Spot Welding Copper Strips to 18650 Battery Cells

[spinningmagnets]

For a variety of reasons, I now believe that anyone who wants to connect copper ribbon to 18650 cell tips as an electrical bus (regardless of your reasons), should seriously consider resistance soldering.

The thicker the copper bus, the harder it is to spot-weld, but...copper thickness has no effect on resistance soldering. Electricity is passed through he two parts (the 18650 can, and the copper ribbon), and the solder gets hot, with that heat radiating outwards until the desired result happens (two seconds?). In fact, with resistance soldering, a thick copper bus would heat-sink the heat away from the joint as soon as the power to the RS unit is cut.

If this interests you, I started a thread a wile back to compile info on resistance soldering / RS.

viewtopic.php?f=14&t=88965
http://technitoys.com/diy-resistance-soldering-outfit/

[spinningmagnets]

Since this thread is specific to copper busses (I recommend nickel-plating), I just received a sample of copper sheets (I also added more listings as a reference).

wire diameter gauge and sheetmetal thickness gauge are two separate things!
the thicknesses listed below are what I could easily find available on the internet.

one-mil is 0.001-inch

0.15mm___6-mil__34 ga
0.20mm___8-mil__32 ga
0.25mm__10-mil__30 ga
0.33mm__13-mil__28 ga
0.40mm__16-mil__26 ga__12-oz per sq foot/B370 architectural 99% copper sheet
0.51mm__20-mil__24 ga__16-oz
0.64mm__25-mil__22 ga__20-oz
0.81mm__32-mil__20 ga__24-oz
1.02mm__40-mil__18 ga__32-oz
1.27mm__50-mil__16 ga
1.63mm__64-mil__14 ga

https://youtu.be/_EtZE7oQJaw



https://www.riogrande.com/Product/coppe ... oft/132126

6, 8, and 10-mil cuts with common scissors, 6-mil crumples easily, it feels like paper. I didn't even want any 6-mil, but it was cheaper to buy the sample pack that includes four types. At 0.40mm thick, its a toss-up between scissors and shears. Scissors could cut a sliver off of the edge, but it was hard, and I had little control in trying to cut a curve. If you need (or just want) 0.40mm, plan on ordering the busses to be laser-cut.

The four samples were from a DIY jewelry site, and look to be very pure copper by appearance. The architectural copper sheet was noticeably cheaper, but is only certified to be 99% pure copper. I suspect it will be conductive enough at 99% that this will not be an issue. However, it is made for the restoration of antique homes that used copper sheet in the roofing joints and rain-gutters. 16-mil / 0.40mm is the thinnest architectural copper sheet I could find. There are several thicker variants at http://www.onlinemetals.com

The common 18650-cell plastic holders have an 8mm wide slot for the common 7mm wide nickel bus-strips. If we limit ourselves to 7mm wide copper strips, then...the flat-bar to square-wire conversion is:

0.13mm X 7mm wide = 0.91 mm squared, area in cross-section
0.20mm X 7mm wide = 1.40
0.25mm X 7mm wide = 1.75
0.33mm X 7mm wide = 2.31
0.40mm X 7mm wide = 2.80

Converting "flat bar" area of cross-section to round cross-section copper wire equals =
http://www.engineeringtoolbox.com/awg-w ... d_731.html

7mm wide
ribbon______Wire
thickness___Gauge___Amps/Cont__Peak

0.13mm____18-ga
0.20mm____16-ga
0.25mm____15-ga
0.33mm____14 ga
0.40mm____13-ga

I have found conflicting info for wire-ga to amps on the internet, do not use these amp-ratings without doing your own research first.

[spinningmagnets]

I have wondered why nobody makes C41410, which is 99.5% copper with a trace of iron. It has very good conductivity, and the tiny amount of iron makes it spot-weld very easily (SW listed as "excellent").

Well, now I think I know why. Thanks to Luke / LFP, I just found C71500, it is 70% copper, 30% nickel, and 0.5% iron...and it is also listed as spot-welding "excellent"

http://www.farmerscopper.com/copper-nic ... 71500.html

CRAP! IACS is 5% for C71500

xxxxxxxxxxxxxxxxxx

C70600 is listed as 90% copper and 10% nickel, with a sprinkle of iron to add better ability to avoid stress-cracking. It's spot-welding ability is listed as "good", and the conductivity should be better than the C71500 listed above, due to a higher copper content. Plus, its listed as available in sheet form.

http://www.farmerscopper.com/copper-nic ... google.com

EDIT: CRAP!...several sources list C70600 electrical conductivity as 9% IACS, worse than pure Nickel...

xxxxxxxxxxxxxxxxxxxx

edit: for electrode holders, I have settled on C14500, which is pure copper with a tiny amount of tellurium added to make it easier to machine (drilling, etc). IACS 93%

[rojitor]

I succeeded spot welding pure copper with my arduino spot welder. No flux on 12v.



It totally defeated 0.05 and made acceptable welds at 0.1
I will try to increase the voltage and use flux.

