here you/they are talking about normal charging though. Not specific to a top balance / balancing in general.

My point was that I think the absorb phase is integral for a top balance. My intuition is that the slow CV tapering off helps all cells in your parallel pack achieve exactly the same SOC (which is the purpose of top balancing) A simple CC charge and disconnect, even in parallel, may not achieve that. I cant say for certain but that is my view.

(I really like that article by the way, its in my sig, I think its one of the best concise beginner friendly articles on LFP)

"There is a lot to be said for simply setting the absorb Voltage to 14.4V or 14.6V, and then just stop charging once the battery reaches that Voltage! In short, zero (or a short) absorb time. At that point your battery will be around 90% full. LiFePO4 batteries will be happier in the long run when they do not sit at 100% SOC for too long, so this practice will extend battery life. If you absolutely have to have 100% SOC in your battery then absorb will do that! Officially this is reached when the charge current drops to 5% – 10% of the Ah rating of the battery, so 5 – 10 Amp for a 100Ah battery. If you cannot stop absorb based on current, then set absorb time to about 2 hours and call it a day."

I agree with this reasoning, well I'm not knowledgeable enough to agree or disagree, it would probably be more accurate to say I believe/adhere to this reasoning. But not for an initial top balance.

 

[ArthurEld]

Here you/they are talking about normal charging though. Not specific to a top balance / balancing in general.

My point was that I think the absorb phase is integral for a top balance. My intuition is that the slow CV tapering off helps all cells in your parallel pack achieve exactly the same SOC (which is the purpose of top balancing) A simple CC charge and disconnect, even in parallel, may not achieve that. I cant say for certain but that is my view.

(I really like that article by the way, its in my sig, I think its one of the best concise beginner friendly articles on LFP)

I think the purpose of the top balance should be to get the battery balanced at the top of the charging range.
Or whatever range the BMS will balance at. Otherwise, the BMS is going to just redo the balance anyway.

 

[Dzl]

I think the purpose of the top balance should be to get the battery balanced at the top of the charging range.
Or whatever range the BMS will balance at. Otherwise, the BMS is going to just redo the balance anyway.

My understanding is the same, and I interpret the CV phase in the initial top balance as being part of achieving that objective.

(Can't say for certain if its true or not, but since all balancing guides recommend CV taper, I suspect it is.)

 

[ArthurEld]

My understanding is the same, and I interpret the CV phase in the initial top balance as being part of achieving that objective.

(Can't say for certain if its true or not, but since all balancing guides recommend CV taper, I suspect it is.)

I think there are many ways to achieve the desired result. And I think the person who started this thread wouldn't have ballooned their cell if they charged the way I plan to.

There are so many different concepts involved here and I am a newbie.
We've got -
1) Charging voltage vs cell voltage vs resting voltage.
2) CV brings the three together. Which is simple and easy for everyone to understand.
3) Highest safe resting voltage is not easy to understand and it is lower than 3.65.
4) Balancing.
There are more concepts involved besides these and they each require pages of explanation.

My goal is to make a simple capacity and internal resistance test so I can compare my cells. Most people here agree that is a waste of time but I'm still doing it.
Balancing is in the back of my mind and these concepts will be involved there too.

 

[Dzl]

I think there are many ways to achieve the desired result.

This is often a wise outlook (But not always)
Since I don't understand the finer points of the theory behind top balancing and it seems most others don't either. I choose to place my trust in the most commonly recommended best practices for top balancing from the sources I trust. I don't see a reason to diverge from it unless you understand what you are doing or want to experiment.

That said, skipping the CV portion doesn't introduce any additional risk as far as I can see, it just may not adequately balance your pack.

And I think the person who started this thread wouldn't have ballooned their cell if they charged the way I plan to.

Maybe, but this is equally true if they properly followed the normal procedure.
Whether you do CC only or CC/CV, charging to 3.65V is in spec for an EVE cell. I think the only way a cell could fail at that voltage in the short term is if it is a weak/bad cell or if the EVE cells are Grade C or maybe B cells.

