lithium polymer battery explosion

[john monks]

I was doing some destructive testing with a 190mAh lithium cell at work to see how durable they are. I ran a continuous 1 ampere and when the cell reached about 12 volts it exploded with such concussion that I jumped back about a foot. So I'm wondering if there is any way that I can place a lithium cell into a product and and reasonable confidence that the cell will not explode.
1. Can I be sure that the cell will not explode if the cell is run between 3.2 volts and 4.2 volts?
2. What danger is there if a cell is inadvertently cut in two or run over?
3. What danger is there if the cell overheats?
4. What danger is there if the cell is shorted out?
5. What happens when a cell is inadvertently charged backwards?
6. What if I charge several cells in series? Are the somewhat self equilizing like lead acid cells?

 

[eKretz]

Lithium ion batteries can be very bad news if they are damaged in any way. Whether it's mechanical damage or internal damage. One guy on another forum had a flashlight explode and bust out the glass in his sliding door and put a pretty massive dent in the aluminum floorplate with flashlight shrapnel. It is a good idea to get rid of the batteries if there are any significant cuts or nicks, as well as if they ever check over 4.2VDC or under 3VDC with a multimeter. The packs normally have circuitry built in to protect them from overcharging or discharging past their proper cutoff point. They often have circuitry built in to properly balance the cells also. They do not self balance in series, and can get out of whack and become very dangerously overcharged if used/charged in series without proper balance circuitry. I always charge mine in parallel for this reason.

 

[john monks]

After posting my questions I connected three lithium cells in series and began charging with a constant current. One cell came up to 4.8 volts, another 4.5 volts, and the remaining cell came up to about 0.5 volts and began to bulge like a balloon.
So at this point, unless the technology changes, I'll have to scrap the project. I am very disappointed because lithium polymer cells have about three times the energy density of the cells we are currently using (nimh). If I can find a safe way of using lipos I would do it. Thanks eKretz.

 

[GreenGiant]

There are many safeguards you can put into place, individual charging of cells, having similarly classified cells (each cell is going to have slightly different characteristics).
Having over and undervoltage protection will help limit problems with the batteries.

What you are asking for is a very difficult thing to answer, I worked for a LiPo battery company for almost 2 years and most of their research was focused in 2 areas, energy density and safety, there are many formulas that are very safe even if the battery is punctured 100% of the way through, but they will have lower energy density than the ones that arent as safe. (so far anyways)

There was a whole department designated to destroying the batteries in as many ways possible (crush, nail penetration, slice, over heating, overcharging, over discharging, etc)

to answer your questions:
1. You cant be sure of this with any battery, but with monitoring circuitry you can lower the risk, 4.2V is a bit high voltage for standard LiPO, their range is typically 2.6-3.6V there are different formulas that will easily go to 4.2 but those are more the "oxide" cells which are a tad bit more... volatile... so to speak.
2. As stated above this is dependent upon the cell, also state of charge, temperature, etc.
3. Over heats is a relative term, some cells are perfectly happy to go all the way up to 150-200C BUT charging or discharging them at these temperatures can cause internal damage usually leading to lowered capacity and sometimes what are called "vents" which is when the pressure inside is enough to cause the battery to leak its electrolyte out without exploding.
4. Some cells are fine with shorts, not for long though, almost any LiPo cell will recover almost 100% after a momentary short (<250ms), longer than that and you either start melting things or causing internal shorts, either way you will have a diminished capacity.
5. Charging backwards is probably the best thing that can happen to the cell of all the bad things, this will firstly discharge the capacity to 0, then it will vent and cause an open circuit if youre lucky, if not it will start charging capacity backwards which, depending on supplied current, may either cause a vent, a meltdown or something a little more violent.
6. With properly balanced cells, and voltage supply, this should be no problem at all, theoretically they should self equalize BUT they are all so different you sometimes get cases like you were saying where one will charge and the others will not.

Please note that in most cases the electrolyte used in LiPo batteries is HIGHLY flammible, so be conscious of that.

 

[GonzoEngineer]

Reminds me of when I was working for Sandia Labs many years ago.

We were doing a project using Lithium Thiony (sp?) Chloride batteries. Turns out the Lab
had no safe disposal procedures in place, and the disposal guy's wouldn't touch them.

We had a fifty gallon drum outside our building. So we twisted the wires to short them out and threw them in the barrel.

I thought I was back in Vietnam! Those damn things exploded like a concussion grenade!

 

[Lord_grezington]

Is there any way you can check the cells to ensure that if you were to connect them in series the charge will be uniform between all the cells? I would assume that if each cell had the same resistance the charge would be uniform? Yes manufacturing processes are not perfect, but could you request balanced cells from the manufacturer? How do we get properly balanced cells?

Individual charging is awkward as it needs individual circuits (and monitoring) with a lot of wires between the charge controller and battery.

 

[john monks]

I used 2500mAh cells out of an old Sony laptop computer. The cells were connected directly in series so I'm confident that the cells were receiving the same amount of current. I never did order new cells because I originally did not have much confidence in using these things in our products safely so I got a whole bunch of lithium polymer cells out of old equipment. So the cells probably were not equal to start with. I noticed in my bunch of cells that I have some batteries with three cells in series and they are fully charged and the voltage between the cells is ruffly the same. And the voltages came out differently so I assume that some of the cells were leaky and I'm afraid that if one cell leaks the other cell will overcharge and I will have an explosion. So that is why I asked if they are sort of self equalizing. But base on what eKretz, Green Giant, and GonzoEngineer I can see why more people haven't done this and frankly I'm a bit afraid of this project. I don't want to hurt a customer. I would feel a whole lot better about this if we only needed about 3.5 volts. To equalize the voltage would require some extra circuitry and that takes room. I'm grateful for all the help.

 

[Audioguru]

Is there any way you can check the cells to ensure that if you were to connect them in series the charge will be uniform between all the cells? I would assume that if each cell had the same resistance the charge would be uniform? Yes manufacturing processes are not perfect, but could you request balanced cells from the manufacturer? How do we get properly balanced cells?

Individual charging is awkward as it needs individual circuits (and monitoring) with a lot of wires between the charge controller and battery.

New cells made at the same time are balanced when new. But they become unbalanced with age and use.
If you have Lithium cells in series then you are supposed to use a balanced charger that automatically senses and adjusts the voltage on each cell. The balanced charger has a connection wire from each cell.

 

[Harald Kapp]

@Audioguru : do you realize that this is an old thread from 2013?

 

[Audioguru]

Why isn't an old thread indicated with a red color or something that it is old?
I rarely look at dates because most dates say "Today".

 

[Harald Kapp]

I'm sorry, I have no answer to your question.
Small comfort: you're not the only one caught in this trap