Spot welding Copper / Copper Nickel / Copper Stainless Steel Sandwich

If these are the same (or similar) to the dummy 18650 cans that 18650batterystore sells, then people should be aware that they weld TOTALLY differently to a real cell.

They are not suitable for dialing in weld settings that you actually intend to use on real cells, in my opinion.

If you just plan to use them as a substrate for messing around with welding techniques, and you will dial in your weld settings more accurately when you move to real cells, then by all means these should work. Just want people to be aware!

3 Likes

Have you found any material that more accurately represents welding to real cans?

No, but honestly I have not looked that hard. I’ve always have spare P42A’s on hand :man_shrugging:

1 Like

Nelvick sent me 15 cells for testing spot weld resistance.

I’ll get on it soon, I promise.

I recently rebuilt a 7S2P DMEGC20P prebuilt battery, with 0.2mm copper under 0.1mm stainless.

I screwed up the busbar between PGroup 6&7, and was able to use a 4mm sharp low angle chisel to wiggle pry lift the stainless steel off of the copper surprisingly easily, and cleanly, and then weld more 0.2mm copper on top of the 0.2mm copper.

Then, made the same mistake again.

I decided that both 0.2mm copper layers, under 0.1mm stainless, needed to come off.

I was impressed with the wide flat width of that lowest layer of copper strip.

No more chisel prywiggle possible.

I had to use dremel and grind the copper off, and it was still well attached to cell between the welds, dare I say flux soldered?

Yes, I am using 3 pair of welds on an 18650 Cathode even though this pack will likely see no more than 8 amps per cell in its intended application, and also added a second fishpaper ring.

I did have one alarming episode of an unintentional welder double fire.

But my foot was still slowly lifting from the pedal.

I started being very deliberate in lifting my foot instantly…a quick stab of the foot, and did not experience a second double unintentional pulse.

For reference

I am welding 0.2mm under 0.1mm stainless, using Nelvick’s Flux.

AwithZ P20B

Gear 389

0.10 ms preheat

4 MS intermittent(delay before first weld pulse)

Double pulse

0.2 second interval between weld pulses.

The relative ease with which the 0.1mm stainless could be chiselprywigglelifted from 0.2mm copper below was interesting.

I used the same welder settings as above to weld more 0.2mm copper on top of the destainlessed 0.2mm copper.

That first layer of copper strip seemed to be pressed extra flat and wide to the cell, similar to the 0.5mm Stainless experiments.

2 Likes

I’m still curious to hear people experimenting and finding solutions to welding copper reliably WITHOUT flux. Anyone with one of these modern welders working on this?

Until then, I’m sticking with pure nickel and my Malectrics

I have welded 0.2mm copper without flux with my p20B, but had to use more than twice the power as using flux below, and stainless above.

I was not impressed with the tear off strength of 0.2mm copper alone even with strong welds, and moved away from raw dog copperification experiments.

When i finally get around to using the big boi for 0.3 i will tag you bb xox

1 Like

Agreed!

Thought they could be a decent ā€œbaseā€ for welding different metal combos onto for resistance testing. Especially since the top contacts are a separate piece from the can making shipping a lot easier.

2 Likes

Im messing around here and there to find something. Did a bunch of testing with pre tinning nickle and… well tin to the copper to try and get as reliable results and haven’t found the magic yet. Welding copper is difficult but not impossible in other welding types and the requirements are pretty specific: fast hot clean welds of short duration - having only the gear as a parameter on the awithz welders is a bit difficult because of the obfuscated parameters but not necessarily impossible. It’s just takes a lot of careful experimentation and good note taking so my progress is slow.

I do think an electroplated nickel on copper strip is going to be the easiest option but sourcing it when nickle copper is also an alloy that’s available has made communication through a language barrier hitting some dead ends - product available is actually not what i wanted :cry:

The significant wear of the welding tips when doing steel or stainless steel sandwich over copper has me continuing to look for that perfect solution. Im going to cad up some single cell ss sandwich pre cut medallion type sandwich pieces to cut and send them around to get laser cut. Just having a million easy cheap sandwich pieces with alignment markings would be nice and hlep simplify sandwiching over copper i think.

Also ya: where nickle is sufficient it really is still the best cleanest option. I still think the hassle of copper isn’t worth it unless it is the best option.

As an example the asymmetrical low profile high output pack i made for @MichaelWA ’s white ember build could have been done with nickel - but the compromise would have required more robust series connections to handle the current and make sure i had the cross section of nickel capacity required - this would have meant adding about 3-5mm more thickness to accommodate bringing the current path up and over the tops and bottoms of the cells on each series connection to use the wider nickle cross section in that orientation and then adding conductors bridging each of those. With the copper i just had to make the copper bridge between the p groups.

