A123 20Ah cells price down to 17.50$ per cells?

[mistercrash]

This is what I came up with but I just went with dimensions I know from what I found on the internet. This is just guess work for now but I would make those clamps more precisely if I had cells of my own. They would be easy to make by just getting bar stock of the right dimensions and then it's just a matter of cutting and making holes in the right places and tapping the right holes. Copper would be best but parts of the clamp have threads in them so copper is probably too soft. Anyway, here's what I have for now. Here are the approximate dimensions.

Top bar - 67 X 20 X 2 mm - 4 mm holes 56 mm apart deeply countersunk
Middle bar - 67 X 20 X 7 mm - 4 mm holes 56 mm apart and one hole in the middle threaded M4X7
Bottom bar - 67 X 20 X 3 mm - Holes threaded M4X7 56 mm apart
Screws - M4X12 mm flat head

So by looking at where the screws go, there is a way to put these clamps on. The threaded hole in the middle of the middle bar can receive a M4 screw to fasten connectors for the main leads and for the balancing leads.
The blue bars are any kind of insulating material placed between the clamps to prevent shorts. Wrap the whole thing in something sturdy and there ya go.

 

[docnjoj]

That is cool, mistercrash. I like the countersunk screw to avoid potential shorts. I was thinking of 1/8" coroplast for insulators. I am guessing you are using alloy for the parts. What about oxidation raising the resistance?
otherDoc

 

[dangerzone]

I don't KNOW they are made in Korea, and I don't know they are NOT made in USA. In fact, the trader insists they ARE made in USA and come from there. I find this extraordinarily unlikely. But it could be so.

I was told by someone who should probably know that all the 20Ah cells are actually made in Korea. But I have no way of knowing any of that.

In testing, we are seeing 18.5-19.5 AH which seem to be in line with their spec sheet and the American concept of almost meeting one. I have a little difficulty accurately measuring at those levels consistently. In groups of six, they seem to average better over 19 Ah and I suspect it is my measurement accuracy as problematical.

They will REALLY do 23C. We did a string of three in series at 475 amperes for OVER 30 seconds. I found it astounding.

My THEORY is as some of you have surmised, an Intellectual Property problem. As a result of Fisker shortfalls, A123 is abrogating manufacturer contracts in Asia, and the factories are simply selling them where they can then. Basically, they've lost control of their own battery manufacture.

The CEO of A123 has predicted the price of Lithium cells to drop in half by 2016. We have found his cells have dropped from $50 to $17 each since November so we think things are dropping faster than he apparently does.

We are doing an almost weekly attempt at a battery module from these cells as they have fallen through the $1 per AH level. I must say in all modesty that it is mostly comical. My packaging skills are truly embarrassing. But it's gotten to be a bit of a gig.

Jack Rickard
http://www.EVTV.me

Great info and great work Jack, good to see you here again.

So do you think this 17.40$ per cell for >500 pcs is gonna be the deal from now on..? I mean, why do they sell them as 'rejected cells', is that advertising or what..? I hope you're right and that there is no need to fear a shortage in supplies cause that cell is what I plan to buy next and the price seems right...

Do you know if that Chinese company sells BMSes, those plastic A123 battery packs and chargers too..?

 

[mistercrash]

That is cool, mistercrash. I like the countersunk screw to avoid potential shorts. I was thinking of 1/8" coroplast for insulators. I am guessing you are using alloy for the parts. What about oxidation raising the resistance?
otherDoc

Coroplast would probably work very well but personally I would go with something solid like some kind of plastic sheets, something really tough like Kydex. I just went with EVTV's idea of clamps for the cell tabs and made a drawing. I'll let someone else worry about oxidation of the alloys :lol:

 

[Doctorbass]

I doubt these little flat bar can keep contact with all the tab surface due to the bending..

maybe considering to make it thicker ?

