battery format - cylinder vs pouch

D

davec

1 kW
Joined
May 22, 2011
Messages
345
http://batteryuniversity.com/learn/article/pouch_cell_small_but_not_trouble_free
http://batteryuniversity.com/learn/article/types_of_battery_cells

i can't make up my mind with which format to go with. i am leaning more towards the cylinder format as they seem to be
more reliable than pouch cells
in the past cylinder cells have always been manufactured with better qc, and it was common to have more failure in the large format
have prismatic/pouch improved in quality over the years? reliability?
can anyone with real life testing confirm

"The prismatic pouch pack is light and cost-effective to manufacture. Exposure to high humidity and hot temperature can shorten the service life. A swelling factor of 8–10 percent over 500 cycles is normal."
 
Chemistry aside I find pouches usually cram more capacity into frame space but I must admit it's been a few years since I used any cylindrical cells.

There's many a good reason to use cylindrical packaging but with some care, many pouch cell packs have worked fine for many years.
 
As with anything you read from the notoriously outdated and/or incorrect battery university, there is some truth and some non-reality.

If you manufacture a pouch cell right, package it correctly, and manage it properly, it's as reliable, robust, and long lasting as any cell format around.

Just like if you make shitty cells in a can, you get a shitty pack as an end result. Or, if you start with excellent cells in a can or pouch, and design a pack that doesn't package them properly or manage them properly, it's going to be shitty and fail.

You can have a long lasting durable design with any format done right, just like you can have a shitty pack from any format of cell or pack construction type if you do a shitty job with it.

The new generation of 18650's with some decent poke and energy density (like the new Sony cell, or the NCR18650PD) make 18650's a more attractive option than ever before for an EV pack. Still though, the pack having a long and reliable service life is going to come down to how the pack is designed/constructed more than anything to do with cell type.
 
Battery U is good for the basics, learning the difference between series and parallel and such.

Deciding on the battery is a very complex decision, and can only be made after ALL other decisions have been made. Once the motor and controller decisions are set, and you know what space you have to carry the pack, you can begin to look at what works within those limits.

If you will be making a bike, a very small vehicle, but expect it to perform big, then you already know energy density and c rate will rule your decisions. Or, if you will just commute at 20 mph on a bike, any lifepo4 battery can do the job.

But as Luke says, the single most important thing may be how the pack is constructed. Broken connections won't get you there, nor will damage from road vibrations or even crashes. The best cells in the best format will fail if poorly assembled. Really poorly assembled packs get damaged in shipping. That's how the cheap 18650 lifepo4 pack got it's horrible reputation.
 
thanks guys
another point is that the pouch format is way easier to manage build/monitor/ troubleshoot if something goes wrong
i would prefer going this route
with the cylinder cells you got alot of work- testing cells, thermal management, and if something goes wrong troubleshooting can be a pain- also if you look at the pack statistically with the large number of tiny cells in it - there's a higher chance of something going wrong and if it does happen and there's a short than there's even safety issues unless the pack has failsafe features built in- increasing the complexity- also as the cells age you're hoping for uniform aging(which could be another problem).

but assuming the failure rate on these smaller cells is so low - than the effort might be worthwhile , and there are alot of folks using these type of packs on here with success. at least cylinder cells don't need compression
with the pouches you need compression- you see alot of puffed up pouch cells on here and im assuming these are older cells that were abused or were in a pack that was not build properly. but even built properly according to battery university they still wrinkle with age, im assuming with newer higher quality this wouldnt be the case. either way if a pack constructed properly can be just as reliable than pouches cant be that bad and easy to work with- i think even ping uses pouches and there's good reviews on his packs
 
Oh yes, pouches will puff with age, or abuse, or if they are sloppily made like RC batteries.

I wouldn't say fixing a problem is that much easier with pouches. Unless, like you do with RC stuff, you can just toss a 5 ah section and plug in another. Either kind can be built in modules that plug together, rather than a gigantic block all glued together. Pings are easy to chop up into 12v sections, the way he constructs the pack.

If a 2c cell will do your needs, then by all means go with ping. I put 10,000 commuter miles on two ping batteries. Each one lasted longer than 3 years. But now I run a heavy bike twice as powerful, and use the RC batteries for extra oomph. I'd have to carry 30 ah of ping to run my longtail.
 
and that`s the thing, ive never seen a quality cylinder cell puff - the swell rate is something im curious about- i might even run some tests on pouch cells under compression to see how much they swell with time-- maybe manufacturing processes have improved to the point where this dosn`t happen as much....

