Pure nickel vs nickel plated & copper sheet?

[sternwake]

I have a very small low angle 1/8" wide chisel that easily gets under the 0.1mm SS, and a little twist and pry popped it off the copper pretty easily, and or lifts it enough to get the side cut pliers in there. I've not done this intentionally at scale though.

I respect math, but I tend to just go overboard on busbar width and thickness, as I want my busbars to be heatsinks, rather than heatsources, as much as is possible, within reason.
I use 0.2mm copper in applications where 0.15mm nickel would be 'just fine' on paper.
I am personally not seeking 'just enough' or weight savings.

I will double fishpaper ring and overlap the cathode well before limiting busbar width.

I'll buy 100mm+ wide rolls of copper or stainless and cut to fit with capable tin snips.

I think a lot of people looks at a cell's theoretical max CDR and try to size their busbars for that, when in actual use that power level will be very short lived and the 0.6mm copper that seems to be required on paper, is well excessive in actual use.

 

[Zambam]

I have some spare Gotian 32135 LFP cells left over from when I built a 22S2P pack by soldering because I didn't have a spot welder back then. Now that I have a 10.5 kw AwithZ UF20B, I thought I'd give it a go even though I know it's not powerful enough due to the 32135's thick copper electrodes.

I did not take a photo but the 0.10mm 304 SS/ 0.15mm Cu sandwich did not stick at all to the 32135 cell with gear 99, 0.15 ms preheat, 0.2ms intermittent, 2 pulses.

Next I tried it with flux. It did stick although it was easy to peel off with pliers, not a weld I would trust.

In this video @ZEUS-FL used straight copper with flux (no sandwich), gear 700, not sure which AWithZ.

One reason I did this test with my UF20B is that I think with 0.10mm 304 SS/ 0.15mm Cu sandwich + flux, a welder with more power than UF20B (P20B or higher?) may be capable of producing good welds on 32135 LFP cells. The only way to know for sure is via actual tests.


Sorry for out of focus photos, my phone does not have macro mode.
You can see 2 silver spots on the copper where the flux melted.


With the flux wiped away you can see the silver spots better.

 

[sternwake]

My phone does not take great macro photos either. I found if I hold phone a bit farther away and zoom in, then it can focus, but the phone needs to be held really steadyy or motion blur happens. One way to reduce motion blur is a LOT of light( I use a high CRI flashlight held just below lens and hotspot aimed at the welds, so the shutter speed is faster.

Cant say if this will be effective with your phone but it is worth a try.

I was thinking about getting some of those Lifepo4 cells, as I got a low$ 12v nominal 22AH AGM pushing 7 years old, and 12v nominal forms the basis of a lot of my lighting and ventilation and charging of Lithium via CCCV boosters.

Thanks for the exploring the limits of your UB20 and posting about it.

I believe I found the upper limits of my P20B being stacked 0.35mm( 0.2 + 0.15mm) copper onto a nickel plated steel cell, using flux and triple pulse at gear 940+ IIRC.

I did later find that homogenous 0.3mm copper required more power to weld successfully, than stacked 0.15mm copper, So if solid 0.35mm copper was available I likely would find the P20B lacking.

When exploring the upper limits of the welder, using flux, try increasing the preheating duration and reduce the 'intermittent' . This 'intermittent' is the delay between the preheating pulse and the first welding pulse, and on the other menu reduce the interval between pulses to 0.1 second. I think trying to get as much heat in as short a duration as possible, not trying to spread it out over longer is a more effective strategy when the welder does not have an excess of power available.

Other things which can help are sanding the copper with new sharp 220 grit sand paper. This not only cleans the oxidation, but leaves small peaks and valleys on the copper into which the flux can flow. It also reduces the amount of copper in contact with the cell below, and thus increases the localized heating under the electrode.

One can also remove the nickel plating on the cell itself intentionally rather than as a byproduct of removing previous welds.

