"Copper/nickel sandwich" buses for series connections

[chuyskywalker]

Having a devil of a time getting this method to work for me. Using a kweld. Tried plain nickel on copper, split nickel, slotted nickel, and even just split copper, but nothing seems to get the weld to hold on the battery. Went everywhere from 50J up to 120J without success.

I suspect my battery just isn't strong enough to actually make it happen (lipo 3s 50c 5500mah). I see other posts about this working when delivering 1200-1400A, but my readout only shows ~1000A on each weld. Also, perhaps my materials are too thick (.2mm copper, .15nickel).

Gonna get pick up thinner copper and a higher C rate lipo to see if I can get this to go.

 

[TrotterBob]

If only the LiPo manufacturers could post honest C ratings rather than '150C BURST CAPABLE!!!!!'. Maybe we should shame them more often on social media. There are meters out there that can accurately measure C ratings and internal resistance.

https://www.rcgroups.com/forums/showthread.php?980290-Cell-Internal-Resistance-Meter-more-useful-than-C-ratings-Now-available

 

[chuyskywalker]

Gonna get pick up thinner copper and a higher C rate lipo to see if I can get this to go.

Yup, moved to this and .1mm copper and I can get solid welds now at 60J and it shows I'm hitting 1300-1400A. Will try the .2 later.

 

[spinningmagnets]

Good info. Thanks for posting that

 

[nalu808]

Trying to find a welder under 200. Any out there that can do the copper nickle sandwich method?

 

[Mourningstar]

Having a devil of a time getting this method to work for me. Using a kweld. Tried plain nickel on copper, split nickel, slotted nickel, and even just split copper, but nothing seems to get the weld to hold on the battery. Went everywhere from 50J up to 120J without success.

I suspect my battery just isn't strong enough to actually make it happen (lipo 3s 50c 5500mah). I see other posts about this working when delivering 1200-1400A, but my readout only shows ~1000A on each weld. Also, perhaps my materials are too thick (.2mm copper, .15nickel).

Gonna get pick up thinner copper and a higher C rate lipo to see if I can get this to go.

Kweld can only weld 0.1 copper. It will not any more and secure a good weld.

 

[Mourningstar]

Trying to find a welder under 200. Any out there that can do the copper nickle sandwich method?

No. Sped the extra $100 and buy the Kweld.

 

[chuyskywalker]

Kweld can only weld 0.1 copper. It will not any more and secure a good weld.

I was able to get the 0.2+0.15 copper/nickel mix to work at 70J hitting about 1300-1400A welds with "can't pull it off by hand" welds. (I may end up having to take the pack apart because it's possibly too big to fit the frame; I'll grab some pics of the weld removal if I end up having to do so.)

 

[SkippVonDoodle]

What is the "best" attachment for copper to copper?

Low resistance and high physical strength.

Ideally non-destructively reversible?

Not just for flat strips/sheets, but wire? Fusible links?

"best" way to join copper to copper, from my experience, is with GTAW (TIG weld), but you ABSOLUTLY, NEVER TIG ON A CELL.

silver brazing will also join copper incredibly well, but this requires a really hot torch and will not work for making connection to a cell (AGAIN, FIRE!!! DO NOT DO!)... maybe useful for those that don’t have a TIG welder, but have an Oxyacetylene torch and want to join large copper busbars together... pure silver is 5% more conductive than copper btw

this Cu/Ni sandwich method seems to be making some sort of Monel alloy, and I expect that the connections will far outlast the cells life cycle.

Has anyone added a dash of silver to their sandwich yet?

 

[hemo]

Recently finished a simple 2p 12s moli p42a pack for my Ute cargo bike, no issue wth slotted 0.15 nickel plated steel over 0.8mm cu sheet with a Malectrics.
5 or 6 y/o 3s 65c turnigy graphene at 45ms setting still performing well.

 

[windtrader]

After reading the entire thread, the following battery design and build should work well.

Battery 16s6p, 59.2v 28.8 Ah, 1705 watts

Battery cells - Tesla 21700 3.7v 4800 mAh cell

Configuration
5 packs - 3s6p
1 pack - 1s6p

Tab spec
0.1mm nickel plated steel
0.1mm copper sheet

Spot Welder
Kweld 1200-1300A welding current @ 60J

Application
e-Bike with controller max about 1800 watts (60v x 30aH), so just over 1C max discharge.