I could be wrong wrong but i just feel that the smaller format cells are more reliable than the pouches- (heck i even seen MIT do a cell cycler and had thousands of cycles on a small cell with only a loss of 10% capacity and nothing more)-assuming the pack is build properly The mass produced 18650's (assuming you get a123,Sony, Sanyo etc. and not generic china brand) are made in factories that strive to keep defects in the parts per million range or less-- this used to be the case in the past, if manufacturing improved on the large format than this might be a total moot point
 
You can get amazingly good cell quality from a tier1 18650 provider if you're paying for good cells and not the B-stock reject stuff the flashlight battery vendors typically sell.

A properly designed and constructed 18650 pack or pouch cell pack needs to support the cells, never mechanically load the connection points, use a connection process that doesn't locally over-heat and damage the cells, be capable of flexing/shock/vibration without failing connections or shorting, and withstand weather exposure (if you're going to ride in all conditions at least). Then you need some method to monitor/balance as well, which can be off-board or on-board the pack depending largely on your use preferences.

Worry about getting those things right before worrying about the cell itself IMHO.
 
your batteries should have some headroom. I wouldn't use a 2c battery at 2c. I think a 5c or better yet a 10c and don't over discharge or over charge. I mean a true 5c and not claimed if you are using for a commuter. A race bike is a different story.
 
We would need to first understand the battery current profile for your motor/controller/gearing/weight/aero/speed range to give a good answer. It requires some system design to go through before you can get a good answer.
 
Start a new thread Giovanni, and give all the details of what the car will do, what the cars systems require, etc. Your question is buried in this thread on another subject.

If your commute is at 35 mph, the requirements for the same vehicle will be very different from a 65 mph commute, for example.

Car size, many that convert a regular car have used large format lifepo4 in the plastic "prismatic" boxes. But not because it's best. It's just easiest to get, cram in the trunk, and hook up much like you would have done with lead. The lame c rate is dealt with by making the pack very large.

I think for a homebuilder, an 18650 can pack that large would be quite a challenge. It's not like us with bikes, who can just stuff 10 pounds of glued together cells in a bag.
 
I would be looking at Enerdel batteries as well. No need to roll your own when a high quality purchasable solution is available.
 
for a car personally i would consider large format
using packs made out of cylinders is possible- but would take forever... for some reason tesla chose to go this route ,
even for an ebike building a small 48v pack would be very time intensive...also big packs made from cylinders can be harder to troubleshoot. seems to be that most of high quality manufacturers are caught up in the smaller format,but for large format cells, a123 20ah,the enerdel 5c cells,kokam, calbs might be worthwhile
it appears that the rimac, busses are using the large format

Based on the links below the cylinder cells seem bullet proof - if the bigger format has become more reliable than it might even make sense due to ease of construction- i'm still trying to find a source or research done on the bigger format to show reliability

some interesting research
http://mit-evt.blogspot.com/2010/07/new-cell-cycler.html
http://praveens.mit.edu/~praveens/wp/wp-content/uploads/2010/06/CellCycler.pdf

informative links about pouch cells
http://www.mpoweruk.com/cell_construction.htm
http://electricmiata.blogspot.com/2012/11/2-month-inspection-of-a123-cells.html
http://www.buchmann.ca/chap3-page3.asp
http://randomev.wordpress.com/2012/01/21/pouch-cell-pack-building/

Mr Yeow, notes the following
disadvantages are: (1) mechanically vulnerable and thus requiring cell cartridges to hold them, (2)
prone to swelling during operations especially when the cells age, and (3) has no mechanism for
gas venting (as opposed to cylindrical and prismatic cells). Gas venting for the pouch cell involves
swelling/breaking of the pouch and hence causing cell failure

another note is pouches dont have safety fatures like vents/ptc/cid so they simply rupture if something goes wrong, cylinders have vents+compression, but regardless i dont think you ever want those toxic gasses to come out(a cell should never vent). bottom line here is i feel that cylinder are a more mature technology, pouches on the other hand still havent been proven but theoretically if not abused should work ok
 

Attachments

  • CellCycler.pdf
    5.8 MB · Views: 49
  • Yeow_AVL_final_2202012.pdf
    3.2 MB · Views: 35
You must log in or register to reply here.

Similar threads

nutnspecial
cans pouches cells & chemistries/ past present & future
Replies
3
Views
707
LewTwo
LewTwo
S
DeWalt 20V Max 3.0Ah Battery Pack Teardown & Analysis
Replies
10
Views
25,133
qwertyES
Q
S
Electric Car Battery Packs and Longevity
Replies
6
Views
1,740
Kvno
K
Lock
Leyden Energy Li-Ion superior energy density,longer run time
Replies
5
Views
1,548
Lock
Lock
TylerDurden
NYT: - The Secret to Solar Power
Replies
10
Views
4,170
arkmundi
arkmundi
Back
Top