Doc

 

[mistercrash]

There's not much room to make it much thicker. The holes are as close as possible to the tabs so the leverage effect making to top bar buckle would be as little as possible. Your top bar is in fact thicker in the middle but it will still bend where it is still at the same thickness as before and that is right at the screw. The design I made was to keep things as simple as possible so that anyone with a hack saw, a drill with bits and a tap can make their own clamps just by buying some bar stock at the hardware store or online. If you're worried about the top bar buckling then you can have it machined like this to match the thickness of the tabs precisely so the harder you turn that screw, the more the bar will buckle the opposite way putting more pressure on the tab. Actually, it wouldn't buckle at all because the tab is there to fill that gap. but the more pressure is on the screws, the more pressure is on the tabs.

 

[bigmoose]

Jack showed that the tabs were long enough that you could parallel the tabs of three cells in one clamp. Why not make your clamp blocks 2 pieces in stead of 3? You would be able to get more threads in the material. The two pieces would be thicker and less prone to barrel bulge as the good Doctor showed above. Size your fasteners so that you can torque them to 85% of yield and use loctite and they won't vibrate out.

 

[mistercrash]

Of course! Even simpler. Me likes simple, just like my mind. Let me get back to the drawing board.

 

[mistercrash]

So there. Same size bar stock is used. It's 7 mm in the drawing but it's kind of hard to find that exact thickness so I guess a 1/4'' would do the trick. The screws are M4X14 mm. Pure C101 oxygen free copper is best for conductivity but I could not find any in the right thickness. So it's C110 copper which is still copper but harder so it's better for putting threads in it. I found sheets of it at Onlinemetals.com. They have a cut fee and you can order all the little pieces precut for you for $48. The bars would be 0.25'' thick by 0.825" wide by 2.75' long. 34 in total. There is also 6101 aluminum that is good for conductivity but it doesn't seem much stronger or harder than C110 copper.

If you want to place an order :lol: http://www.onlinemetals.com/basket.cfm?id=3114054&action=delete

Now with a thickness of 1/4" on the bars, I don't think some M4 screws would be able to bend them, you'll strip the threads before you are able to do that.

 

[bigmoose]

OK, looks much better to me. The hole for the voltage tap in the center of the bar... how about making the bars two different lengths and moving the tap outside of the compression screws? Also The conductivity is from tab to tab, so for all intents and purposes the compression strips could be made of steel perhaps. For the copper compression blocks to pass current the path would have to be through the screws. I have been thinking about this for months and one additional idea I had was to cut one tab shorter and fold the other over the top of the short tab so that the folded tab had contact on both sides of the short tab. I think I would fold the copper tab over the aluminum one.

If I wanted to enhance the compression to the extreme, and I made the compression blocks from steel; I could surface grind a bit of a barrel roll to them so that the centers had perhaps 0.005 to 0.007 more interference and hence compression in the center. I like your design. One could use some 0.020 or thicker calendered nomex insulation between the blocks. That is the super tough stuff used as "slot paper" rewinding motors.

 

[mistercrash]

Now we're getting somewhere. Great suggestions moose. You're folding tab idea is something I would probably do but I wouldn't cut one tab, I'd leave them to full length, bring them together as best aligned as I could and fold both together and clamp the suckers real tight.

Here's the clamp.

And here's what it looks like on a 16S1P pack. Now the balancing connectors on the right could be facing inwards with the BMS on top of the pack somewhere in the middle I guess.

 

[caspo6]

What about a clamped system that uses copper or aluminum 19mm x 90mm x 6.4mm bars on the outside, and 19mm x 160mm x 7.25 non-conductive bars on the inside to compress the tabs. Cells would alternate positive & negative tabs, 3/16" bolts would have non-conductive sleeves & washers. Balance leads would be sandwiched with tabs.

Also, you could first cast the pack into the silicon rubber, and then mold a hard urethane case around it. That way if you ever had problems with a cell, you could still pop it apart.

Sketch shows a 12s1p pack.

Also found something similar here:
http://www.google.com/url?q=http://...twIwAw&usg=AFQjCNH4TBYLQ8nsAkPnHeZpx-Qvg249GQ

 

[ohzee]

I really like what you guys are doing. Keep up the good work.

