kWeld - "Next level" DIY battery spot welder

So I am about to start spot welding a 22S/20P pack. I have read the discussion threads regarding cleaning the electrodes and battery cell surfaces and I think I am all set. Most people recommend fine grade sand paper to keep the electrodes clean but what is the best cleaning method for the cells? I will be using brand new INR Samsung 25R cells but I want to make sure the weld surface (ie. anode and cathode surface) is as clean as possible. Thanks again for all the great advice on this forum. Gary
 
use a few taps of 80+ grit on the tips every 100 welds or so. just clean the cells with IPA before putting the nickel on.
 
I got mine today!!!!
I assembled it....
Now trying to learn about powering it.

I have lots of 5p4s LiFePO4 packs A123
12.8v nominal 9.6ah. I’ve tested these packs with no voltage sag up to 5-7C (45-65amps). All parallel groups hold at 3.2v.

I’ve read the Kweld manual of 4-30v For supply but I’m reading through this thread that 12v is best.

Anyone have a guess of how many I need in parallel???
I believe the pulse CCA at either 250 or 400CCA for 10 seconds.
 
Tommy L said:
0.1mm copper
0.2mm nickel

90 joules 1150-1200 amps
Looks a little hot, but during tests, this is where
Good adhesion is.

https://drive.google.com/file/d/1-K7nLbpHJJgJjSZHHYbwyDEvOcehPNMV/view?usp=drivesdk
90 joules sound like a lot to me. I think I used 50-60 just for 0.1mm copper. It is not very strong weld since the material is just too weak by itself. You should try thinner nickel and thicker copper. Also if you have more batteries you should try to ad one more paralel so weld current goes up to 1500-1800A and that gives you shorter pulse times what means in copper welding that more energy goes to weld and less for heating the copper around the weld. I bet you can turn down joule setting with more current and get same results and less hot electrodes..
 
ossivirt said:
90 joules 1150-1200 amps
Looks a little hot, but during tests, this is where
Good adhesion is.

https://drive.google.com/file/d/1-K7nLbpHJJgJjSZHHYbwyDEvOcehPNMV/view?usp=drivesdk
Also if you have more batteries you should try to ad one more paralel so weld current goes up to 1500-1800A and that gives you shorter pulse times what means in copper welding that more energy goes to weld and less for heating the copper around the weld. I bet you can turn down joule setting with more current and get same results and less hot electrodes..
[/quote]

Ahhhhhh this sounds like an awesome tip!!!
Yes my electrodes get hot. Time to try 3
12.8v nominal 9.6ah LiFePO4. I’ll do some tests.

Thank you 🙏🏻 so much

Tommy L sends.
 
ossivirt said:
Also if you have more batteries you should try to ad one more paralel so weld current goes up to 1500-1800A and that gives you shorter pulse times what means in copper welding that more energy goes to weld and less for heating the copper around the weld. I bet you can turn down joule setting with more current and get same results and less hot electrodes..

I put another battery in parallel just now!!!
Three 12.8v nominal 9.6ah LiFePO4 in total.
You are absolutely correct. 60 joules and reads
1515amps. Electrodes are cooler!!!

You sir are a godsend 🙏🏻🙏🏻🙏🏻🙏🏻
Greatly appreciate your efforts in helping me.
 
Tommy L said:
ossivirt said:
Also if you have more batteries you should try to ad one more paralel so weld current goes up to 1500-1800A and that gives you shorter pulse times what means in copper welding that more energy goes to weld and less for heating the copper around the weld. I bet you can turn down joule setting with more current and get same results and less hot electrodes..

I put another battery in parallel just now!!!
Three 12.8v nominal 9.6ah LiFePO4 in total.
You are absolutely correct. 60 joules and reads
1515amps. Electrodes are cooler!!!

You sir are a godsend 🙏🏻🙏🏻🙏🏻🙏🏻
Greatly appreciate your efforts in helping me.

No problem. You might be able to use even 4th battery in paralel but that could be very close or over 1800A that is max recomended I think. I have not tested the nickel over copper method but 0.1 to 0.15mm nickel should be thick enough and cheaper and even easier to weld.

Edit : After changes in battery or electrode setup its recomended to recalibrate the welder from the menu..

