whats inside a123 20ah cells

[whatever]

I thought it might be useful to repost some pics in a new thread of the internals of an a123 20ah cell which I opened up.
( the pictures I posted were in this thread http://endless-sphere.com/forums/viewtopic.php?f=14&t=38761&hilit=a123+20ah&start=900)
Some info on the pack the cell came from I posted here some time back
http://endless-sphere.com/forums/viewtopic.php?f=14&t=41589







View attachment 4


note should read 25 copper sheets

 

[whatever]

some info on the cell
korean made a123 20ah source via taobao
min. voltage 2.4v
max amp draw 20amp
This cell was the second cell to fail, it was on the top of 16cells lying flat, there was no compression clamping of the cells,
the top cell of the pack had no pressure on it. I suspect the reason for failure was lack of compression, it happened twice so the top cell of the pack with no pressure on it seems to be an issue.
So can conclude pressure on the cells is important.

The cell also had some 'bubbling' of the aluminium pouch along a seam, the aluminium pouch has a thin layer of plastic on each side,
it appears that a breach in the thin coating has allowed electroyte to react with the aluminium pouch layer, a study on this phenomenon here seems to support this
http://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/20120000040_2011025423.pdf
So can conclude that bending the perimenter outlying plastic around the cell ( where its sealed around the edges) is not a good idea as it might breach the very thin plastic layer that protects the aluminium from electrolyte.

The plastic membrane holding the electrolyte between cells appeared in good condition without tears or holes.
The copper layers graphite coating appears to have failed in many places, coming away from the copper, suspect lack
of compression might had played a role in that.
The lithium iron phosphate layer on the aluminium plates was in a1 condition.

The copper plates were not in electrical contact with the outer pouch , protected with strips of kapton tape.
The aluminium plates were in electrical contact with the outer pouch aluminium coating, no kapton tape.

If you have a look at this thread
http://endless-sphere.com/forums/viewtopic.php?f=14&t=41589
I point out that short cut tabs where the plastic pouch is scraped off ( to make a bit more tab useable for connecting cells), can short out the aluminium pouch with the copper tabs. After opening up the cell I can see the reason for this is no kapton used on the aluminium tabs to isolate from the
outer pouch. The amount of current that shorted was small but enough to unbalance a pack, using sealastic was sufficient
to stop any shorting occuring, a thin smear of silastic along the copper tab where the pouch had been scraped off the tab
electrically isolates the pouch and copper tab.
The short cut tab cells with some outer pouch scraped off are still very useable if silicon used to stop shorting problems.

 

[whatever]

in the cell
25 sheets of copper (graphite deposit on surface), 24 sheets of aluminium( lithium iron phosphate deposit on surface)

copper sheet size = 15cm by 19.5cm
aluminium sheet size = 15 by 19.5 cm
surface area of on sheet copper in metres squared
0.15*0.195=0.02925 m2
25 lots of copper total area = 0.02925*25=0.73125 sq metres of copper in one cell
24 lots of copper total area = 0.02925*24=0.702 sq metres of aluminium in one cell

For a 16 cell pack just for fun
The pack would contain
16*0.73125=11.7 sq metres of copper sheet( approx 12sqmetres)
16*.702=11.232 sq metres of aluminium sheet ( approx 11 sqmetres)
So if your carrying a 16s pack ( 48v nom) of a123 20ah cells on a bike
you are carrying the same as a :
1. copper sheet 1m wide by 12m long
2. aluminium sheet 1m wide by 11m long
If your pack was just one long sheet of copper and aluminium separated by a membrane it would be a very thin, but a very long battery.
Thats quite alot of copper and aluminium.

 

[wb9k]

some info on the cell
korean made a123 20ah source via taobao
min. voltage 2.4v
max amp draw 20amp
This cell was the second cell to fail, it was on the top of 16cells lying flat, there was no compression clamping of the cells,
the top cell of the pack had no pressure on it. I suspect the reason for failure was lack of compression, it happened twice so the top cell of the pack with no pressure on it seems to be an issue.
So can conclude pressure on the cells is important.

The cell also had some 'bubbling' of the aluminium pouch along a seam, the aluminium pouch has a thin layer of plastic on each side,
it appears that a breach in the thin coating has allowed electroyte to react with the aluminium pouch layer, a study on this phenomenon here seems to support this
http://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/20120000040_2011025423.pdf
So can conclude that bending the perimenter outlying plastic around the cell ( where its sealed around the edges) is not a good idea as it might breach the very thin plastic layer that protects the aluminium from electrolyte.