One other trick that might help is to reduce the pressure of the welding pens compressing the sandwich. Press down firmly to press the sandwich, but then lift off the pressure a good amount, and then hit the pedal.
If one is pressing down super firmly then there is actually less resistance and less heating.
So as long as the sandwich is in contact, no air gaps, then a huge amount of downward force is unnecessary and not helpful when the welder is already at the limits of its power. When there is an excess of power then more pressure can be used and achieve more consistency from weld to weld as the amount required to compress and then the degree to which one lifts pressure takes time to develop the muscle memory and feel how the electrodes sink into the sandwich.

My old purple PCB welder had no timer or pedal and was underpowered for 0.15mm copper under 0.1mm Nickel plated steel, So I really had to do a LOT of test welds figuring out the timing, and pressure of the electrodes ( and tip shape) on the sandwich.

So the P20B is like a Luxury in comparison, as is stainless, and Flux.

 

[Zambam]

I found if I hold phone a bit farther away and zoom in, then it can focus, but the phone needs to be held really steadyy or motion blur happens. One way to reduce motion blur is a LOT of light( I use a high CRI flashlight held just below lens and hotspot aimed at the welds, so the shutter speed is faster.

I know the hold phone away and zoom in but not the CRI flashlight. I'll try better lighting next time.

I was thinking about getting some of those Lifepo4 cells

They work good in my scooter and was a good deal.

When exploring the upper limits of the welder, using flux, try increasing the preheating duration and reduce the 'intermittent' . This 'intermittent' is the delay between the preheating pulse and the first welding pulse, and on the other menu reduce the interval between pulses to 0.1 second. I think trying to get as much heat in as short a duration as possible, not trying to spread it out over longer is a more effective strategy when the welder does not have an excess of power available.

I was never sure what those settings do. Thanks for the info. I will try your suggestions and see if it helps. I have a feeling it won't help much since my welder is under powered.

Other things which can help are sanding the copper with new sharp 220 grit sand paper.

Cell was a bad one unsoldered from the battery. I used a coarse file to remove the solder, down to bare copper.

One other trick that might help is to reduce the pressure of the welding pens compressing the sandwich. Press down firmly to press the sandwich, but then lift off the pressure a good amount, and then hit the pedal.
If one is pressing down super firmly then there is actually less resistance and less heating.
So as long as the sandwich is in contact, no air gaps, then a huge amount of downward force is unnecessary and not helpful when the welder is already at the limits of its power. When there is an excess of power then more pressure can be used and achieve more consistency from weld to weld as the amount required to compress and then the degree to which one lifts pressure takes time to develop the muscle memory and feel how the electrodes sink into the sandwich.

A lot of good tips that I will experiment with. Thanks!

 

[Zambam]

try increasing the preheating duration and reduce the 'intermittent'

Here are the settings used on AWithZ UF20B (edit: my welder is underpowered for Gotian 32135 cells)


Before weld surface prep


Weld spots


Didn't help much, welds popped off easily.


With flux removed, solder spots visible on copper strip, not visible on cell copper tab


illuminating with a bright white LED flashlight helps a lot with photo quality though

 

[sternwake]

I am not saying it would have made the difference between weld and not welded, but those scratches look more like you used dull 100 grit sandpaper instead of new sharp 220. wvwvwvwvw shaped as opposed to unmumunun

Was on of the welds atop those round obstacles on either end of the copper contact?

Regardless, when there is so little tinning next to/ under the attempted welds means the flux did not get hot enough, So more power is needed. The scratches and downward pressure likely only come into play when the welds are just barely enough, and maybe then the ministrations can make them a bit more than barely enough.

I have not tried to weld to a Solid thick piece of copper though.

I just tried to weld 0.2mm copper under 0.1mm stainless using flux onto 4mm thick copper at gear 949 and .50Ms of preheating 2ms intermittent double pulse with my p20B, and it absolutely laughed.
No chance.
Did not stick at all.
Two silver dots on the underside of the 0.2mm strip only, can barely see where on 4mm thick copper bar I tried to weld, cannot feel anything.

I see no point in trying gear 999 or going to triple pulse.

Just tried 0.1mm copper under 0.1mm stainless using flux at gear 949 onto the 4mm copper bar, and it did stick, but did not weld. Lifted off easily. There is two little pinpricks of copper left on the bar.

I do not know how representative the 4mm thick bar is compared to the contact on the Gotion cell, but I would not buy the P20B hoping it could do what the UB20 could not.
The P30C might not even do it, probably need P60C or P120D or a laser welder.