Edit2: kweld has setting caled kable lenght. Check that that is also correct in case your battery changes make your wiring longer. Long wires mean more inductance and that can break the welder so maximum allowed current decreases with wirelenght. Check the manual for that..
 
tatus1969 said:
I haven't used Pb batteries myself, but have collected feedback from other users on the kWeld product page. A 65AH battery (the 'normal' size for cars, likely also your Subaro) seems to be in the 1000A region, and larger 100AH models are around 1400A.

:warn: :warn: Can I make a poll from you guys using the welder with a starter battery? If you tell me model / age / reported current by kWeld, maybe along with a picture, then I would share that info for others.

Using a single used H8 AGM battery with two 2 gauge 15 inch starter leads (what I had in the drawer). The batteries are rated new for 1000A for 10 seconds at freezing, or 95Ah total. The used ones are still plenty strong to quickly turn over a v10 diesel engine. I calibrated trying to adjust the lengths to match what 8 gauge leads would be (roughly 195mm plus the welding leads), so 1.0m. That worked. Current from short showed 1679A with the used battery.

Welding 0.2mm nickel to some scrap 0.52mm copper plate took 80J consistently. No issues whatsoever.

Then I tried the 0.2mm nickel with 0.8128mm copper plate and I couldn't quite get a strong weld. Last successful was at 150J, 158ms and 1649A. Then I tried 175J and got a timeout. That copper thickness will be for most of my bus bars on the pack (nickel copper sandwich).

After that I will also need to spot weld 0.2mm nickel strips to 2mm copper plate as well.

Given that I can try:

1. Paralleling the second H8 battery to double the amperage, with the battery leads about the same length.

2. Change to a different energy source / and or add a cap bank.

Recommendations?
 
This morning for the 0.2mm nickel with 0.8128mm copper plate I tried 100J and placed one electrode on the nickel side and the other on the opposite side of the copper plate. Better weld, but still not complete. Increasing the strength caused a timeout, so I'm going to try two of these batteries in series (easier to connect physically and a voltage over current advantage).
 
If you are connecting copper to nickel, and there is no cell involved, you might want to consider a microwave-oven-transformer / MOT...I found several 1500W units for free, and if you pay shipping I will send you one.

"Making a Resistance Soldering Unit, and maybe a spot-welder too"
https://www.electricbike.com/resistance-soldering-unit/

file.php
 
I pre-ordered this(being that they are out of stock at the moment) and whenever they get all the components in stock it will be on the way. I look forward to putting some packs together with it.

I have x96 A123 26650s that I want to make a 24S4P LiFePO4 pack with, and another x150 Panasonic NCR18650GAs on the way which have been stripped from unused packs and are claimed to deliver 95%+ rated capacity, that hopefully will have enough good cells in that order to build a 20S6P LiIon pack with. Planning to use the nickel copper sandwich method.

I think I can get away with using two large 12V AGM automotive starting batteries in parallel as a power supply.

I do not know when I will have everything needed to get to work due to the current series of events. I've been waiting on a 24S LiFePO4 BMS since July and it's been sitting in a warehouse in Indiana since August 5th. I'm tempted to order a vruzend kit that I know is probably going to give me problems, just to get my trike converted to an EV already...
 
agniusm said:
Received mine:

B9dcQTh.jpg


8T6qQz8.jpg


Thanks

Would you mind sharing the stl files for the casing?


I've ordered the kweld with the kcap module. I'm planning on charging the caps with my 60v 5A bench power supply through a murata rbq-8.2/45-d48 isolated DC-DC converter to use as much of available power as possible. Do you think that is a good alternative?
 
I have recieved a kWeld, kCap and kSupply yesterday, and am able to successfully weld 0.15mm nickel with it (that's another Sunkko on its way to the big workshop in the sky)

I have set the output voltage of the kSupply to 8.1V without anything connected as described in the manual, and when I connect everything up the the input side of the kCap reads 8.1V but the output side maxes out at 7.66V. Is this expected behaviour? I assumed the output side of the kCap would give the full 8.1V when the caps are fully charged.
 
Sorry if there is info in the forum about this topic but I could not find it using the search tool!

Is there a way to connect 2 kWelds together to be able to handle more current?
I don't need to double what one welder can do but I certainly feel the need for mine to handle probably 20 or 30% more juice to weld .20mm, pure nickel strips WITHOUT sticking electrodes.

