Terminating the 15Ah and 20Ah Cell-Man cells

[peterperkins]

I'm starting to assemble my cell and wonder if insulating sheets between the cells in the stack is a good idea?

Is there a current/voltage leakage issue through the pouches?

I was thinking that very thin plastic A4 stuff used in files/folders. With cutouts so the alternate tabs can be soldered.

It will insulate the tabs from one another as well which will be required anyway.#

Also thinking about temperature sensing in the middle of a stack. It would be nice to put a thermistor in the middle, but a thermistor bead sandwiched between two cells will deform them creating an indentation and possible problems internally.

How about a thermistor on the copper tab. Will the heat conduction to the tab from inside the cell be suffcient to gauge the internal temp?

[bigmoose]

Look at Luke's (liveforphysics) thermal images of these cells under high discharge rates and you will see where the intrinsic hot spot of the cell is. That is where I would put my temperature measurement sensor.

[liveforphysics]

I'm starting to assemble my cell and wonder if insulating sheets between the cells in the stack is a good idea?
With any pouch cell it's a good idea, even something simple like a layer of wide double side tape is enough.


Is there a current/voltage leakage issue through the pouches?

Every pouch cell, Kokams, RC LiPo, and these A123's can and normally do develop some weak connection to the foil pouch over time, so if you've got them stacked touching other foil pouches and things, then you can get current leakage (the reason to insulate).

I was thinking that very thin plastic A4 stuff used in files/folders. With cutouts so the alternate tabs can be soldered.
That would work fine, but it would be slippery. Something adhesive would make for a strong pack.
It will insulate the tabs from one another as well which will be required anyway.#

Also thinking about temperature sensing in the middle of a stack. It would be nice to put a thermistor in the middle, but a thermistor bead sandwiched between two cells will deform them creating an indentation and possible problems internally. Like BigMoose said, check my thermal pics.

How about a thermistor on the copper tab. Will the heat conduction to the tab from inside the cell be suffcient to gauge the internal temp?

In my IR testing, the tab itself was kept cool by the large copper block it was clamped upon, so depending on your termination plan, I would say the tab isn't the best indicator of the cell temp.

More important than monitoring temps though, if you simple have a think sheet of aluminum between each cell to act as a heat spreader, you're never going to overheat a cell. I gave them the worst case scenario type testing under the IR gun with no heat spreader, and 80% of the cell stays pretty cool while one area gets warm, which is common with all pouch cells I've seen, and just distributing the heat from the hot area to the rest of the cell with a heat spreader is all they would ever need.

[cell_man]

I wouldn't worry about a heat spreader between the cells unless you are going to be pulling serious current like 150A per cell or more. You don't need to go overboard on the insulator if you decide to go with 1. Anything will do as it's only ever gonna be a small current leakage. The double sided tape as Luke suggests would work well for some termination methods but not for my mechanical termination as the cells need to be accurately lined up and moved around a bit to get it all fitted in place. When I use aluminium sheets between the cells I simply tape each cell before assembling. The packs are wrapped up pretty tight so I can't see the cells moving around much once it's been built.

[CroDriver]

Does someone know if a bitt softer cells will have different performance?

[dangerzone]

As was mentioned I'm kind of sold out of 20Ah cells. That's not strictly true, I have a customer that has already purchased a big quantity who is due to take most of the remaining 20Ah cells. With what's left I could probably do some smaller packs but not sure I'd have enough for a bigger pack. Situation on the 15Ah cells is similar but have a few more 15Ah cells to spare. There's no news on more batches of cells at the moment.

Any idea when u gonna have some more of those 20Ah cells..?

[cell_man]

Sorry no news I'm afraid and it's not looking good.

[peterperkins]

I've Just about finished the initial pack build with my 50 x 20ah A123 cellman cells and managed to get 50 cells into my Honda Insight mk1 OEM battery case for a nominal voltage of 165V x 20ah. Same voltage as in my Phev 40ah Insight.

The 20ah is a five fold increase in capacity over the stock usable 4ah. And it fits in the same space and weighs about the same as the oem nimh pack.

A few pics.











The blue wires are temp thermistors buried in the pack.

The yellow tabs are safety covers over the power connections points. The power connections are not yet completed.

The pack is split in half 2 x 82.8V

The pack negative connection is at the left hidden, and the positive is on far right.
Two pack split point connections in middle.

