Pouch Packs -- Stapling Tabs

[JackB]

I just finished testing over 300 pouch cells. It took weeks to do them all, but it was neccessary,
as they were used and some dead, so only way was to break them down into single pouches and test each one.
The good news is most of them were good to very good. Some poor and some totally dead.

Anyway, I've read through all the AMP20 pack builds including all of the 36 pages of one thread.
Some of the pouch assembly kits look good, but I need a fast, cheap, and LIGHTER way.
The pouches originally used some plated copper 'sandwhich slabs', and I can't believe how heavy there are.
So my first attempt was using PVC strips bolted together, and that works and is pretty light,
but it is time consuming method. So I came up with something very fast, cheap, and light,
which is folding the tabs and then stapling them. My only concern with this method is the flammability of
using wood, which is needed to hold staples. However, I don't see the battery can get hot enough to ignite the wood,
and one rationalization is that wood smokes before catching fire, so one could smell the danger coming.

So here are pics of my first PVC strip pack, and the old copper bus connectors and the new staple method.
What do you think?

 

[agniusm]

I think it's as bad as it can get.

 

[Teh Stork]

Oh dear...

This is an alternative that is lightweight. I used this recently to put together some A123 cells that have been resting for a year (Still at 3,29V). Holes in tabs are done with a "hole hand punch" (for aviation). I use M3 nylon nuts, two washers and one M3 screw (x4). The nuts are retained in a 3D-printed plastic "holder" that allows easier assembly. The joint is coated with electrical grease before assembly (will give better corrosion resistance) and are tapet to prevent shocs. Pictures to follow:









You should care more for reliabillity than fast-cheap-easy nonsense.

 

[JackB]

Realize that I have ~300 pouches to connect. The current idea is 20s blocks.
Punching hundreds of holes and screwing hundreds of screws does not have much appeal.
I have made another 8s pack using clamps and bolts/nuts, I find it quite dangerous using screw drivers
and wrenches when working with these. An air stapler is pretty darn fast and easy.
Doing the folding is what is more time consuming, but the folding is what is ensuring good contact.

Another 'outside the box' idea is to connect them using a staggered layout,
this pretty much eliminates having shorting contacts near each other during attachments.
It is very dangerous assembly these are if you are not careful,

 

[999zip999]

Yes this cell has 18ah even at 3.6 volts they can be shorted for some fire works and at 20s or 70 volts and 18ah can make some heat. Like shorted sense wires ect.
What will this pack be used for ?

 

[JackB]

Cells are going into my Porsche Boxster EV

 

[JackB]

So I stapled up a 4s pack from some pretty toasted cells to test.
Charged them up for a few hours, and then did a little arc welding with them,
but not really, just used a welding electrode for the dead short test,
and was dumping 120amps for a minute or so, after a while the electrode gets hot,
and then the cells start to get hot, but the stapled tabs are fine.
100amps is the max for a string, 12 strings = 1200 amps all the controller can handle,
So looks good to me.

 

[999zip999]

That's the way we do it.

 

[999zip999]

I just used a basic table hole punch ( 3 hole punch ) and tape for a guide line for the taps, just two holes. For Durtleght stlye build. But 300 cells ? So 20s3p packs. What you thinking ?
Plus if you use threaded rods for compression, put some plastic tube on it.

 

[dmwahl]

Oh dear...
You should care more for reliabillity than fast-cheap-easy nonsense.

+1
Staples are just not meant to provide even and consistent clamping pressure. Given how much time/money has obviously gone into the rest of the build I would not cut corners here.

 

[JackB]

It is the folding of the tabs that provides good contact, the staples just keep them folded.
Folding increases the surface area contact. Compare this to using just bolts, or even
the bar clamps, which only contact the tabs where the bar is.
Staples are very reliable attachment method when stapled properly.
The wood itself can rot, but a battery will wear out before the wood does.