I have a collection of welders I have accumulated through the years and the kWeld is the best :bigthumb: and it would be just perfect for my needs if it could handle currents in the range of 2400/2500 amps without the over-current protection kicking in.

Since I have several it occurred to me, can I somehow hookup two in a way that would increase the 2000 A current limit of just one?

Crazy idea, I know, but you never know until you ask.
 
NetPro said:
Is there a way to connect 2 kWelds together to be able to handle more current?
I don't need to double what one welder can do but I certainly feel the need for mine to handle probably 20 or 30% more juice to weld .20mm, pure nickel strips WITHOUT sticking electrodes.

if the kweld is not enough for .2 pure nickel you are doing something wrong. first guess would be not enough pressiure on the electrodes and too fine a point on the electrodes.

if the welder says you actually hit 1k amps or over you should be around 50J if you are using a decent supply like the kcap.
 
The fact is, kWeld can do the .2 pure nickel well, but ONLY with the supplied electrodes.
My beef with the supplied electrodes is (as good as they weld just after they arrive or you re-dress the tip) lack of repeatability.
For consistent results, electrode gap, pressure and tip shape/size must not vary and it is impossible to achieve this hand-holding them for each weld.
Re-dressing them properly takes a lathe and cannot be done with the electrodes mounted in a welding head because if you file the tip it gets bigger and bigger quite fast.
Speaking of which, I bit the bullet and bought a so-so-quality pneumatic one in the far East (picture below) and installed Sunstone's Copper Glidcop Electrodes EL125BATT. Also tried the next size up they sell.

Now, these electrodes, having a cylindrical shape with flat tips, can very easily be re-dressed to perfection while mounted in the pneumatic head.
Since I have a lathe, I have made electrodes from different copper alloys, of different tip diameters and I am still having trouble with those that produce a good weld nugget because they stick to the weld.
Have tried tungsten of different diameters and sources and played with a wide range of pressure from very light to very strong and everything in between.

The kicker is, if I weld .15 pure nickel with the kWeld (and many of the electrodes mentioned above), I get an excellent weld in a pretty much sticking-free operation.

Aha, you say, it is the thicker .2 strips that are the problem. While this may be indeed the problem, I have purchased 3 different rolls of this material and done both, the salt and grinding wheel spark test and they all appear to be made of pure nickel.

For power source I have purchased the following from Hobbyking:

2 x Turnigy Graphene Panther 5000mAh 3S 75C
4 x Turnigy Bolt 5400mAh 3S 11.4.2V 65~130C
4 x Turnigy nano-tech 5000mah 3S 65~130C Lipo Pack

As well as an Optima Yellow Top Deep Cycle Battery, Group Size 31T, 900 CCA, which I cannot use because it trips the kWeld's over-current protection.

At the end of the day, as I increase the current to my kWeld, the stickiness begins to subside but I can't go past the 2000A built-in protection, hence my asking the question yesterday.
I have a feeling that if I could increase the power to around 2400A I would hit the jackpot and win a perfect, consistent welding nugget.

Now, I am not doing this for a commercial application (I know it sounds like I am) I just like to optimize things and get them as close to perfect as I can. I love the challenge and enjoy the time spent at the shop building stuff and testing it.
You should see the weld head I built from scratch before I purchase the pneumatic one. (I thought that was the component responsible for my troubles but it wasn't)

Plenty more info to provide on this subject if anyone cares to read it: Just ask and I shall respond.

Thanks!
image.png
 
I would highly recommend investigating the copper/nickel sandwich method for welded buses. It's a proven method that provides more than three times the series current, while also running cooler and generating less voltage sag under load.

Nickel is a poor conductor, so increasing thickness from 0.15mm to 0.20mm is fighting to desperately increase conductivity 33%...using copper for series is a 300% improvement.

If corrosion from bare copper ribbon is a concern, Keith at 18650 heat shrink and cell holders has nickel-plated copper ribbon.
 
spinningmagnets said:
If corrosion from bare copper ribbon is a concern, Keith at 18650 heat shrink and cell holders has nickel-plated copper ribbon.
I haven't been not worried about corrosion of copper, but then I plan on at least an IP65 case, maybe full potting too.

But I like the idea of nickel-plated copper anyway; any idea of its amps rating without over-heating, or AWG equivalency?

 
Back
Top