The plastic membrane holding the electrolyte between cells appeared in good condition without tears or holes.
The copper layers graphite coating appears to have failed in many places, coming away from the copper, suspect lack
of compression might had played a role in that.
The lithium iron phosphate layer on the aluminium plates was in a1 condition.

The copper plates were not in electrical contact with the outer pouch , protected with strips of kapton tape.
The aluminium plates were in electrical contact with the outer pouch aluminium coating, no kapton tape.

If you have a look at this thread
http://endless-sphere.com/forums/viewtopic.php?f=14&t=41589
I point out that short cut tabs where the plastic pouch is scraped off ( to make a bit more tab useable for connecting cells), can short out the aluminium pouch with the copper tabs. After opening up the cell I can see the reason for this is no kapton used on the aluminium tabs to isolate from the
outer pouch. The amount of current that shorted was small but enough to unbalance a pack, using sealastic was sufficient
to stop any shorting occuring, a thin smear of silastic along the copper tab where the pouch had been scraped off the tab
electrically isolates the pouch and copper tab.
The short cut tab cells with some outer pouch scraped off are still very useable if silicon used to stop shorting problems.

Sorry, I have to interject here. You are way out of your depth with the statements you are making along these lines. The corrosion you saw on the outside of the cell all but proves there was some electrical contact between the anode and pouch. Just because you didn't find it doesn't mean it wasn't there. I can tell by your verbiage that you have no idea what the most common failure mode that causes this is....and no, I'm not going to tell you what that is.

It sounds like these had reworked short tabs? If so, this is very likely the source of isolation trouble. DO NOT, I repeat, DO NOT ever use a pouch cell that has had tabs reworked interior to the pouch. This is one of the most irresponsible tricks ever perpetrated on this community by purveyors of stolen material that was supposed to have been destroyed beyond usability. How in the world do you know that there was no breach here? Do you know how to test isolation between the electrode stack and pouch PRIOR to teardown? I'm thinking no, since I wrote the procedure and have never shared it here. Turns out this is the ONLY way to know for sure if there is/was a fault, and even then the cell may have to be under compression to show the fault---and even then there are degrees of severity. Simply put, you are just not qualified to judge the state of isolation that existed on this cell at any time as near as I can tell.

Lack of compression was one major factor in your failure. Where this the only problem, the cell would still have degraded at an accelerated rate but what really killed this cell so quickly was overdischarge. The dissolved copper visible coupled with the flaking powder off the anode tells this story quite clearly. Had you been monitoring voltages and balancing regularly as you should have, this cell would probably have gone on for several more years before you would have noticed anything at all--except that is also had an isolation fault that led to breakdown of the aluminum pouch. So you have THREE problems here--all within your control from the sound of it--that tend to compound each other. The isolation fault could have been from reworked tabs--this will kill the cell in a certain amount of time no matter what you do with it at that point. Mishandling/folding a cell could also have done this, but it is far less likely. Along with rot of the bag come loss of capacity because Li is plating onto the Al, making it unusable for energy transfer. The lack of compression led to elevated Peukert loss on the failed cell (which was already suffering capacity loss because of the isolation fault). This fostered an ever-worsening imbalance in the pack that you were unaware of because you used no BMS and neglected to check voltages yourself on a regular enough basis to see the problem until it was too late. Please feel free to cut open an photograph all the cells you like, but don't pretend that you have a good understanding of what you're looking at here, because you just don't.

 

[Punx0r]

I think the OP's attempt to dissect and analyse the failed cell was well-meaning and about as much as any of us non-experts could attempt.

 

[wb9k]

I think the OP's attempt to dissect and analyse the failed cell was well-meaning and about as much as any of us non-experts could attempt.

I think you're probably right. My only concern is that some very dubious statements are being made as though they are fact. The OP cannot say there was no isolation fault in this cell. That is blind speculation that stands in direct contradiction to the main externally visible problem (aluminum decrepitation caused by an isolation fault leading to spotting and leaking electrolyte). [Please note that a certain amount of speculation is required on my part as I have had to read between the lines of the OP and new information continues to come to me regarding the construction and history of this pack. No pictures are provided that can prove that the aluminum was embrittled and not, say, punctured from the outside, so I don't even know for sure if the cell could have been accidentally wounded six months ago by a two-year-old. I also have to trust that his numbers are good to begin with, which always contains a certain amount of risk. The truth is however, that virtually all failure analysis on fielded units works like this. Information from the field is always suspect--even from "technical" people (some OEM's are better than others)--but it MIGHT be very useful. The difference when I get the cell back is that I know how to collect all kinds of data from the cell before I even think about touching it with a knife. All of this information has to be used to correlate against any teardown results. Several details in the photos essentially prove he doesn't know how to find that type of defect except in perhaps the most glaringly obvious case---and internal faults in cells are rarely glaringly obvious.] I already made a more brief rebuttal to this claim (of no isolation fault in the cell) in another thread, yet the claim has been reinforced here again. So I have posted a more detailed, and slightly grumpier rebuttal to this claim. I hope this makes it clearer where I'm coming from.

