13

I'm designing a device with a small lithium polymer battery (4x12x30 mm, 120 mA-h). Looks like this:

enter image description here

I've heard that there is a rule of thumb that the space left for the battery in a case should be around 10% larger (I suppose primarily in thickness) than the nominal dimensions to allow for expansion. Extra 10% seems quite large.

  1. Where does this rule of thumb come from? Is there any official recommendation for how large a compartment to put lithium polymer cells in?

  2. How much do these batteries expand and shrink in normal use? For example during charging/discharging cycles, temperature cycles over normal temperature range (-20C to 60C), etc.

  3. What happens if the battery is in a rigid compartment in the case of malfunction? It's pretty common to have batteries "puff out" if they get shorted internally, but what happens if the battery is in a compartment that prevents expansion? (Assume the compartment is strong enough to withstand the pressure build up) Does the pressure/walls make the short worse, or better?

Voltage Spike
  • 75,799
  • 36
  • 80
  • 208
Alex I
  • 3,341
  • 2
  • 36
  • 59
  • Could you source your rule of thumb comment from? I've never seen such a thing in any device that I've taken apart that has a lithium polymer in it from laptops to phones. Maybe that's because I was opening old products? In the few cases where battery expansion happened, 10% wouldn't have been enough to allow it. In most cases, it starts to rupture/sever/open up the hard case trying to contain it. – horta Feb 01 '19 at 21:06
  • @horta - That's from a couple of MEs I've talked to (and they couldn't source it either - just engineer lore I guess). They always want to add significant extra thickness over what the battery is specced at. Eg for a 041230, they would want 4.5mm space. Btw the battery spec is the max thickness, it has a tolerance like +0, -0.5, so this is all for expansion after assembly. I've taken apart a few devices too, I think it's common to have a bit of foam between the battery and case, but agreed less than 10%. – Alex I Feb 01 '19 at 21:17
  • Maybe it's a safety rule, in case the battery blops open, leaks, swells, melts, heats, or any other hazardous reaction. The tolerance prevents the case from opening exposing electricaly charged parts – Fredled Feb 01 '19 at 21:25
  • I've definitely seen the sticky foam in most of them. I wonder if the extra 10% is more than anything for room for the foam for shock protection. – horta Feb 01 '19 at 21:27
  • I have been told by cell manufacturer that the cells can swell and that we need to leave a bit of room. Don't remember if it was 10%, but that sounds about right. They can swell not only due to temperature and cycling, but they can outgas a little bit as well. But the cell manufacturer should be able to answer the question. Anything that allows pressure to build-up will make violent rupture more likely if the battery should vent due to failure. It's not practical to design a structure around the battery strong enough to prevent venting gasses from escaping. – user57037 Feb 01 '19 at 21:53
  • 1
    In the papers I read, it was the anode or cathode that swelled 10%, so that's where I think 10% comes from. – Voltage Spike Feb 02 '19 at 17:46
  • @laptop2d, academic or trade publications are probably more accurate regarding the cause of the swelling than random manufacturing reps. Of course someone at the manufacturing company probably knows why the batteries swell. But my conversation was not with that person, so I will assume what you say is true. – user57037 Feb 02 '19 at 18:41
  • 1
    @mkeith I think the most important thing is to do the swell testing yourself, or get it from the manufacturer. It's not hard to do, just need a set of micrometers and a way to charge and discharge the cell – Voltage Spike Feb 02 '19 at 21:18
  • 1
    As far as puffed up batteries, the battery must be designed to either expand or vent. Bag type batteries like the one in your photo expand, hard shells like 18650s vent. A failure point must be provided in order to prevent catastrophic failure, so if you design something with a bag type battery, if possible, put failure points in the casing to ensure it will pop/split open if the battery bag expands, and allow no nearby shapes that will puncture the expanding battery. Only if it is not *possible* to have the case expand, a failure point must be provided to ensure the case will vent. – K H Feb 02 '19 at 21:35
  • 1
    Most of this is basically preventing a fire from becoming an explosion. If a battery type is used that must vent, IE 18650, then the casing must be able to vent or expand as well. A weak point in plastic, a rubber/soft glue plug or snap apart tensioned hooks are all acceptable. Anecdotally, this year I ended up using my Samsung Galaxy S6 until the battery could hold only enough charge to power the phone for about 2 minutes. I use Lifeproof cases, which are waterproof, so totally sealed and the front and back clip and hold together with quite significant force. – K H Feb 02 '19 at 21:46
  • 1
    When I finally got around to replacing it and opened up the outer case, without the strong pressure of it holding the phone together, the front and back covers were both popped right off the frame by the expanding battery(designed failure point), so I can provide a pic or two for your question if you like. Note that despite the compression, the waterproof case does have adequate failure points, as if the battery did vent (expand so much as to pop), the usb port cover, ear phone jack cover and assorted thin membranes would open/fail, preventing an explosion. – K H Feb 02 '19 at 21:46
  • @KH Thank you - this gives me enough to work with, I think. The case I'm working on probably won't have enough room to let the battery expand, but could easily have designed failure points and that's what I'll do. Btw: does increased internal pressure make a battery failure worsen over time? (Ie: should the designed pressure limit for the case be set relatively low or pretty high?) – Alex I Feb 03 '19 at 00:27
  • 1
    Rather than thinking of it in terms of an arbitrary pressure level, think about it under the assumption that the battery will fail in a slow way(if possible as it expands, it should indicate the need for replacement before the destruction of the device. Allowing the case to split open accomplishes this. Then in the worst possible way (rapid fireball and production/expansion of gas) and when you look at it this way, look to minimize the dangers presented. Spend some time when you think you have a good design and really think "What *will* happen when this fails?". – K H Feb 03 '19 at 00:56
  • 1
    Don't let pressure build up to make a grenade(how much this is depends on material and thickness). Avoid creating a jet from things like a narrow channel with plastic edges that will melt the plastic while still concentrating enough pressure to launch it or even just to direct a flame. Think about the intended use as well. Battery in the handle of something? It should vent out the top or the bottom, and definitely not sideways through the hand. Battery in something with an eyepiece? I can think of one direction I wouldn't want it to vent! Fire, pressure, heat, gases, projectiles. – K H Feb 03 '19 at 01:00
  • 1
    I would try to figure out how strong it *needed* to be, and make it only 5/10% stronger than that. Then look at that value the other way and make sure the value it "needs" is not so high as to be dangerous. You would need to have a pretty high pressure/robust application for these two things to conflict though. – K H Feb 03 '19 at 01:04
  • 1
    @KH Maybe you can post these comments as an answer, it's a good way to think about this. – Alex I Feb 07 '19 at 09:47
  • @Alex I I was anticipating the bounty drawing in the usual mob of engineers, so assumed this would end up being later-deleted brain fodder for them. I'll see if I can find time to parse it into an answer. – K H Feb 07 '19 at 23:21