[Willow]

The DN10 with custom timing circuit (thanks to Ridethelighting and his Uncle) and tig welding tips - 0.1 copper is a breeze.

Have the electrodes suspended from the ceiling for ease of handling.

copper.jpg (224.29 KiB) Viewed 4934 times

[vex_zg]

The DN10 with custom timing circuit (thanks to Ridethelighting and his Uncle) and tig welding tips - 0.1 copper is a breeze.

Have the electrodes suspended from the ceiling for ease of handling.

copper.jpg

Not to start any flame war... this looks visually appealing... but...
I can spot weld up to 0.2mm copper but I choose not to because the copper weld point it's weak mechanically, and the energy required to weld it is a bit high. 0.1mm copper I gave up immediately because it is really weak, welds points are weak (even perfect welds) it's, paper thin, and I would not trust it in my pack because of these weak mechanical properties of copper.

with copper I would go with either resistance welded soldered,pulse arc welded, or the combination of nickel spot welded to cell and then welded/soldered to copper bar.

[Willow]

...even attempting to weld 0.2 copper to an 18650 is really a big risk. I see you have had a lot of failures and quite surprised you have not seriously injured yourself or started a fire.

Never attempt to hold a pack together 'mechanically' with the electrical conduits - they should only ever carry electricity, and should do so with the least resistance possible.

Attempting to remove the copper strips will tear the copper and leave the welded/fused points stuck to the cell. If your welds are not sticking, then you are not doing it right.

[vex_zg]

...even attempting to weld 0.2 copper to an 18650 is really not wise. I see you have had a lot of failures and quite surprised you have not seriously injured yourself or started a fire.

Never attempt to hold a pack together 'mechanically' with the electrical conduits - they should only ever carry electricity, and should do so with the least resistance possible.

Attempting to remove the copper strips will tear the copper and leave the welded/fused points stuck to the cell. If your welds are not sticking, then you are not doing it right.

I designed, built and coded a 600Ws CD welder from scratch, it's 3F so aroudn 10V is enough for welding .2mm copper. No risk of injury or fire , I use all protection equipment and tested with old and mostly empty cells.

As I said, the risk in using copper is that a bit too high energy is needed for .2mm copper, since I found .1mm to be too thin and weak , and copper is so much less mechanically durable than nickel. But .1mm copper is really like thin plastic bag, I would not trust it to endure all vibrations it will take over year or two. Please share your experience when you are done and tested it for a while. Would be interested also how you encase the pack and keep it's structural integrity.

The best combo IMHO would be to have nickel welded to cell and then somehow connected to copper sheet or bus bar, that's best of both worlds.

[kdog]

I'm nearly there vex!
In my extremely limited available time, I've scraped together several hours and done this:

image.jpeg (151.21 KiB) Viewed 4884 times

These are .16mm copper punched with 8mm holes directly over the centre of terminal. I will through weld 8x.2mm ni strips to cover holes. Then I weld the nickel to the battery in the normal fashion. The .16cu is nice and strong and cheap, through welding is quick and easy with pincer style electrodes...it's just a bit more time consuming in punching holes ( very quick) and the extra welding.
It's looking promising so far...

[vex_zg]

I'm nearly there vex!
In my extremely limited available time, I've scraped together several hours and done this:

image.jpeg

These are .16mm copper punched with 8mm holes directly over the centre of terminal. I will through weld 8x.2mm ni strips to cover holes. Then I weld the nickel to the battery in the normal fashion. The .16cu is nice and strong and cheap, through welding is quick and easy with pincer style electrodes...it's just a bit more time consuming in punching holes ( very quick) and the extra welding.
It's looking promising so far...

yeah kdog I think your solution is probably the best of both worlds I've thought about it for some time now.

one modification though: I would also solder the connection of nickel and copper, copper sheet & nickel solder really easy even with my 60W soldering iron. I think it would decrease the copper-nikel connection resistance. In a pack I just built I used copper strips soldered here and there on some thin pathways where I could not get enough nickel.

how do you punch holes? Why not just drilling with a drill? How do you mark where they should be? This seems most tricky to me, marking exact location of the hole.

[kdog]

Punching the holes is way quicker and makes very clean holes. drilling thin materials usually results in a mess. I built a small hole punch and it takes about 3-5 sec per hole.
As for aligning the holes I used a very low tech trick. I cut the copper template out and trimmed to desired fit. Then I got the left over inside circle of the insulators (that you use on the positive end) and placed them upside down in the end of the batteries ( sticky side up). Placing the template carefully down on top I ended up with them all stuck in exactly the right spot. Then I traced and punched. A little slow but not too bad, and except for once, every time was spot on.

[spinningmagnets]

Snath made a steel jig to press dimples into the copper strip. Very easy to make something similar for a hole-punching jig. The baseplate and punches should be steel, but...I think the top-plate could be aluminum or wood/plastic cutting-board.

When using a bolt as a hole-punch, I would grind the threads off, and then drill into the tip a very small distance, to create a sharp edge. Copper is soft, so I doubt a hole-punch made from a hard steel bolt would ever need additional sharpening.