OTOH, I think the poster may have made a good decision to carry the teardown discussion to a new thread. It was a detour for the thread it originally appeared in. So I appreciate that. I don't seek to excoriate anyone or be any kind of flame lord. I do seek to be crystal clear in a medium that can be easy to misread on a subject about which people are easily confused. What the users of this forum seek to do is not easy, and is completely beyond the grasp of a good many people out there. When misinformation is presented as fact, I seek to correct it, in as clear a manner as possible. Further questions are always welcome, but a restatement as a valid finding of what has been pointed out as unsubstantiated, is not. That is my only gripe here. Sorry if my response seems unnecessarily heavy-handed.

 

[wb9k]

I should have added that I think it's fair for me to have pointed out that all of the problems that conspired against the user were all within the control of the user...and out of control of the cell maker. The only exception is if there was indeed an internal defect in the cell. However, it sounds as though these cells were reworked trash with tab repair done within the seal area of the pouch. From a mfg's standpoint, this automatically would void any type of warranty--and is by far the most likely source of the isolation problem. Worse than this, the practice of using cells reworked like this is dangerous and irresponsible. If I'm correct, chances are his whole pack is made of cells like this. He should feel lucky this is all that happened. So, I guess I do have more than one little gripe here.

Hobbyists and experimenters must take control of that for which they are responsible. Haphazard design, trash materials, and general neglect will usually result in disappointment, especially when you have all three together at once.

 

[Punx0r]

I already made a more brief rebuttal to this claim (of no isolation fault in the cell) in another thread, yet the claim has been reinforced here again. So I have posted a more detailed, and slightly grumpier rebuttal to this claim. I hope this makes it clearer where I'm coming from.

Ah, I was missing this context as I didn't see the other thread.

Perhaps this thread might be a useful opportunity to clarify some of the different failure modes and their likelihood for A123 pouch cells? We might be able to apply some of the lessons to pouch construction cells generally. Perhaps ranging from manufacturing flaws to mishandling to abusive charging/discharging. Tell-tales in terms of physical appearance or electrical characteristics and perhaps what could be expected to be seen inside the cell if it were dismantled.

Some I can see already in this thread:

1) Isolation failure (in this case due to reworked tabs). Could this be caused by anything else? This is an electrolytic process that breaks down the aluminium pouch material? Is this visible from the outside? Is this also the result of mishandling (denting, folding, creasing, bending)?

2) Lack of compression. Causes accelerated cell deterioration due to what means? Results in increase Peukert losses. How much compression should be used for the AMP20 cells? Is puffing a visible result?

3) Over-discharge. Causes the copper anode to be dissolved and its surface to flake away?, lithium plates the aluminium cathode, both factors reducing active plate area resulting in loss of capacity and C-rate. Any external visible cues? Presumably increased internal-resistance and possibly self-discharge?

 

[wb9k]

PunxOr,

I appreciate your interest. Most of the items on your list have been discussed in some detail on this forum in the past. You might try prodding the forum search engine with a few choice phrases.

In your item #3, you actually combine two separate failure modes into one: Plating Li onto the cathode (not previously discussed in this thread) will cause loss of capacity, but has no direct relation to the growth of copper dendrites. Li plating on the cathodes is usually the result of overcharging the cell. Another well-known way to do it is hitting a very cold cell at a very low SOC with too much charge current. I was talking about Li "plating" onto the Al pouch. Actually, "alloying with" would be a more accurate term (so the metallurgists tell me), but this robs cyclable Li from the cell just the same. Hope that helps.

 

[999zip999]

I got a coffee grinder and grind that stuff up to eat, smoke and to put on a salad. I just don't play with it.

 

[wb9k]

I got a coffee grinder and grind that stuff up to eat, smoke and on to put a salad. I just don't play with it.

Oh yeah, I do the same. You ever sprinkle the shavings into your eye? That's the best!