I am certainly rethinking buying those cells if I can't weld to them. I am not keen on trying to drill and tap them for threads.

 

[Zambam]

I do not know how representative the 4mm thick bar is compared to the contact on the Gotion cell

4mm is 2.5mm X thicker than the Gotian 32135's copper tab, which 1.6 x 7 x 20mm. Do not give up on your P20B yet.

 

[Zambam]

Was one of the welds atop those round obstacles on either end of the copper contact?

Those are rivets which secures the copper tab to the cell. No, neither welds are on the rivet.

 

[Zambam]

I am not saying it would have made the difference between weld and not welded, but those scratches look more like you used dull 100 grit sandpaper instead of new sharp 220. wvwvwvwvw shaped as opposed to unmumunun

I tried with another Gotian 32135 cell which was never soldered. Cleaned with the sandpaper which came with my AWithZ (says 800 grit). Same settings.

Didn't stick at all. Slid right off. The previous cell was sanded with a approx 60 grit diamond file which stuck better.

 

[Zambam]

@sternwake Have you seen this video ? He did a lot of test welds with different materials, including aluminum.

 

[ljbred08]

Reading this, one idea did come to me: we know welding copper-to-nickel/steel/cells works with these welders (standard copper-nickel sandwich), so what if we did it backwards? Could we weld a nickel or steel strip directly onto the cell's copper with no extra copper in between? Of course then we wouldn't get the low resistance copper connection, but it would be a start if it happened to work.

 

[sternwake]

@sternwake Have you seen this video ? He did a lot of test welds with different materials, including aluminum.

I think the author of that video and I have different opinions on what constitutes a strong weld.

If only one of the two welds sticks strongly, I do not consider it a successful weld.

I did like seeing that model of Supercap Spot welder in action. I would certainly be considering it if I did not have the p20B.

I was unaware of the NAOH on aluminum, I was under the assumption that aluminum oxidizes so quickly that one had to have a way of preventing oxygen getting in proximity of the recently deoxidized surface.

I did see one guy solder wire to an aluminum soda can by using some High flashpoint mineral oil.

I did try and weld to aluminum before but am not sure what alloy of aluminum that it was, nor did I try very hard regarding surface prep, or use a lot of power.

 

[Zambam]

I think the author of that video and I have different opinions on what constitutes a strong weld.

There needs to be some standard in spot weld pull test. I have 2 dumbbell weights, each 5 lbs. I'm going to use vise grips on a welded copper strip to lift one 5 lb weight. If it holds, add another to make it 10 lbs and see it it holds. After that I need more weights but I have no more dumbbells.

Edit: Come to think of it, I just remember I have a digital fish weight scale. Maybe I can rig something up with it.

 

[sternwake]

I just want to see the copper or Nickel strip stuck to the cell when I roll the strip off with some narrow needle nose plyers.
If I can just grab it with my fingers and easily wiggle peel/pop/lift it off, then I consider it not strong enough.

That video he seemed to think when one weld popped off cleanly with little or no effort, but the other one stuck and tore the strip that it was strong enough.
I want both spot welds torn remnants to be as near in diameter as possible.

When only one weld of the pair is strong, it has always been a sign to me that more power is needed.
With every bump up in power that less strong weld starts sticking and leaving more and more diameter or remnant left on the cell.

I have used a high carbon steel razor blade for test welding, to get in the general range of power needed to weld, but Those settings established on other materials, is just a starting point for determining where to start on an actual cell in my opinion.

If the copper or nickel strip is tearing and leaving bits on the cell, then I have confidence that vibration and thermal expansion and contraction is not just gong to have the welds pop off.

Also lifting straight off on a cell can deform the can on strong welds, where rolling the strip off to the sides can prevent this deformation. Rolling a weak weld will have it pop off same as lifting straight up. When sandwiching rolling can have the Stainless or steel or pure nickel pop off the copper below easier, but this is of little consequence in my opinion, on 0.2mm copper anyway.

As long as the copper tears and leave nearly equal size portions on the cell, I am content.