It's all contact adhesive stuck together so i hope it all works the cells are a solid lump but can be split by peeling them apart. I certainly don't want to take it to bits!! I soldered the tabs using an 80W soldering iron some spray flux and standard solder. i used a sheet of self adhesive plastic sheet between each cell to negate any issues with voltage leakage from cell to cell via the pouch. The sheet was cut so it also folded over the tabs once the joint had been made as each cell was added to the stack.

The pack will also be strapped with 300kg plastic banding to limit cell expansion.

Still lots to do I'll post more pics later.

As i was assembling the pack i found one of the Cellman cells was strangely flexible so i put that to one side. All the others are very rigid. It has a normal voltage so a bit strange.

[RoughRider]

look realy nice....

[cell_man]

Good work Peter, it looks really nice. Don't know how I missed your post previously. How long would you estimate it took to put it together?

Doesn't sound like you had a bad time soldering them, I found them very difficult. It seems fairly common practice to spot weld a nickel tab to the alloy tab of pouch cells but I think it requires a special spot welder. If that's done it's very simple to solder them together. If I'm able to get some more cells I'll see if I can get them prepared in this way. The mechanical method is pretty good but it is a lot of work.

[peterperkins]

Probably took about 10-12hrs to assemble/solder the stack of 50.

I've also strapped my cell block to give it some compression with plastic packing strapping 300kg x 3 breaking strain. I've tensioned it to give some mild compression at present to allow room for slight expanison which will also increase the compression hopefully into the optimum range for these cells. The strapping should also help to prevent the cell expansion damaging the case. I used the same strapping technique for my old TS 200ah cells years ago in my van. that worked fine.

The soldering of the aluminuim tab needs 4 things. The copper just needs a bit of light sanding.

1) Power fileing/sanding of the aluminium tab to expose fresh surface. I prepared 5mm of the top edge on each tab.

2) Some spray flux

3) Standard multicore solder

4) A minimum 80W soldering iron with a decent flat bit.

Remember my joints only have to cope with 100A max.

To make the high power connections for cables to the tabs I soldered on some copper connectors which then allowed a simple bolt connection between the cable and the copper connector.

I'll post a few more pics when i have the connection board installed

[nerk]

Hello all, i have a few Questions I'm sure you guys get this a lot, theirs a lot of noobs out there like me. Anyway i'm looking into building a 20aH 72 volt battery pack and i'm trying to find the best way to go about it. I think ideally we'd all like to use LiFePO4 batteries but their so expensive and until price drops a bit i don't wanna drop 300 bucks on the cells, So then i looked into NiMH which are pretty cheap, then i was thinking to myself i'll just look into mass produced items that use decent batteries, makes cents to me that the price would be lower on such items, I found that Roomba replacement batteries are quite cheap they have iRobot Roomba APC NiMH 14.4V 3.5 aH batteries for around 25 bucks with shipping and they weigh some were around 25oz in the plastic housing they come in. Thats one option i'm looking into, Also this next one is a little far fetched but since i'm so cheap i thought it would be cool to see others prespective, So the iphone you can buy 3g batteries for 1.00 on ebay lol and lots for even cheaper, they are Lipo 3.7v at 1.3aH, and for a dollar how can you beat that? even if i had to buy 100 of these it seems like a cheap way to throw some packs together?

Please enlighten me with all your awesome battery knowledge!!

[slayer]

Well first of all i went to check for the prices on ebay and you are right i saw some at that price ... its a great find. you say in your post you don t want to spend 300.00 bucks...humm

now to build 72 volts 20 ah you would need 300 cells, and if you are a noob it will be very difficult, also if you have to buy the cells from one of the seller i checked he is asking 1.79 for shipping so almost bring your cost to 900.00 then he says he is not reponsible for the coston and duty charges ...ouch i paid some charges like that that almost double my prices on the order...ok now you have to put them together ....

i could continue but i think you get the point...

[patrickza]

they are Lipo 3.7v at 1.3aH, and for a dollar how can you beat that?

Hmm, that works out to 4.81wh per dollar or $0.21/wh. Good price but not all lipo is equal. It might have an exagerated ah rating, lousy C rating, or a lousy number of cycles. Best bet would be to buy one or two and hook them up to a cell tester and get some real figures.

[cell_man]

they are Lipo 3.7v at 1.3aH, and for a dollar how can you beat that?