 

[999zip999]

No but I pull them out of my bum after waiting 2 days, checking daily. I don't advise as it can be in bad health as cutting open a cell in a unprotected test zoon or assk first Please understand first then try to move forward or die. The man who went into Chernobyl wheir heroes those who jumped over the grand canyon ?

 

[whatever]

No problems guys, each person is entitled to their own interpretations and opinions, if you have some evidence that newer a123 20ahr cells have kapton tape under the aluminium tab ( without pictorial evidence of this your arguments dont stand up).

 

[whatever]

I have a different view to wb9k and 999zip999 on why the aluminium layer on the outer pouch deteriorating, based on what I have seen inside this cell.
My opinion is, From what I have seen from the inside of the cell, its not due to shorting through the plastic membrane with electrolyte held within it. It appears that the plastic lining of the outer pouch has opened up along the crease line inside.
This would then allow the aluminium layer to have direct contact with electrolyte, normally the aluminium layers inside the cell dont have direct contact with electrolyte, as its coated with lithium iron phosphate. ONce the outer pouch allows electrolyte to have direct contact with the aluminium of outer pouch it seems to react with it causing the bubbling. Thats my view from looking inside a cell. Since there was no breach in the white electrolytic membrane it would be reasonable to come to this conclusion. Once that very thin plastic layer inside the pouch is breached, the mylar type pouch will become an active part of the battery, acting as another plate. A slightly thicker inner membrane of the outer pouch might solve this issue, but I'm sure a123 is well aware of it.
You guys are quite entitled to a different opinion based on whatever your experiences are but it might be nice to keep the conversation civil and polite. Sorry I dont agree with you guys, it might be better to back up your arguments with some pictorial or other evidence, or at least some articles where studies have been done, that would be useful information for all.

 

[999zip999]

All type of batteries can have a manufacturing defect. Sla, lico,lifepo4 ect. And need to be recycled. But some times they get pick up and resold.

 

[wb9k]

I have a different view to wb9k and 999zip999 on why the aluminium layer on the outer pouch deteriorating, based on what I have seen inside this cell.
My opinion is, From what I have seen from the inside of the cell, its not due to shorting through the plastic membrane with electrolyte held within it. It appears that the plastic lining of the outer pouch has opened up along the crease line inside.
This would then allow the aluminium layer to have direct contact with electrolyte, normally the aluminium layers inside the cell dont have direct contact with electrolyte, as its coated with lithium iron phosphate. ONce the outer pouch allows electrolyte to have direct contact with the aluminium of outer pouch it seems to react with it causing the bubbling. Thats my view from looking inside a cell. Since there was no breach in the white electrolytic membrane it would be reasonable to come to this conclusion. Once that very thin plastic layer inside the pouch is breached, the mylar type pouch will become an active part of the battery, acting as another plate. A slightly thicker inner membrane of the outer pouch might solve this issue, but I'm sure a123 is well aware of it.
You guys are quite entitled to a different opinion based on whatever your experiences are but it might be nice to keep the conversation civil and polite. Sorry I dont agree with you guys, it might be better to back up your arguments with some pictorial or other evidence, or at least some articles where studies have been done, that would be useful information for all.

I am one of the top failure analysis guys in all of A123. You've torn down one cell, and you're telling me I need to back up what I'm saying. Well, I've said quite a bit already and it would be hard for me to tell you more or show you what I've written up on the subject without showing you the property of A123 Systems. Since I like my job, that isn't going to happen. If you want to think this makes your "analysis" more valid than mine, I can't stop you. I can make sure others can make up their own minds, however. Better luck with the next build.

 

[Alan B]

Thanks for the teardown and the comments here folks. Good to see. Always something to learn from it.

 

[999zip999]

Wb9k how dangerous are the insides of the A123 cells to us.

 

[wb9k]

Wb9k how dangerous are the insides of the A123 cells to us.

There is an MSDS included in the A123 Handling Guidelines published here on ES. The short answer is that it's not terribly dangerous in most circumstances. The solvents in the electrolyte can cause irritation of the lungs, skin, and eyes--hardly a surprise with organic solvents. The very worst thing that can happen is if any significant quantity of the electrolyte is mixed with water, it can form hydroflouric acid (HF), which is nasty, nasty stuff. However, it's pretty tough to do this with the amount of electrolyte you might spill from a single cell. Still, it's good to have a tube of calcium carbonate [CORRECTION--calcium gluconate] cream on hand to neutralize HF should you manage to make and get any on you. Fumes should be well-ventilated. Nitrile gloves and eye protection as well as full body cover by clothing are strongly recommended. Keep all water well away from opened or leaking cells. The remains of leaking or torn down cells should be kept in a heavy, sealed ziploc bag. Cells should generally be discharged to 1V or so immediately prior to teardown.