3 Answers3

14

This paper measured a cell, they reported a max of 0.5% expansion over a charge cycle:

enter image description here Source: Expansion of Lithium Ion Pouch Cell Batteries: Observations from Neutron Imaging Figure 7

Over the lifetime of the battery, it swelled more than 1.5%

enter image description here
Source: Expansion of Lithium Ion Pouch Cell Batteries: Observations from Neutron Imaging Figure 9

One could use these numbers for a baseline, but it's not that hard to make these measurements to a reasonable degree. Since batteries are made with different anode\cathode and electrolyte combinations that vary from manufacturer to manufacturer, it would be wise to consult the manufacturer on swelling or measure it.

If you really want to find out how much your battery is swelling, get a micrometer and measure it discharged, then measure it fully charged and see what the difference is. Measure the cell under maximum discharge, because the cells swell more with thermal expansion. Give your self additional margin to account for differences in batteries and manufacturing tolerances.

Cells also bulge more in the middle if they are heated then the outside. So make sure you measure the middle of the cell.

enter image description here
Source: https://www.researchgate.net/publication/283720424_A_novel_thermal_swelling_model_for_a_rechargeable_lithium-ion_battery_cell

Voltage Spike
  • 75,799
  • 36
  • 80
  • 208
3

Although I cannot compete with the wealth of detail given in previous answers, I think it might be helpful to give you an example why this much extra space is used for a good reason.

A Lipo does not only swell during normal operation with temperature and charging/discharging, but also when it ages. Have a look for electrolyte decomposition to find more about this phenomenon, but ultimately it breaks down to the creation of gases (mostly oxygen) inside the LiPo.
Just to give you an idea about what you can expect: A frequently used 3.5 year old battery I measured during writing this expanded from 25mm (according to the reseller, couldn't find a datasheet) to almost 32mm, which is more than 25%! I guess the datasheet values are not as optimistic as the reseller's ones, but still this is a substancial increase, which should be taken into account while designing your product.

If the battery has no room to swell, it will become a possible risk of fire and - in the worst case - even explosion. See the comments on below your question, this has been described by K H in detail.

K. Krull
  • 336
  • 2
  • 8
  • Hmm, what type of battery was it? 25mm is a lot for a pouch cell, do you mean 2.5mm? Does it seem puffed? I think the typical tolerance on cell thickness is almost 0.5 mm. – Alex I Feb 07 '19 at 09:45
  • I was talking about a 4S 2200mAh LiPo battery used for RC cars and multicopters, the big variant of your small single cell (the 25mm are intended). It is not puffed yet, but I would not recommend using it anymore. I dug a bit deeper and found an old 1S battery by now, which is quite similar to yours. It looks swollen as well, but not as horrible as the big one. Unfortunately I can't tell you numerically, since I don't know how thick it was originally. – K. Krull Feb 07 '19 at 12:45
3

There should be a distinction made between inevitable electrode expansion/contraction due to electrochemistry of electrodes itself at nanoscopic level (which was presented by lapto2d), and "battery swelling/puffing" of pouch-type cells (presented by K.Krull) due to electrolyte decomposition/outgassing, which is a sign of malfunction and/or poor manufacturing quality of a cell.

Regarding puffing, there are several theories about it, but it looks like the main cause is some electrolyte decomposition and metal build-up when the cells are left in nearly over-discharged state for long time.

The manufacturing issue is related to production process, where the cells are "formed" before being sealed, letting the electrolyte to outgas. If the forming is done sloppy/too fast, the sell still has some gas build-up and will puff over time.

Obviously the overall cell expansion in practice is a combination of the two effects, and some studies of well-made cells show expansion up to 4% after 50 cycles, see this publication.

While the 0.5%-1% of electrode thickness expansion is natural and can/should be accommodated with some oversize of battery compartment, excessive irreversible puffing is a serious precursor to catastrophic failure. At one point in dealing with customer's issues I came to realization that it would be very beneficial to have a pressure sensor to detect this state before the whole device case gets torn apart. It appears that this idea is already patented, US8717186B2. I strongly recommend to put a pressure sensor inside your design.

Ale..chenski
  • 38,845
  • 3
  • 38
  • 103