[vex_zg]

Punching the holes is way quicker and makes very clean holes. drilling thin materials usually results in a mess. I built a small hole punch and it takes about 3-5 sec per hole.
As for aligning the holes I used a very low tech trick. I cut the copper template out and trimmed to desired fit. Then I got the left over inside circle of the insulators (that you use on the positive end) and placed them upside down in the end of the batteries ( sticky side up). Placing the template carefully down on top I ended up with them all stuck in exactly the right spot. Then I traced and punched. A little slow but not too bad, and except for once, every time was spot on.

hah! sticky inside insulator circle! excellent idea, never would have thought of it.
Next pack I am building this way. With additional soldering of copper-nickel.

[kdog]

Just over a month ago I did these...

image.jpeg (120.36 KiB) Viewed 4684 times

Which I was really happy with. A really solid DIY method for providing a high current solution for series connects. The only draw back was that it took ages to do all the welding. It probly took me 20hrs to cut, shape, weld all the nickel to the copper ready to join to the battery.
I've been delayed ( another baby) so I have been waiting to finish my pack and build.
Two nights ago I picked up my un welded but glued 20s10p pack and it promptly snapped in half. I'd used the wrong type o hot glue. I had to take the whole pack apart. It was frighteningly easy. Preparing to reglue it I realised I won't be able to get it back together in exactly the same way so all these end plates are useless!!!
Nooooooo!
Gutted! But it's a blessing in disguise, it would have been a fire ball after several hundred k's. Also Fixed up several issues with the original layout. I just have to figure out how I'm going to weld it all up. Nickel is looking very user friendly right now.

[spinningmagnets]

I'm really starting to like this idea of using a copper sheet with holes over the cells. If someone wants to spot-weld nickel ribbon onto them (or brass?), it can easily be done. If you want to use resistance soldering, that's available. If someone wants to use a fuse-wire on either end of the cell, this basic pattern allows that too (fuse wire can be soldered or spot-welded)...

[ridethelightning]

I succeeded spot welding pure copper with my arduino spot welder. No flux on 12v.



It totally defeated 0.05 and made acceptable welds at 0.1
I will try to increase the voltage and use flux.

welds 10-14 on the second blade look like the right amount of heat. that may need to be re adjusted on the 18650s though, it will have different resistance.

[vex_zg]

I did a lot of experiments with welding copper, I have a 600Ws fully programmable spot welder, can weld copper including 0.2mm and probably thicker, but my conclusion is that for sufficient mechanical strength you need at least 0.2mm copper, but to weld 0.2mm copper you need to use power levels which are borderline with energies needed to damage the cell can.

Next issue is that the electrode wear is quite heavy at these high energies. Tested molybdenum, wolfram/tungsten, copper/tungsten alloy, but pure copper was the best. Haven't tested yet that zirconium /copper alloy AKA glidcop, but have them at home.

Another take I had is that cheap yellow brass I bought at aliexpress was much better in current carrying capacity than nickel, easy to weld, and better mechanically than copper.

My choice would be either: brass, 0.2mm nickel in layers (0.3mm nickel also needs a bit high energies), or copper somehow connected to nickel layer which has been welded to the cell. I liked what kdog was doing and would probably do something similar if I had time.

[spinningmagnets]

I am really liking the bus-plate style in the pic below (easy and cheap to make at home), and then the idea of using brass for the ribbon to make spot-welding easier is a great addition:

"Brass for spot-welding 18650 parallel cell groups?"
viewtopic.php?f=14&t=85059

[rojitor]

Looks very efficient and easy to do. I like it.

[rojitor]

I found this video today. I don't know if it has been discussed already. Up to 5mm thick. Looks amazing.

[spinningmagnets]

Thanks for that link. At 2:44, they list the benefits of being able to weld thick buses (up to 5mm)...and also specifically the high-conductivity materials, such as copper and aluminum.

[qwerkus]

Anyone tried to weld or solder 5*0.2mm nickel plated copper solar bus wire ? It's cheap, supposedly copper and already protected against corrosion and, with a cross sectional area of 1mm2, it should be able to handle 12A. I wonder how the plating would influence the welding process.

[rojitor]

i think that's not nickel plated but tinned. Good for solder, not so sure for welding.

[vex_zg]

Anyone tried to weld or solder 5*0.2mm nickel plated copper solar bus wire ? It's cheap, supposedly copper and already protected against corrosion and, with a cross sectional area of 1mm2, it should be able to handle 12A. I wonder how the plating would influence the welding process.

I just ordered, will report in few weeks.

[ridethelightning]

Anyone tried to weld or solder 5*0.2mm nickel plated copper solar bus wire ? It's cheap, supposedly copper and already protected against corrosion and, with a cross sectional area of 1mm2, it should be able to handle 12A. I wonder how the plating would influence the welding process.

I just ordered, will report in few weeks.

great. im interested to know what happens.
i suspect the tinned coating might sort of explode with welding, as the current will need to be very high to weld the copper.