Need some advice recovering some batteries

[pdarnett]

Not got to spot welding yet!

Update on the battery, got my charger and following Dave's advice slowly upped the cell count to 6S and managed to get it charged up to that ok (22.2v in about 10 minutes).

Cells are still unbalanced after leaving for a bit, so on the Life setting (3.3.v) doing the cells individually then will do a full Lipo charge on each cell.

As its stands

Cell1: 2.2v
Cell2: 1.8v
Cell3: 2.4v
Cell4: 2.56v
Cell5: 2.18v
Cell6: 2.8v
Cell7: 2.14v
Cell8: 2.88v
Cell9: 2.3v
Cell10: 2.4v

So a bit better than 0.5v a cell, so I'm going to let it sit overnight then repeat etc etc etc.....

The other pack with the 0v cell is a scrapper, so I may do some testing on it (maybe salvage some cells for the other one) and use the BMS and the case for some 18650 cells I seem to have a lot of!

 

[Deleted member 4366]

That looks promising so far. It only takes a small amount of charge to go from .5v to 3.0v, so i'd expect there to be quite a lot of variation in the cell voltages at this point. Now, charge the single cells to about 3.3v, and then the BMS should allow normal charging.

 

[camerart]

Are your cells the protected type or un-protected? i,e, do they have the over-under protection circuit on the +end.
Some of mine are, so I doubt I could spot weld them.

I have made a balancing cradle so I can connect and charge 3 cells to my charger.

Camerart.

 

[pdarnett]

My 18650 cells are all the unprotected type, with tabs already on. My plan to solder them is to rough up the tabs, then very quick, high temp soldering.
Protection wise I'll be using a BMS for the final pack.

 

[camerart]

I think the weaker cells in a pack will dictate the length of you bike ride, so it's best to have all the same if poss, so after you got any going probably best to match like with like.

I think temperature fuses might be a good idea, there should be some in the packs you've got, they should stop fires.

If new cells are bought they will need spot welding. I had a look at cells where the weld has been snapped off, and there are 4 dots, where they were welded. I have an arc welder, I'll try some tests to see if I can do some gentle spot welding. Obviously I'll need strips, so I'll have a look for some.

Alternatively, I was looking into making a pack where the cells are in springs, like e,g, a radio. It might be impractical, if I can't guarantee good contacts.

Cheers, Camerart.

 

[pdarnett]

Don't do springs, they're not up to the current. If you're dead set on holders, look for something like this http://www.conrad-electronic.co.uk/medias/global/ce/6000_6999/6500/6510/6513/651327_LB_00_FB.EPS_1000.jpg

ps my bms has a temp sensor, which I'll bury in the pack.

 

[camerart]

I did doubt normal springs. I'm thinking about replacing cells, but I suppose cutting out and re-welding is probably the best bet.

Cheers, camerart

 

[pdarnett]

My advice would be do small packs. Mine will be built from "blocks" of 4 parallel cells. I'm going to build a 36v 9Ah pack, 10 blocks of these 4 packs. A lot easier to fault find as well. Each block will be soldered tab to tab then the whole thing heatshrinked together with some output leads.

 

[camerart]

Today I soldered a NI-MN cell into an 8 pack (repair) and it got so hot, that the cover melted, and I thought I had shorted it. I am careful about the solder time. This has never happened before. I tried out the pack later, and it appears ok. Mystery!

I was talking about testing a spot welding idea, but soldering is easier. Is it just as easy to buy tabbed cells as tab free?

Do you have a shareable source for cells? If yes, a Link would be appreciated.

 

[pdarnett]

All mine are from a job lot of Dell cells, mostly samsung ones, so I've been dismantling packs and being very careful not to rip the strips off, leaving a solderable tab on each end (very sharp though). I reckon I've got enough for 2 36v 9Ah packs. hence me building them in fours, if one block is sagging, then I can swap it out easily. I'll post some pics once I'm done with reviving the prismatic pack. have a look on endless sphere for DrkAngel, he's the king of laptop cell packs.

Plus sometimes some bargains come up under this ebay category http://www.ebay.co.uk/sch/i.html?_odkw=bulk+laptop+battery&_sac=1&_osacat=40005&_from=R40&_trksid=p2045573.m570.l1313.TR10.TRC0.A0.Xlaptop+battery&_nkw=laptop+battery&_sacat=40005

 

[camerart]

How do you think the best recovered cells will compare with new ones?

I'll watch your progress with interest. When do you think you will be testing them on a bike?

 

[pdarnett]

Not testing for a bit, once I've done the big pack I'll build some packs of the 18650's. Going to use the dead battery's case and BMS for the 18650's.

 

[Deleted member 4366]

How do you think the best recovered cells will compare with new ones?

A lot of the cells you reclaim will have most or all of their capacity. The problem is that most of them are older spec, so between 1.5 and 2.0 aH, and low C-rate, which makes a relatively heavy battery.

 

[camerart]

ok, so good for the price but not the best.

I was thinking about your 4packs (not your stomach) and wondered if it might be better to do strings of 12V X3 then double and quad them for the current, then each cell could be isolated.

 

[Deleted member 4366]

You have to sort them into their designations, then look up the datasheet to see the capacity, then test them to see that they have that capacity, then test them. Next, arrange them into strings that will give the current you want, and finally join the strings in series to get the voltage you want.

The best way to test them is to make a jig that can hold several in parallel. I made one out of MDF for 6 cells. I nicked the springs and contacts from an old Nicad/Nimh charger. Alternatively, you can simply solder up a 6S1P pack with power cables and balance wires, but it'll take a lot longer, and there's a chance of damaging the cells by repeated soldering. You can then charge a group of cells as a 6S1P, then do a discharge test and observe if any fall off in voltage prematurely. If you charge them in parallel, you can't see any duff cells because the good ones will support them.

The whole exercise of testing and sorting takes a long time. It's not a good idea to cheat on it because if one weak cell developed an internal short, though low probability, it could be catastrophic.

My test jig looked like this:

 

[pdarnett]

For me I've kept them in the same groups they came out of the packs in. Then made up the packs of 4, keeping the same make and model together. Prior to that I've individually charged each one and let stand for a day then re measured.

Like daves jig though! The mdf may come out this week for one of my own!

Not sure about your avatar though...

 

[camerart]

With an avatar like that who needs batteries..

I just made a 10 minute spot welder!!

Two bits of spring from an old tape measure. The wires are only speaker wires and of course they got a bit warm. It's a start anyway.

 

[Alan Quay]

With an avatar like that who needs batteries..

I just made a 10 minute spot welder!!

Two bits of spring from an old tape measure. The wires are only speaker wires and of course they got a bit warm. It's a start anyway.

Loving your work sir! A spot welder made from wood, who'd a thunk it!

Excellent job dude!

 

[Deleted member 4366]

... but your electrodes are the wrong way round. They should be side by side like this:
http://www.avdweb.nl/tech-tips/spot-welder.html

 

[camerart]

Ok, d8veh, I'll move them round and add thicker wires.

Thanks, Camerart.