Seem reasonable??

 

[GalFisk]

I've had quite good results now with the tungsten electrode+0.1mm copper foil, though it took some experimentation and tweaking.
I'm using a pure tungsten TIG stick.

Which is quite brittle, but can be cut and polished with a Dremel cutoff wheel without cracking.

Here's the beginning of the first of 5 batteries (I didn't grind off the nickel remnants as you can see, which was a mistake):

Thigs to consider:

• This type of spot welding requires quite a lot more energy than welding nickel using copper electrodes. I don't have the actual numbers, but I went from 14 ms@11V to 41 ms@14V on my welder.
• It is very important to have nice flat electrodes and use even pressure, to ensure a large contact area. Sharpened electrodes can't work consistently, since as soon as they get rounded, more energy is required for a weld. Flat ends is the only way to get a consistent surface area. A lot more energy is needed then though, which carries its own risks, see below.
• On used cells, grind down the remnants of old spot welds, or you'll have problems getting consistent welds. I've been doing this for my subsequent batteries, and the welding goes do much better now.
• Some copper will eventually stick to the electrodes. Be prepared to grind it off, or you'll get a hotspot there which excerbates the issue on the next weld.
• If the electrodes are not pressed down firmly, evenly and at a right angle, the high energy of the weld pulse can blow a hole in the foil, possibly even in the cell when welding the negative end. I've blown 5 foils and ruined one cell while perfecting my technique.
• Stay away from the center "no weld" area on the negative end. It's more heat sensitive than the rest of the cell.

Edit: I vaslt increased the contact pressure between the electrode holders and tungsten sticks, in order to solve another issue I had, and thanks to this I was able to lower the pulse time all the way from 41 ms to 2 ms! This also increased the consistency of the welds.

 

[hemo]

From my post up page.
My very simple 12s2p M42 moli build using the malectrics.


I built the battery for my Kona Ute with 250w 48v TSDZ2, enough range for me to travel 4 x 6 mile journeys towing a trailer with loads up to 200lbs. The route being 70% rough surface (loose to firm made up old railway trails for traffic free use and 30% tarmac roads) ascent isn't much at approx.313m one way and -258 in the opp direction.

 

[larryjamescobb@gmail.com]

I've had quite good results now with the tungsten electrode+0.1mm copper foil, though it took some experimentation and tweaking.
I'm using a pure tungsten TIG stick.
P1130721.JPG
Which is quite brittle, but can be cut and polished with a Dremel cutoff wheel without cracking.
P1130720.JPG

I also am trying Tungston w. Lanth. electrodes. I bought 1mm diameter, but I think they limit the current.
What diameter electrodes are you using ?

 

[spinningmagnets]

My very simple 12s2p M42 moli build using the malectrics

What battery did you use for the welding current?

 

[GalFisk]

I've had quite good results now with the tungsten electrode+0.1mm copper foil, though it took some experimentation and tweaking.
I'm using a pure tungsten TIG stick.
P1130721.JPG
Which is quite brittle, but can be cut and polished with a Dremel cutoff wheel without cracking.
P1130720.JPG

I also am trying Tungston w. Lanth. electrodes. I bought 1mm diameter, but I think they limit the current.
What diameter electrodes are you using ?

I'm using 1.6mm diameter sticks, and I found that when I shortened the electrodes and increased the clamping force, I could go from a 41 ms to a 1.5ms welding pulse. Before, there were shims around the electrodes, and tightening the screws squeezed the shims more than the electrodes. Without the shims, the electrodes sit better and the force from the screws go directly onto them.
What's your reason for using tungsten+lanth?

 

[larryjamescobb@gmail.com]

I'm using 2% Lanthanated Tungsten Electrodes, since they were recommended for copper spot welding.
I have finished my 13S 3P 21700 battery pack

 

[GalFisk]

Looking good. Maybe I'll try lanthanated W when my current stick runs out. Do you know why that one was recommended? Did you have any issues with using it?