It doesn't look sexy and high tech, but that is not really relevant.
I would prefer to laser weld them, not easy to DIY, and the staples can be removed
with some effort if you have to.

 

[Punx0r]

It is the folding of the tabs that provides good contact, the staples just keep them folded.
Folding increases the surface area contact. Compare this to using just bolts, or even
the bar clamps, which only contact the tabs where the bar is.

If the tabs in your photo were anything like flat I might agree that the folding helps, but I imagine you have few points of contact. Bolts with large heads, or washers, will tightly clamp a sufficient area (think of the size of a ring terminal). Bolts with a bar are even better. You simply don't need to clamp the entire tab area. A123 didn't bond the entire tab area.

A tight joint (contact pressure) is also important. Staples provide little clamping force.

Staples are very reliable attachment method when stapled properly.

Do you have anything whatsoever in the way of evidence to back this up?

I imagine you will take the responses in this thread to be unwelcome criticism, but trust me, it is constructive criticism you get on this forum. These people know their stuff. Also consider that many good engineering solutions look "right".

 

[JackB]

I have just as much evidence that staples work as the evidence posted by the ones that say they don't.

After more research, I found this quote on how Bill Dube connected his tabs.

"This is going to sound very simplistic, too easy to work - but hear goes. Bill Dube said he has had excellent results using just a metal spring clip, like you use to clip a stack of papers together. He doubles the 2 tabs over on one another with a small dia, toothpick size, wire between the two so that the clip will not slip off. He uses a very small amount of conductive paste. He says he has gotten very high amps out of the 20ah pouches. "

I'm sure everyone would say this would not work, it has uneven pressure, it doesn't make good contact, Bill Dube is a cheap fool, but if you didn't know, he has one of the quickest EV's in the world, so maybe he knows a little.

I have found in my research, that some people make it a PhD thesis on how to connect these cells,
and have more parts and time invested in making a simple 12s pack than it takes to build an entire EV.
The more fancy and difficult the better it seems.
Believe me, I always do things the hardway, I can spend weeks crafting custom parts,
but we are just connected two tabs for christ sake.

And I have already demonstrated that 120amps creates no heat in the stapled tabs.

I will say however, that I don't particularly like it. Doing more of them, the staples have a hard time going through
all four layers of tabs, particularly when using the air stapler.
Works quite well going through just two, so I need to test them without folding the tabs,
which makes it faster, but now its more dependent on the staples.

 

[Hillhater]

If something doesnt look right..then its not right !
..and sorry but that looks bad !
Do you think Porsche (or Fiat even) would secure anything important using staples into wood ?
If you want to stay with staples, use one that is bigger and folds onto the back, rather than just sticking into a piece of wood.
But there are better "permanent " metal crimping systems that could be used to give a tighter, permanent clamp for folded tags.

 

[Teh Stork]

This is possibly the most elegant solution to joining tabs.

TE-connectivity TERMI-FOIL line of products are, together with a suitable crimper, perfect. Digikey stock most of these.

Even for Alu-copper joining, like the A123 20Ah, this solution with some ASI compound #70 (Grease like galvanic corrosion suppressant) should give fast and reliable connections.

Crimpers for underfloor heating, rather than the Tycho brand for TERMI-FOIL, are much cheaper.

Or finding suitable dies might be all that is needed (I found one for 9 USD on aliexpress).

 

[Punx0r]

I had no idea such connectors existing. They look very promising?

 

[Hillhater]

This is possibly the most elegant solution to joining tabs.

TE-connectivity TERMI-FOIL line of products are, together with a suitable crimper, perfect. Digikey stock most of these.

Even for Alu-copper joining, like the A123 20Ah, this solution with some ASI compound #70 (Grease like galvanic corrosion suppressant)

Crimpers for underfloor heating, rather than the Tycho brand for TERMI-FOIL, are much cheaper.
).

Spot on Teh Stork.
You have more patience than me for digging up the link details !!