To answer the question OP made earlier "no I do not have the resources or even the courage to do a group buy on these cells
like you are discussing. I very much appreciate your potential consideration of doing so and if it goes thru I plan on getting in"

Yea I am not responsible , but sorry for the harassment you have gotten for whatever the reason.. Seems misplaced.

Thanks for your hard work. It is appreciated.

 

[999zip999]

If you fold the tab I think there should Be some cruse in the tab for contact. What are the tabs made of I believe the pos. and neg have big differences yes.

 

[Spacey]

Can the tabs not be soldered and clamped? Is soldering these tabs a no go?

 

[VoKuS]

Where does one buy these cheap 20Ah A123 Cells?

Whats the min order?

I need bulk energy storage for my camping van, to run AC/Electric Heater with out a gas generator...

Would love to build a large pack, using these light weight packs, instead of something like lead acid...

 

[Harold in CR]

There will be a group buy coming up within the next week or so. Keep an eye on the For Sale NEW board in this forum.

What country are you in ?? Shipping to group members will be from south Florida.

 

[mistercrash]

Can the tabs not be soldered and clamped? Is soldering these tabs a no go?

I believe one tab is aluminum and the other is brass or copper. The aluminum is the problem for soldering. But with good stiff clamps, the problem would be solved and the contacts might actually be as good or maybe even better.

 

[dnmun]

these clamps you guys are making with the screws could have a third screw go right through the middle of the tab sandwich. use a drill bit the diameter of the hole and just drill through the tabs when they are locked in place, then run the third screw through that hole. should not reduce the current capability at all, removes the warping in the middle problem. these screws do not have to be large, #6 or #8 max. three #6 screws would create a lot of force.

 

[docnjoj]

That sounds like a good idea dnmun. I am still hoping that there is enough room to use a nut and bolt rather than tap the holes in the metal. We will see.
otherDoc

 

[dnmun]

i agree, a nut and normal flat head screw would be best. since the pouches would have the staggered terminals separated by a large space, a nut and flat head screw would work without fear of shorting between the pouches, 2 over in the series.

use copper as the central buss, and the outer clamp metal could be steel since it doesn't carry the current which would go through the middle. the copper bussbar in the middle just needs the sense wire soldered on anyway, not the big nut job that jack did.

not meant to diss mr jack either since i am totally in agreement with him about these endless fear idiots saying these are all reject cells. they have left a bad impression about the sphere for no reason. just their bigoted hatred of the chinese.

 

[IBScootn]

I like the new clamp idea, but I was thinking of a low tech solution. Simply take a copper bar with a thin slot in; a slot just wide enough for the two tabs being joined and for another thin angle piece of copper that is jammed into the slot to make a compression fitting. That latter jam piece could have a tab to bolt down to the copper bar - locking it in place.

Take some Noalox anti-oxidant and a emery cloth and buff the Al tabs until shiny, then wipe with dry cloth. Should leave a very thin film of Noalox to keep Al from oxidizing.

Just a thought

 

[mistercrash]

There's lots of room for improvements or different interpretations. Like using a nut instead of tapping the metal bar like docnjoj said, there would be enough room if you follow the idea of dnmun of putting a third screw and nut in the middle of the clamp. That means thinner metal bar can be used, thin enough to give enough room for screws and nuts. There's more than one way of doing it.

 

[mistercrash]

Well I took some ideas from many in here coupled with the original idea I saw on EVTV and here's what I would definitely do if I had A123 prismatic cells to join in series. Simple copper bar stock of 0.08" (2.03 mm) thick and 3/4" wide, cut to pieces 60 mm long with M4X12 mm bolts, M4 washers and ordinary M4 nuts with blue Locktite, no Nylocks. I found that using bolts instead of counter sunk flat head screws is much better if only one cell needs to be taken out. Ring terminals crimped and soldered on all wires. I think this would make a very solid pack.

 

[nieles]

I found that using bolts instead of counter sunk flat head screws is much better if only one cell needs to be taken out.

make sure you make slots instead of holes in the middle of the cell tabs, for the bolts to pass through. otherwise you can still not slide out one of the cells.

also use lock nuts, otherwise the connections could vibrate loose (they do take a little more space)