My only issue (after increasing the clamping force on the electrodes) is copper sometimes sticking to the probes and having to be ground off. If I don't remove it, I get a hot spot where more copper will keep sticking. Also, one electrode seems to be wearing down slightly faster than the other over time.

 

[Hummina Shadeeba]

reason not to use steel tabs? A bit cheaper and probably need less current.
https://m.aliexpress.com/item/3256802932898898.html?pdp_npi=2%40dis%21USD%21US%20%246.91%21US%20%243.18%21%21%21%21%21%402101f6b516613646102352977e8097%2112000024199774763%21btf&_t=pvid%3A1925aed1-3e48-433f-92b1-a27a0b0978a7&afTraceInfo=1005003119213650__msite__c_ppc_item_bridge__xxxxxx__1661364610&spm=a2g0n.ppclist.product.0&gatewayAdapt=Pc2Msite


But related to just above and welding copper directly, I’ve welded a couple batteries with the kweld using solely .2mm copper sheet.
Took a lot of power. I forget how much. I don’t want to do copper solo again though because took so much power and I worry it has caused problems with the cells.

 

[GalFisk]

That looks neat. .2mm is probably overkill though. I used .1mm on the cells and .15mm for interconnects. I only used .15mm where I could solder it, as it didn't weld well with my capacitor welder. With .1mm I had no worries about the energy deposited into the cell when welding.
Edit: I also used the infinite slot method, which helped lower the welding energy required.

 

[hemo]

My very simple 12s2p M42 moli build using the malectrics

What battery did you use for the welding current?

Beg pardon re the late reply.
I did post the battery a few posts up the page Ron, but in case it was missed a Turnigy 3s 65c Graphene lipo from Hobbyking. My one is about 5 or possibly 6 years old now but still going strong, it gets used a bit also for my bee keeping in powering a vape pan and a diy beevac for the humane capture of swarms in awkward places. Another bril use for it is turning a car engine over with a flat battery.

 

[Vincentp]

I have done some pretty extensive testing on the Copper/Nickel sandwich with the Kweld. I have successfully welded up to 0.25mm copper with 0.1mm nickel plated steel. The biggest problem I suffer from is that the copper electrodes have a tendency to stick to the sandwich. My weld energy is 300J at about 1,370 amps. I am going to run some tests with tungsten and molybdenum electrodes, maybe even elkonite if I get desperate($$$).

My question: Is 0.25mm of copper absolutely overkill? The reason I went for such thick copper is that I want the busbars to experience minimal heating while pushing 35 amps per cell with molicel P42A cells. Moreover, I am building packs that are non rectangular, as a result there are situations where the busbar must handle more than 35 amps at certain locations. I did a basic simulation and saw a maximum of 50-60 amps (can't remember exactly) at some busbar locations due to less than ideal current sharing as a result of the pack geometry.

Using thinner copper, say 0.2mm or even 0.15mm copper would be really beneficial, but I am no expert on ampacity and so I have played it safe for the time being.

 

[Frank]

300 joules sounds like a lot! I'm using .15 mm for a pack I'm now building using the Molicels; I calculated the thickness based on conduction area known to work (from another application with different batteries). I didn't have to account for uneven current distribution though. If you post a picture of your welds others with more experience can probably comment.

 

[caleb7]

I scanned the thread, I'm sure I'm missing something but how do you people come to conclusion that 1mm2 of copper sheet is capable of carrying, say as high as 30A? I'd like to use this design in future but not without knowing this. AWG ampacity charts say much lower ratings for solid wire, for example 1.3 mm2 copper wire is 15 Amps according to this list:

https://www.engineeringtoolbox.com/wire-gauges-d_419.html

I see on at least in one design, one cell was only getting a lip of copper sheet where it has like 1 cm width.

Is copper sheet better conductor than solid wire or its more capable because it's basically a sheet in 'air'?

 

[spinningmagnets]

0.20 copper is exceptional, and thanks for posting that. I'd feel comfortable taking high bursts of amps from that, then...if there is a hot spot in one section of the busses, I would look at the possibility of adding a jumper over that one spot.

The exact nature of the jumper would depend on the existing buses in that spot.