To get 250a cont in nickle i would either need to make a thicker strip than i wanted with nickel or add conductors like bussing. The width of the smaller sections of the enclosure wouldn’t take the extra as it a tight fit as is. Going over the top with more robust series connections was going to make it pretty crowded up top reqireing a small additional gasket. and splitting it between the top and bottom would have required a bigger gasket.

Pick your poison - difficult battery method or add gasketing and reduce ground clearance on a low scraping build. Any reasonable build would have done nickle but i don’t build reasonable things :laughing: i make the crazy stuff

3 Likes

Im not finding excessive wear of the tips welding 0.1mm SS sandwich over 0.2mm copper.
I’d think welding 0.3mm copper without the S steel would require much higher power levels, which does seem to wear the tips faster.

The replacement tips in Nelvick’s store are now 7$.

Ive been using an Arkansas sharpening stone to clean and shape the tips.

The aggressive diamond file the welder came with has to accelerate tip wear in two methods. It removes a lot each pass, and leaves the tip rough, concentrating the weld energy into less space on the tip which then melts easier and sticks to strip.

I find the sharpening stone easier to use and keep consistent tip shape.

I get very little to no wear directly on copper and don’t need to sharpen or shape the tips when welding around gear 4-600 on the p60 except when doing sandwich tests or welding nickel. Interesting. Going to add a onthe list of things to look into :laughing:

You moved away from raw copper experiments… but what about testing your stainless-over-copper techniques without flux?

The flux is the big concern for me personally. I don’t really trust it, nor do I see it as a viable part of a battery production process. That’s just me though, not looking to debate the merits of the flux. Just curious to hear people’s experiences without the flux.

1 Like

Electroplating nickel onto copper sheets can be done relatively cheaply and easily at home with some hardware store chemicals and a lab power supply, iirc.

Perhaps someone like @jack.luis could make a business out of plating copper sheets and cutting custom shapes on his laser :man_shrugging: just an idea.

I will look into that. Either way I should start selling laser cut copper since that’s a pretty much solved problem on my machine. Still working on optimizing for 0.2mm nickel.

Edit:

I think I got nickel figured out

2 Likes

.2mm copper under .1mm ss w/ no flux. Potted in 60d polyurethane.

I have built a few small batteries with 0.2mm copper under 0.1mm, and 0.07mm stainless, no flux.

I was using gear 550 on my machine( AWZ p20B)

I didn’t find any difference between 0.07mm SS and 0.1mm stainless.

I was moving up towards gear 550 until both welds seemed equal strength and size which is how I determine minimum power. for example gear 510 had one nice solid weld and the second weld tore the copper, but the patch of copper left on the cell left was half or less the size of the good weld. Gear 530 was 2/3 to 3/4 the size, and gear 550 was nearly exactly equal size.

Same ā€˜up the gear by 20’ process and try again, with 0.2mm copper under 0.1mm stainless but using Flux, yielded gear 320 on some cells but gear 385 on some others.

Since then, I just use gear 385 and flux and everything has been getting 0.2 copper under 0.1 stainless even if it is overkill for the application, and perhaps only 320 was required for the cell in question.

Both Glitter and AwithZ say that copper thicker than 0.2mm needs the flux, but @Dinnye has welded 0.3mm under nickel plated steel without it.

I am never using NPS again as stainless seems so much more welder and electrode friendly.

I punctured a few ā€˜Wintonic’ 18650s from a laptop battery pull, when trying fluxless stainless 0.2mm copper and am a bit shy of repeating that experience. It is hard to file the steel from the copper alloy electrode when that happens, and the sickly sweet smell of electrolyte is unpleasant.

I take electrode to the garage and the dremel with diamond cut off wheel as the file takes forever, and the punctured cell gets put into a plastic baggie with gravity in mind.

I do like welding the stainless to the copper using gear 008, all other settings the same, before positioning the sandwich on the cells for the Gear 385 pulse.

Adding a dab of flux to the cells before placint the sandwich on the cells is pretty painless to my process, but I am not building huge batteries nor view time as money, I just want low resistance and high weld strength for my own builds, and build super slow anyway.

I am of the philosophy that the lowest amount of energy, and thus heating of the cell, that can be used when spot welding, should be used, and the flux and the stainless significantly reduce the energy required, so I will use them.

I have no idea if there is any significant difference to ultimate cycle life of the battery, but part of the reason I want to use less power is I don’t want to be maxing out the machine every weld, and using 38.5% of its power has to be better for it, than using 55% every weld, or needing 85% of the power required to do the same job with No Stainless and no Flux.