 

[whatever]

there were some pics of cut open a123 cells back in 2012 by agnesium
http://endless-sphere.com/forums/viewtopic.php?f=14&t=38545&start=500&hilit=youtube

 

[Ypedal]

On the topic of safety, once the electrolyte evaporates and the black carbon stuff dries up, any dangers of fine nano-particle size dust in lungs ? does it remain clumpy or does it break down in fine powder ?

 

[wb9k]

On the topic of safety, once the electrolyte evaporates and the black carbon stuff dries up, any dangers of fine nano-particle size dust in lungs ? does it remain clumpy or does it break down in fine powder ?

No risk here that I am aware of. It stays pretty clumpy. I wouldn't go breaking the stuff up and snorting it, but I've never heard such cautions from anyone.

 

[arkmundi]

LOL, I'm not going to tear one open. Just recycle when terminated.

 

[whatever]

I'm not sure why wb9k is being so negative against the taobao supplied cells.
He says they are

reworked trash

Once again I disagree with this view, considering the price of these cells compared to new cells, and the fact that you can buy them with good IR and capacity, its a bonus to ebikers that they are available ( you need a contact in china to organise buying them). The negativity against tabs that have some plastic rubbed off is I feel unjustified. It can lead to shorting with quite small currents involved, there was only one cell in the pack that had a shorting issue anyhow ( out of 16cells) and the solution to this shorting is so simple ( just a smear of silastic on the edge of the cut plastic)
I did quite a bit of searching on taobao to identify a supplier that actually tested the IR/capacity of the cells before buying them, from memory the price was 80rmb per cell ( approx $12usd, current prices vary from 60rmb $10usd upwards). The issue now is how to get the cells out of china, as its becoming increasingly difficult to send them out of china.
If you buy from a random chinese supplier of the secondhand cells there appears to be very few that actually take the time to test the cells, so its hit or miss, hence the need to do some research at the chinese end.
The source of the cells appears to be large ev packs but its not known by me what program in china is releasing the packs.
They are 100% used packs, but if you can get a hold of cells that still have good IR/ahr ratings why not use them?
So how about wb9k instead of being negative about these cells, how about back up your statements with sound reasons why you think they are "reworked trash", purely negative speculation is not useful in these discussions.
My guess regarding the source of the cells is either
1. after a given time ( one or two years) period the packs are replaced regardless of condition
or
2. more likely, one or more cells fail in a pack, rather than replace the bad cells the whole pack is replaced and those packs
sold off cheap to be taken apart and reused by small businesses, perhaps how victpower is sourcing its cells.
If whole packs are simply replaced with new packs, rather than identify bad cells and replace them, it would suggest a
govt funded program ( perhaps electric buses or taxis).

Either way there are going to be plenty of good cells in these packs. Its unlikely we'll ever know the true source of these cells.

With regards the cell that had a short, it was located in the middle of the pack I built, I noticed its voltage was always lower than the other cells ( when I built the pack I did for a time actually monitor voltages via cell log). Once I removed it from the pack I noticed I could make a tiny spark puting a screwdriver on the edge of the cut plastic on the copper tab, this was the source of the leakage, the aluminium on the mylar outer casing ( being in direct contact with the aluminium tab as there is no kapton tape used on that tab), the cut plastic allowed the mylar aluminium to contact the copper tab. A smear of silastic stopped the problem. I just tried to measure the resistance of the aluminium of the mylar type pouch, its hard to get a stable reading but appears to vary from 2 to 60ohms, that would allow quite high currents to flow via the mylar type pouch aluminium if shorted to the copper. Quite surprising I didn't think it would have thought it would have a higher resistance to current flow. Anyhow the silastic method worked well to solve the problem.

 

[wb9k]

Whatever,

Take a gander at this thread: http://endless-sphere.com/forums/viewtopic.php?f=14&t=53962&hilit=a123+b+grade+info#p802947

The provenance of most of the grey-market cells is well-known. The thread addresses some of your other concerns as well.

Were I a moderator, I'd consider banning you if you continued to suggest such dangerous practices are perfectly fine, even in the face of a multitude of expert advice to the contrary. It's irresponsible and damages the useful utility of this board for others.