But OP's cell is extremely bloated and was at 3.71V resting voltage after settling for a few hours, meaning it was probably substantially above that. Most likely the problem was an improperly set voltage as someone pointed out earlier.

My goal is to make a simple capacity and internal resistance test so I can compare my cells. Most people here agree that is a waste of time but I'm still doing it.

I like that you are pursuing this, no need to worry about what others think they are your cells, and the more data the better so long as it interests you. Some people told me I was wasting my time when brought up and began looking into cell compression, but it didn't both me, I was a dog with a bone, and look how far our opinions and knowledge have evolved over the last 6 months.

I love all the little projects and experiments and research projects people take on on this forum, we have all learned a lot from each other, so keep at it. I'm curious to learn what you learn.

Balancing is in the back of my mind and these concepts will be involved there too.

I think this explains the disconnect.
Everything I have said here about Absorb/CV is specifically in reference to top balancing. But I think you may be applying thoughts you have about normal charge/discharge and/or capacity testing to top balancing which are two very different things with two very different purposes.
I would basically agree with most everything you said in reference to normal charge/discharge and to an extent capacity testing. But not balancing.

 

[ArthurEld]

Interesting that you skipped my comment that 3.65 resting voltage is over charged.

My reasoning that my method would never balloon is because since I am not doing absorb.
When I get near 3.6V I am watching it. Things move quick at that voltage. I don't have to hang around long to get there.
And there is no leaving it unattended for hours when the cells are near full state of charge.

 

[Dzl]

Interesting that you skipped my comment that 3.65 resting voltage is over charged.

Not that interesting really . I wasn't intending to ignore anything, I just didn't have anything to say about it, and I didn't think it relevant to anything I had said. Is it relevant? The EVE spec sheet explicitly defines a standard charge as having a CV taper at 3.65V. Nothing wrong with you wanting to be more conservative than that, but charging to spec should not lead to failure unless the cell is not to spec.

My reasoning that my method would never balloon is because since I am not doing absorb.

A cell charged within spec should not balloon regardless of whether you absorb or not. But your cautious approach is reasonable for normal charging, a little lower voltage or a little shorten absorb or no absorb at all is a bit safe, we agree on that.
I just don't get why you are so hung up on the absorb here, when we have already identified a clear issue with his charging method.

When I get near 3.6V I am watching it. Things move quick at that voltage. I don't have to hang around long to get there.
And there is no leaving it unattended for hours when the cells are near full state of charge.

This is a good approach (with or without CV)

 

[ArthurEld]

I just finished bringing 16 cells from 3.5V to 3.6V since my last post. They had been charging from 3.35 since yesterday.
I didn't charge while I slept.
At 3.5V I started with 20A
When I disconnected the charging at 3.6V I was down to 2A so there was some absorb.
But the total process from 3.5V to 3.6V took about 10 minutes.

 

[ArthurEld]

I like that you are pursuing this, no need to worry about what others think they are your cells, and the more data the better so long as it interests you. Some people told me I was wasting my time when brought up and began looking into cell compression, but it didn't both me, I was a dog with a bone, and look how far our opinions and knowledge have evolved over the last 6 months.

I love all the little projects and experiments and research projects people take on on this forum, we have all learned a lot from each other, so keep at it. I'm curious to learn what you learn.

So far my testing hasn't shown any significant difference between the cells.
I enjoy learning through experiments but if I can't find anything useful for sorting these cells I will give up.

 

[Deleted member 9967]

But the total process from 3.5V to 3.6V took about 10 minutes.

Very useful to know. Also kind of scary too.

 

[Gazoo]

Just a FYI, my cells were at 3.443 volts when the Riden decided to go in to CV mode at 3.60 volts. I had it set to 3.65 volts and that is when it should have went into CV mode. Charger was set at 12 amps. This is charging 8 cells.