This is obviously less of a concern with the higher power welders like the P30C P60C and P120D

Lesser gear pulses require less time for the 15 amp 9 volt power supply of my P20B to recharge the capacitors.

Nelvick sent me some 0.3mm and 0.2mm laser cut copper and stainless that have the slots to weld across. I need to experiment with that, and see how much less energy is required when slots are present.

I think it would be good to have little 10 gram sample size syringes of Nelvick’s welding flux for those who are flux curious or naysayers so they can judge for themselves whether they want to adopt its use.

1 Like

First,
Thank you for the AMAZING job all you guys are doing. This thread is probably the BEST resource in all internet for this experiments. Having said that, If anyone want to experiment with the flux and make public findings, good or bad (just the true) send me a PM and i will donate the flux.

3 Likes

It is near that time when I start welding some samples to send to Mooch to test the resistance of spot welded strips and try and answer some questions we have.

Nelvick donated 15 cells to the cause, and 6 of those are in front of me, and ready to go.

I am still deciding on what to weld, to answer the questions we have.

I am personally interested in seeing whether the flux changes the resistance of the spotwelded strip.

It tins the copper in immediate proximity to the weld, and when a lot of power is used, more than is really required for adequate strength tear off tests, the flux can tin the whole area between welds and surrounding area.

I theorize this fills the small airgap between cell and strip that forms between and adjacent to the welds, and thus reduces resistance between cell and strip.

I am thinking I can weld two strips, one pair of welds on each strip, with tabs opposed to each other on the same side of cell. One strip uses flux, the other does not. I got the settings and technique pretty dialed in on the 0.2mm copper under 0.1mm stainless.

Raw dogging copper, no sandwich, no flux, means I need to reestablish the welder settings before making a sample where the result can be relied upon to be comparable.

I am willing to do this but not start off sample making with the raw dogging of copper.

I am pretty sure My welder can NOT raw dog 0.3mm copper, so I would be limited to 0.1, 0.15, 0.2mm and maybe 0.1 and 0.15mm stacked for 0.25, but it seems no one is using 0.25mm copper so perhaps no point, and the stacking of strips has to be a variable.

I am also curious about the 0.5mm stainless experiments. Copper welded under this 0.5mm stainless with flux between cell and copper, seemed to be pressed wide and flat to the cell. The tear off strength required was enormous, and the copper between the welds appeared to be soldered to the cell itself.

I noticed this same tendency when I welded 0.2mm under 0.1mm stainless( with flux) , removed the stainless, and welded more 0.2mm copper, 0.1mm stainless sandwich to that original 0.2mm Strip. When I tore off all three layers. That copper on bottom was wide and flat and adhered to cell as if soldered between the welds.

Tearing off just 0.2mm copper under 0.1mm stainless with flux, the copper does not appear to be attached strongly anywhere but the welds, but the 0.5mm stainless seemed to press and hold the copper flat and wide and perhaps even solder the the copper to cell between the welds.

If this is happening then the resistance between cell and strip should be lower and measurable with the quality instruments Mooch has.

Some of these 6 cells look brand new, and all are at ~3.6 voltage. All but one are Ampace inr22/71/ JP40’s/Am04.

The other 9 Donated cells need a little prepwork and are set aside for now.

Mooch, do you have any designs on the functionality of these cells, like torture testing, or are they destined for recycling soon after testing?

Should I just drain them all the way, or leave them at ~3.56v before mailing them to you?

I’d like to get at least one sample welded up each day and send no less than 6 at a time.

Input as to what others want to see tested are appreciated.

This is a great opportunity, and I want to not waste it, or Mooch’s time.

On another Note back in August had a test weld cell, a Samsung 50G, that I left outside in the heat and humidity, inside a plastic case. I had used flux and had cleaned flux with rubbing alcohol which diluted the flux and allowed it to creep around cell below the shrinkwrap.

There are no signs of corrosion anywhere. Could be it never got wet enough or that the flux is simply non corrosive, or not enough time has passed. I would not want my Esk8 battery subjected to the conditions this cell was subjected to, but it is not like I had it in the dirt and direct sun and got it wet daily for 6 months. it was just outside in a plastic case in some bushes and forgotten about.

3 Likes

Sounds like a solid start. Depending on what turns up we can come up with follow up tests to check specific things. Let me know what thicker copper combinations you want and ill fab them up For comparison on round 2. Once mooch chimes in about preferred charge state ill make some head to head raw copper vs flux vs ss sandwich in 2,4,6 welds. Maybe @BenjaminF is interested in offloading some old cells as we are pretty local to each other?

3 Likes

If the copper is welding itself to the battery between welds it seems like it could be dangerous if your welding between the no weld zone on the negative cap