The size of the portion left on the cell can depend on which direction one holds the tapered needle nose plyers and rolls the sandwich off, so the diameter of the welded dot is a better representation so good lighting and magnification can really help fine tune the settings and method.

 

[Zambam]

4mm is 2.5mm X thicker than the Gotian 32135's copper tab, which 1.6 x 7 x 20mm. Do not give up on your P20B yet.

@sternwake
Have you found a piece of copper approx 1.6 x 7 x 20mm to test with your P20B?

 

[sternwake]

I have some 1.46mm thick 10mm wide and 200mm of copper bar stock.

I wonder the size and shape of those rivets holding that copper plate to the cells. I'd love to be able to use threaded fasteners.

 

[Zambam]

I have some 1.46mm thick 10mm wide and 200mm of copper bar stock.

Worthwhile to give it a go with P20B?

 

[Zambam]

I wonder the size and shape of those rivets holding that copper plate to the cells. I'd love to be able to use threaded fasteners.

Head is 3.7mm dia, from there it gets smaller, being a rivet. Probably not a good idea to drill and tap.

 

[sternwake]

Worthwhile to give it a go with P20B?

I cut off a 21mm long piece.

No chance whatsoever welding 0.2mm copper to it under 0.1mm stainless at gear 890 triple pulse.

Moved down to 0.1mm copper.
Gear 890 triple pulse. Stainless did not weld to 0.1mm copper but copper was welded to copper.
One weld was OK but still pulled off way too easily with no stainless on top.


Bumped up power to gear 940

This time stainless stuck to 0.1mm copper, but even less copper was welded to the 21mm section of extruded copper bar. I was using a bit more downward pressure.



I think extruding the copper work hardens the copper and makes it more difficult to weld to.

I just tried two more welds across the extrusions striations rather than with them, and bumped to gear 960.
Barely any difference. Two tiny pinpricks stuck. Lifted off way too easily.

I have to believe my bar stock of extruded copper is not a good analog to your Gotion cells as it seemed you were getting slightly stronger welds with 0.15mm copper with an ~45% less powerful machine

 

[Zambam]

You didn't use flux? Try that next?

 

[sternwake]

You didn't use flux? Try that next?

I always use flux, I just remove it after tear off with some rubbing alcohol on a cotton swab or paper towel to see how much tinning occurs, and to get a better idea of the size of the hole torn, if any.

 

[Zambam]

I always use flux, I just remove it after tear off with some rubbing alcohol on a cotton swab or paper towel to see how much tinning occurs, and to get a better idea of the size of the hole torn, if any.

I should have known you always use flux! So P20B is not powerful enough for Gotian 32135. Next ones to try: P30C and P60F?

 

[sternwake]

I don't know how representative my 1.46X10MMX 21MM extruded copper bar is compared to the the Gotion cell contact.
it does seek the p20b is underpowered for this task. I would not want to use less than 0.2mm copper and do not want to solder to cells ever again.

 

[Zambam]

My experience welding w/o flux and with flux when welding 0.20mm Cu/ 0.10mm 304 stainless sandwich to 18650 cells with my AWithZ UF20B is that it is just as strong. When it comes time to build a pack with 18650/21700 cells, I will not use flux! Why do more work and be exposed to the smoke when not needed?

 

[sternwake]

With my P20b:
I find 0.2mm copper under 0.1mm SS without flux requires gear 550 on 18650/21700
I find 0.2mm copper under 0.1mm SS WITH flux requires gear 385 on 18650/21700.

I have made an multi layer activated charcoal hepa filter to suck all my solder or welding flux fumes/smoke.

I find copper weld consistency is greatly improved with the flux and it reduces the power required.

Lesser welder power required means less cell heating which is desirable, but perhaps inconsequential.

I am watching Nelvicks/ Zeus-FL video with the Gotion cells. I am 95% sure he is using gear 700 on the awithz P90C as he says 56kw. He is welding 0.2 copper with flux, but no stainless. The P120B is 69kw.

I skipped the more economical Gotion cells, and just purchased 4 SAFT VL30Pfe 28ah power cells instead from batteryhookup. Threaded terminals.
The 22ah AGM on my workbench is too old to trust, and is well below 89% original capacity.