Welding to top of cell 21700 cell can rim (negative tab)

[ShockE]

Has anyone in the DIY space successfully welded nickel foil to the cell rim on the top of the cell after stripping away the insulation? I know most packs here are aircooled so this might not be needed, but I am working on a liquid cooled pack and am trying to determine the best way to weld a few prototype packs.

I know OEMs use laser welds usually in this application, but curious if anyone can point in me in the direction for someone who's tried with spot welders, or if any "prosumer" laser welders exist.

Looking forward to reading peoples thoughts on this.

Thanks

 

[mrmojorising]

From what I've seen, this is usually done with wire bonding, which is ultrasonic. You'd need a very precise laser to do that successfully

Have you considered alternative cells? For example, 4695s can be readily sourced and are much easier to design bottom cooling for as they have large positive and negative terminals on the top. The positive terminal is Aluminum, while the negative is nickel plated steel like standard 21700s, but you can find many suppliers pre-laser welding a nickel strip to the positive which allows these cells to be easily spot welded with a half-decent welder.

The downside to 4695s is they are large, limiting the shapes they will fit, and in order to get high amps they require cooling

 

[Balmorhea]

If liquid cooling is necessary then the pack is too small for the load. Liquid cooling adds size and weight to the too-small pack. It seems like it would be better to spend that same size and weight budget on more cells rather than abusing the cells and having to cool them. More cells give more range and longevity but liquid cooling only adds expense and complexity.

What corner case are you dealing with that might benefit from this elaboration? Maybe you could use a phase change potting material to do the same thing without having to relocate the negative tabs.

 

[sternwake]

I saw a video where a guy was trying to get a CID to pop on a lifepo4 26650 cell, and simply bridged the shoulder in a direct short.

It nearly instantly released some pressure when shorted, but this was likely not the CID tripping, but the fact that the insulation suspending the + CID and cathode bridge instantly melted and allowed the pressure to escape. The result would have been far different if it was not a shoulder short, in my opinion.

I built a 2pack that was getting too hot. The cells were. on paper, able to handle the max load of the ESC with some rom to spare.
The next 2pack I built had much better cells, and temperature was no longer a concern.

They also sagged far less and the throttle was extra spicy, and my range far higher.

 

[ShockE]

From what I've seen, this is usually done with wire bonding, which is ultrasonic. You'd need a very precise laser to do that successfully

Have you considered alternative cells? For example, 4695s can be readily sourced and are much easier to design bottom cooling for as they have large positive and negative terminals on the top. The positive terminal is Aluminum, while the negative is nickel plated steel like standard 21700s, but you can find many suppliers pre-laser welding a nickel strip to the positive which allows these cells to be easily spot welded with a half-decent welder.

The downside to 4695s is they are large, limiting the shapes they will fit, and in order to get high amps they require cooling

Wire bonding is used by Tesla but several OEMs laser weld foil instead of using wire bonds.

Sounds like there’s little in the DIY space to support this, was hoping someone tried though.

 

[mrmojorising]

Wire bonding is used by Tesla but several OEMs laser weld foil instead of using wire bonds.

Sounds like there’s little in the DIY space to support this, was hoping someone tried though.

Stark's varg also uses it on their 21700 pack (unsure if design has since changed), and I believe Alta redshift did too when they existed

 

[docw009]

Welding to the top of the cells at the edge of the can? It seems pretty risky. You want the probes to hit the rim of the cell perfectly underneath the nickel. You will likely miss and throw a huge spark when that happens, burning a hole on the rim.

Even if you make it perfectly every time, that throws a lot of heat into an area that wasn't intended to get hot. In addition, it places the negative nickel strip uncomfortaby close to the positive terminal of the cell, If it contacts while welding, that's a disaster.

Did you mean welding to the side of the 18650 or 21700 cell? One of the long time battery vets here, he rarely posts these days, warned against that with some good reasons. Google AI repeats his warnings, so I won't. I think it's unsafe,