Hmm, that works out to 4.81wh per dollar or $0.21/wh. Good price but not all lipo is equal. It might have an exagerated ah rating, lousy C rating, or a lousy number of cycles. Best bet would be to buy one or two and hook them up to a cell tester and get some real figures.

AND on top of that you have to actually terminate the cells into a pack. 300 might not sound a lot when you say it quickly but I can assure you if you had to terminate 300 cells that were likely never meant to be terminated in such a way, it will take a looooong time....

[amberwolf]

Just *testing and sorting* them will take a long time. Ask me how I know.

[peterperkins]

I now have my cells installed in my MK1 Honda Insight and working ok. I soldered all the tabs as mentioned a few posts ago. I'm pretty happy so far with initial testing they seem to have good power and low sag at my 100A max drain.

http://www.insightcentral.net/forums/mo ... stall.html

I'm glad at this stage that I jumped in and bought some when i did

[Nowar]

Hallo,
I read a lot on your board. Now I want to show up.
I have a TWIKE. The NCs are 8 years old. Now I ordered at my local e-bike seller LIONS from BHW which is the best you can get hear, affordable. I was pondering pouches but they are so difficult to get. I would have piled them with two 1mm spreaders. Separated with foil. Slightly squeezed with springs. For the connectors, I would solder one end and clip the other with the Molex. My BMS will be a simple 120 position switch with a musical VCO charge checker. The rest is done by the Twike. For winter insulation I want to use the new Aerogel-Mats. But now I use the cylindrical 38120S.
Maybe in some years I´l do it flat.
So keep charging against the oil.

[dangerzone]

Good work Peter, it looks really nice. Don't know how I missed your post previously. How long would you estimate it took to put it together?

Doesn't sound like you had a bad time soldering them, I found them very difficult. It seems fairly common practice to spot weld a nickel tab to the alloy tab of pouch cells but I think it requires a special spot welder. If that's done it's very simple to solder them together. If I'm able to get some more cells I'll see if I can get them prepared in this way. The mechanical method is pretty good but it is a lot of work.

Is there any chance to buy them prepared that way with a BMS, packed and ready to install in the near future maybe..?

[jimw1960]

This thread has been dead awhile, but with the recent availability of a123 prismatic cells, I thought it worth resurrecting. I'm trying to build a 16s (48v) pack out of these cells and am thankful to this thread for the learning experience. In regards to terminating these prismatic cells, one thing that occurred to me is that prismatic cells are in widespread use for lipo and low-c lifepo4 packs like ping. So, why reinvent the wheel? Looking at the previous threads on how those packs are built, it looks like the proven cell termination method is to use a solder connection to a simple fiberboard with copper substrate. Since these are in widespread use, my question is whether anyone knows where one could buy circuit boards like they use to make lipo packs.

[MitchJi]

Hi,

This thread has been dead awhile, but with the recent availability of a123 prismatic cells...

Where are they available and what's the price?

In regards to terminating these prismatic cells, one thing that occurred to me is that prismatic cells are in widespread use for lipo and low-c lifepo4 packs like ping. So, why reinvent the wheel? Looking at the previous threads on how those packs are built, it looks like the proven cell termination method is to use a solder connection to a simple fiberboard with copper substrate. Since these are in widespread use, my question is whether anyone knows where one could buy circuit boards like they use to make lipo packs.

Asking Ping seems like a good place to start.

Would clamping or bolting to the circuit boards work equally well?

[jimw1960]

Hi,

Where are they available and what's the price?

For one, chinese reseller a123rc.com has them for $51 per 20ah.

Asking Ping seems like a good place to start.

good idea maybe he'd sell the blank fiberboards.

Would clamping or bolting to the circuit boards work equally well?

Perhaps, but I'd like to try the proven method first

[olaf-lampe]

The battery-tabs from HK Lipos and in Pings LiFePO packs are much smaller and thinner than these a123 tabs. Much easier to solder, but still a risk for unexperienced DIYers. For the same reason, the pcbs Ping could offer wouldn't fit for a123. They need to be tailormade. No big deal, but something to consider.
-Olaf

[jimw1960]

I wonder where I could get something like that made.

[wojtek]

just made a quick experiment - i think this idea has potential..

1. the ring terminal is too big
2. i think i could go with just one of them - two will increase the touching surface
3. will use m3 screws [this one is m4 - too big]
4. with smaller ring terminal, i might not have to use extra grass washers
5. will obviously custom cut the screw..

so it will all be smaller and lighter...

what do you guys think?