 

[dnmun]

if you have seen how the hobby king pouches are connected now, you know they don't solder them anymore but use these multitooth crimpers to connect the aluminum positive tab to the copper negative tab. since there is not a copper tab crimped to the aluminum positive tab on these, they cannot be easily split apart into two shorter series which i have to do to build batteries when the series count is not a multiple of 4S or 6S sorta thing, and when i have to repair it.

this is the biggest problem i have with building the big lipo packs from the nanotech pouches. lotta wasted pouches when i could not split the tab down the middle between them to conserve copper tab for soldering on each side of the joint.

finally had to give up repairing the early failures in my big 21S3P 8Ah nanotech pack. as you can see it ended in failure when one tab was not big enuff to solder to after i cut the replacement pack apart. so in cutting down to 2P i discovered a ruptured pouch on row #1, cell 6. it was not even leaking down when i had tested each of them on that channel too.

now the only solution with what i have is to make 21S2P out of what is left, never gonna buy these nanotech again.

this is major problem for building real batteries wit the HK packs that are not multiples so the 4S or 6S, and repairing them will be difficult too.

 

[John in CR]

Jack,

Fold and crimp. Don't forget the most important part, which is compression. That's a big factor in what happened to those cells you tested, and improper/inadequate/non-existent compression is big reason why so many of these guys have had premature failures of their AMP20 packs. All pouch cells need nice even compression.

What voltage are you planning, because I don't understand why you're doing 15 20s packs instead of paralleling first. Parallel first makes bolt together a lot easier, and also avoids a rat's nest of parallel wiring for balancing.

Don't try to rush too much, because the time you invest in building a good pack is time well spent. Folding tabs together, though I think I would have gone for a second fold, and power stapling is a novel idea though. The only part I don't like is the physical stress you've got on the cells. The end of the cell rightmost looks quite bent over. Cut one of your dead cells open and see how many layers are in there. Disturbing those separate layers in any way reduced capacity and increases the risk of premature death.

Please be sure to share with us how your car turns out.

John

 

[dnmun]

i disagree. these big pouches should be assembled in series initially and then tied together in parallel through a small wire to allow the BMS to balance and monitor the pack. it is impossible to isolate a bad pouch if they are tied together in the same node by connecting all the pouches on one channel into one parallel set of pouches.

if the rows of serial pouches can be isolated from each other by unsoldering the paralleling sense wire then you can establish which pouch is failing by isolating that series from the others. this would be impossible if they were all soldered to each other first, and it is not needed either.

what i don't understand is why he doesn't just solder the tabs to each other instead of stapling them. the A123 pouches have tabs that you can solder to.

i did see dissolved aluminum inside the mylar in the first picture so that pouch is already dead. all of the pouches with the aluminum in the mylar sandwich dissolving are gonna die. you can inspect them before putting them together to avoid having them fail when assembled into the battery.

 

[docnjoj]

. the A123 pouches have tabs that you can solder to.

Not the 20 amp ones. One tab is alloy and the other is copper. Soldering is a fail here.
otherDoc

 

[dnmun]

the A123 pouches i worked on were soldered together. lotsa solder.

 

[johnrobholmes]

. the A123 pouches have tabs that you can solder to.

Not the 20 amp ones. One tab is alloy and the other is copper. Soldering is a fail here.
otherDoc

I soldered my pack just fine. Gotta have the right flux and filler to get the aluminum wetted :wink:

 

[Doctorbass]

If you want tofast test your A123 pouches with minimal heat and R loss at high current here is an exemple for you:

thi sis the jig I built for fast connection for testing all cells I have at 100A.

Cell remain under 40 Celsius and tab too!

http://www.endless-sphere.com/forums/viewtopic.php?f=14&t=42322&hilit=a123+test

Doc

 

[Hillhater]

Jack, Are these actually the A123 , 20Ahr pouches ?
..or some other brand / capacity ?