DIY copper/nickel buses for 18650's

[spinningmagnets]

Edit...if anyone wants to make these, feel free, this is an open source project. I hope somebody makes these for sale to the public, and I hope they make a lot of profit. I publicly make no claim to any intellectual property, anyone can use these for any purpose.

Xxxxxxxxxxxxxxxxxxxxx

I have seen nickel-plated copper sheet on Makita, Metabo, Bosch, DeWalt, etc cordless tool batteries. This gives you the high conductivity of copper, with the easy to spot-weld nickel. It's possible to DIY plate nickel onto copper, but that experiment is for another thread. An idea popped into my head, and this is just a proposal.

I will attempt to use very high amps from an RSU (https://www.electricbike.com/resistance-soldering-unit/) to bond copper sheet over the common 0.20mm thick nickel strips. The 30Q can safely provide 15A per cell, and HG2 is rated for 30A, so 15A-30A is the goal without getting hot. If you simply double or triple the layers of 0.20 nickel, it will just get hot and have voltage sag. (not a theory, its been done several times with the same result)

I have no concerns about being able to adequately bond the copper to nickel. I believe that if I just bond a simple copper strip over the entire nickel strip, then...when I go to spot-weld the combo onto the 18650's, the current will be able to pass through the copper from one spot-welding probe to the other (this is also why common nickel strips have a slot over each cell)...so I think the copper should have a full-length split down the middle.

I'm concerned about this because I believe it is important to get the copper as close as possible to the actual spot-welds, to reduce the distance the ebike operational current has to pass through nickel. Anyways, this is my initial idea. I thought I'd share for anyone who wanted to also experiment.








[one-mil is 0.001-inch thick, when researching sheet-metal options]
0.15mm__6-mil__34 ga [copper this thin will crumple like paper]
0.20mm__8-mil__32 ga
0.25mm__10-mil__30 ga [recommended for initial experiments]
0.33mm__13-mil__28 ga

[sizes below require sheet-metal shears to cut]
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__28-oz

 

[john61ct]

simplicity is a virtue

 

[flippy]

how much do you expect to gain?

 

[MJSfoto1956]

Sweet. I assume these are (or will be) water jet cut copper?

M

 

[spinningmagnets]

There's a good chance that this won't be worth the time or effort, just an experiment... The initial set will be hand cut with scissors.

Even if this works as well as I hope, I doubt there would be enough demand to make it worth ordering some to be laser-cut. But if there was a demand for these, laser-cutting would be the way to go.

 

[john61ct]

Have you researched the galvanic corrosion issue?

Maybe a thin bonding coat of conductive epoxy between, rather than heat. . .

 

[Taswegian]

Have you researched the galvanic corrosion issue?

Maybe a thin bonding coat of conductive epoxy between, rather than heat. . .

Should be a non issue? Nickel & Copper are both very low and very close in anodic index, so neither becomes a preferential site of oxidation/reduction supplying the other.

 

[liveforphysics]

I think it makes sense and is more simple than trying to get high current out of nickel busing alone.

 

[Headrc]

Definitely interesting ...I will be watching your progress on this.

 

[spinningmagnets]

I just remembered that we're not supposed to spot-weld onto the center of the negative ends. Here's my latest version...the red dots are the proposed spot-welds.

 

[liveforphysics]

I just remembered that we're not supposed to spot-weld onto the center of the negative ends. Here's my latest version...the red dots are the proposed spot-welds.

BatterySpotweldTab1.png

Getting warmer (close!)


Now think about the current density map through your conductor. How do those pair of linear connections look for currently density traveling over to that next cells positive?

 

[spinningmagnets]

I tried to think of a way to bond the copper to the nickel strips AFTER the nickel had been spot-welded onto the cells, but I havent found any ideas that I think would work. Sooo...I think bonding the copper to the nickel first will be the most "doable" compromise.

I originally thought it would work to bond a solid copper strip to the nickel strip (no central split) with a couple of cutouts where the spot-welds would go. But during the actual spot-welding of the nickel strips onto the cells, I fear that much of the current would pass through the copper, instead of going through the cell caps (which is also the reason for the "split" in the nickel over the cell connections).

the nickel has so much resistance the slot over each cell only needs to be just long enough to make the spot welding current take the slightly shorter route through the cell tip. The copper, however...is so conductive, a long slot is still not enough, and that means the copper must be made in two pieces (*infinite slot?)

Because the nickel has much more resitance than the copper, I imagine most of the current will go from the 18650 electrodes, up through the spot welds into the nickel strip, and then immediately onto the copper right next to the spot-welds. If the current path is what I imagine it to be, then the middle of the nickel strip could simply be removed, and the two ends can be based on two separate nickel "squares"...

I want every spot-weld to have a very short path to jump onto the copper. I don't think the split down the center of the copper can hurt, and I think there's a chance it may help. Even though the nickel is fairly flat and fairly stiff, I think any current that tries to pass from the cell to the nickel strip has a very poor path, except for the actual spot-welds themselves (six per cell-end shown above). And the current from the nickel to the copper is also poor, except for the actual spot-welds.

In this pic, I used dark brown to show where I will use 1500A to spot-weld the copper onto the nickel. Even though the entire length of the copper is up against the nickel, I think only the actual welds will carry current.

I believe this series bus style eliminates 3/4ths of the nickel from the series path, and the copper also provides a very good heat sink to absorb and then dissipate the heat from the peak acceleration amps.

 

[spinningmagnets]

At the top of this thread, I knew that it was very hard to drill holes in sheetmetal, and almost impossible in copper sheet this thin. However, its not hard to make a punch to pop round holes in sheetmetal.

0.25mm thick copper should easily conduct temporary peak amps of 30A per cell, and it can be cut by scissors. Even 0.33mm copper is readily available (28ga) and can still be cut by scissors. The nickel slots can easily be cut longer too. Now I'm thinking that this shape below would be easy to make, and would hopefully perform well?