The amount of current that went into the cells between 3.443 volts as measured at the cell terminals when the Ridens display was showing 3.60 volts, and 3.65 volts as measured at the cell terminals and 3.65 volts on the Ridens display, was 5.59 amps as measured by the Riden. It took 4 hours in CV mode for the absorb.

I am very happy to have that part behind me as it took a long time. I started the top balance on 10-30-2020 at 7:45PM, and it finished on
11-5-2020 1:10AM.

 

[ArthurEld]

Just a FYI, my cells were at 3.443 volts when the Riden decided to go in to CV mode at 3.60 volts. I had it set to 3.65 volts and that is when it should have went into CV mode. Charger was set at 12 amps. This is charging 8 cells.

The amount of current that went into the cells between 3.443 volts as measured at the cell terminals when the Ridens display was showing 3.60 volts, and 3.65 volts as measured at the cell terminals and 3.65 volts on the Ridens display, was 5.59 amps as measured by the Riden. It took 4 hours in CV mode for the absorb.

I am very happy to have that part behind me as it took a long time. I started the top balance on 10-30-2020 at 7:45PM, and it finished on
11-5-2020 1:10AM.

What voltage are your cells now Gazoo?

 

[Gazoo]

So I finished charging Friday morning at 1:10 A.M. I had to wait on some parts to start capacity testing the pack and I started that yesterday around 6:00 P.M. During the time there was no load connected to them, other than the BMS, they all held above 3.5 volts and the biggest difference between the highest cell and the lowest cell was around 25mv. I know I should be keeping better track and I will if there is a problem when I capacity test them. I am going to post an update in the group buy thread shortly.

As of now the total pack voltage is 26.49 volts and the highest cell is 3.314 and the lowest 3.310 volts.

 

[Deleted member 9967]

So I finished charging Friday morning at 1:10 A.M. I had to wait on some parts to start capacity testing the pack and I started that yesterday around 6:00 P.M. During the time there was no load connected to them, other than the BMS, they all held above 3.5 volts and the biggest difference between the highest cell and the lowest cell was around 25mv. I know I should be keeping better track and I will if there is a problem when I capacity test them. I am going to post an update in the group buy thread shortly.

As of now the total pack voltage is 26.49 volts and the highest cell is 3.314 and the lowest 3.310 volts.

So with no load except the BMS. And after resting how long? They went from 3. to just above 3.5 and now 3.31X.
After how long?
Thanks

 

[Gazoo]

They rested from Friday morning at 1:10 A.M when I finished my top balancing until yesterday around 6:00PM when I started the capacity test. During that time the voltage stayed above 3.65 volts. I put the cells in series Sunday and wired up the BMS.. The voltage always drops quickly when a load is connected. As of now the pack voltage is 26.01 and the cells are 3.25 with a 7mv difference between the highest and the lowest voltage cell. Does this answer your question?..

 

[Deleted member 9967]

They rested from Friday morning at 1:10 A.M when I finished my top balancing until yesterday around 6:00PM when I started the capacity test. During that time the voltage stayed above 3.65 volts. I put the cells in series Sunday and wired up the BMS.. The voltage always drops quickly when a load is connected. As of now the pack voltage is 26.01 and the cells are 3.25 with a 7mv difference between the highest and the lowest voltage cell. Does this answer your question?..

It does yes thank you very much.
I did not know though that cells could stay at or above 3.65 volts once the charger was removed.
This is interesting.
Thanks again Gazoo.

 

[Gazoo]

It does yes thank you very much.
I did not know though that cells could stay at or above 3.65 volts once the charger was removed.
This is interesting.
Thanks again Gazoo.

Happy to help. I was surprised by that too. It looks like my my first capacity test will be finished tomorrow. I might do one more full cycle. The only problem is I don't know when my inverter will start sounding the low voltage alarm. The specs indicate between 19.5 and 21.5 volts. 21.5 would be good enough. After that I plan to test between the knees to see what difference I get.

I find all of this interesting with these big cells. Just as I did when I was testing my Valence batteries. My only regret was not having ordered the JunTek when I should have. I will be getting it as soon as I can. Money is tight here too.