kWeld - "Next level" DIY battery spot welder

Acido said:
bigbore said:
Acido said:
I keep blowing my lipos, will i reduce the time required to make the weld by increasing their voltage?
And there by reducing the heat and stress on the thin aluminum terminals

Im currently at 3s and can make a 5s batters from the cells I have left that are not broken (all the same 5ah 75c panther)

My old LiPos are still working pretty well .... did you see what I made?
Those batteries worked for some years on a e-scooter and then I rebuilt them in that configuration.

https://endless-sphere.com/forums/viewtopic.php?f=14&t=89039&start=275#p1359933
This is what i have done, similar to your setup but the solder unsticks probably from heat and then on the next weld a big spark happens and blows the terminal
I used the proper flux for aluminum terminals and the solder and wire sticked really well
https://imgur.com/a/s7h18lx

I made a 3S5P battery with massive copper conductors and in your picture I see a 3S1P. Only one cell alone isn’t enough for the task.
 
bigbore said:
Acido said:
bigbore said:
Acido said:
I keep blowing my lipos, will i reduce the time required to make the weld by increasing their voltage?
And there by reducing the heat and stress on the thin aluminum terminals

Im currently at 3s and can make a 5s batters from the cells I have left that are not broken (all the same 5ah 75c panther)

My old LiPos are still working pretty well .... did you see what I made?
Those batteries worked for some years on a e-scooter and then I rebuilt them in that configuration.

https://endless-sphere.com/forums/viewtopic.php?f=14&t=89039&start=275#p1359933
This is what i have done, similar to your setup but the solder unsticks probably from heat and then on the next weld a big spark happens and blows the terminal
I used the proper flux for aluminum terminals and the solder and wire sticked really well
https://imgur.com/a/s7h18lx

I made a 3S5P battery with massive copper conductors and in your picture I see a 3S1P. Only one cell alone isn’t enough for the task.

Actually the lipos can give enough current but the terminals are the problem so Im going to make a 2p pack out of 4ah red graphenes
 
For you guys who have problems with lipos... Who need to rebuild them.. ...


I keep blowing my lipos, will i reduce the time required to make the weld by increasing their voltage?
And there by reducing the heat and stress on the thin aluminum terminals

Im currently at 3s and can make a 5s batters from the cells I have left that are not broken (all the same 5ah 75c

Build em like I do. You dont need solder, just some copper, some good force, and some pure rivets. I've blown soldered connections before.. Using these batts with a Kweld I haven't had a single problem with crimped clamped pressed connections.

52845360_2439050809458567_1561764509574496256_n.jpg


52766796_2439050566125258_1842951860915273728_n.jpg


53023862_2439050539458594_2200316783737438208_n.jpg


Pulled 1900+ amps repeatedly. No issue. There are alot of circular mil in the tabs, enough for the current. Geeze the I bet id do 600+ amps with the tabs on the EV cells, hobby cells are weaker but they still need to be good enough ( to handle the high current pulse... ? ) to be a product, and should have enough metal in them ... paralleling is also a good idea.. ( I still wouldnt solder, to many dangers with the compromising of the metal quality, ie embrittled ( softened) annealed o r/ and ect... ). Maybe cause my cells are all rather large that I dont have problems... Idk. Runs the Kweld flawlessly.
 
I am currently building a pack using .2mm nickel. I have to crank the Kweld up to 125j for good welds and as a result the electrodes and their lead wires get pretty hot after just 4 spot welds and I have to let them cool off. Is this setting normal for .2mm nickel? And what about the heat in the electrodes and their lead wires? This really slows down the process of building a pack. Has anyone tried using a lug and larger gauge lead wires and different electrodes? The lug would be needed to be able to connect larger gauge wire. Or would this be harmful to the Kweld itself? Thanks
 
Headrc said:
I am currently building a pack using .2mm nickel. I have to crank the Kweld up to 125j for good welds and as a result the electrodes and their lead wires get pretty hot after just 4 spot welds and I have to let them cool off. Is this setting normal for .2mm nickel? And what about the heat in the electrodes and their lead wires? This really slows down the process of building a pack. Has anyone tried using a lug and larger gauge lead wires and different electrodes? The lug would be needed to be able to connect larger gauge wire. Or would this be harmful to the Kweld itself? Thanks
No, that's way too much for 0.2mm. I'd say 30-70J, but its always best to experiment with different settings, and make peel-off tests.

How much current do you get from your battery? And maybe repeat calibration while making sure that all bolts are tight (most importantly the electrode set screws, but don't overtighten since the material is copper), make sure the electrodes are clean (use fine sand paper), press hard during SHORT calibration. There should be no sparks at all.

The 8AWG cable shouldn't heat up from just 4 welds of 0.2mm nickel. The electrode system gets hot, but mostly because it picks up heat right from the molten metal in the spot - that's unavoidable.
 
The battery is a 24V 8S built with HEADWAY 38120 HP 3.2V 8AH LIFEPO4 LITHIUM BATTERIES 25C 200A cells. So it is capable of 200A. I did tests on compatible batteries ....Boston Power Swing 5300's .... before starting to build the battery. I could not get a solid weld until I hit that 125J setting. Any lower setting and the welds did not hold. I will try repeating the calibration and check the connections again.
 
Have you tried using these cells in 4S / 2P? Also, the cell interconnects can sometimes cause some resistance in the circuit. Perhaps a pic of your pack would help?
 
Headrc said:
The battery is a 24V 8S built with HEADWAY 38120 HP 3.2V 8AH LIFEPO4 LITHIUM BATTERIES 25C 200A cells. So it is capable of 200A. I did tests on compatible batteries ....Boston Power Swing 5300's .... before starting to build the battery. I could not get a solid weld until I hit that 125J setting. Any lower setting and the welds did not hold. I will try repeating the calibration and check the connections again.

those are shit batteries so expect poor results.
 
You can also see how much time it took to dump those 125J by keeping the foot pedal pressed.

My welds for .3mm nickel required ~60J and 12ms when the system was cold, 17ms when it was warmed up.
I could do something like 20+ welds before it got too hot to touch, which was still bad so now I mostly use .2mm nickel at 35J.
 
I recalibrated etc. and was able to do successful welds at 75J . I don't agree on Flippy's comments on the quality of the batteries. Regarding the battery interconnects they are 3/4" solid copper about 1/8" thick but I will look into taking a picture. I would like to be able to weld .3mm nickel ...or even nickel plated copper for that matter. Thanks folks ....I always value the constructive critique and suggestions here. Alway very much appreciated ...it is how methods and concepts get improved.
 
I apologize if I sounded dismissive. I simply had no idea about your battery set-up. The issue could have been any one of a dozen problems that were not obvious. I'm glad it is working now.

If you are welding 0.20mm thick nickel at 75J, and the kWeld was working at 125J, I am even more impressed with it...
 
No worries Spinningmagnets. I did not take it as such. Your input is always helpful and insightful. Yes I am quite happy with the Kweld ...still wish I could weld nickle plated copper but so far that has not been a successful endeavor.
 
Will someone please be kind and explain why something like kSupply exists?
Why can't i just connect an appropriate/cheap power supply up to 8.1v to the kCap?
However 2s 18650 wouldn't work->4.2x2=8.4v?
 
eikido said:
Will someone please be kind and explain why something like kSupply exists?
Why can't i just connect an appropriate/cheap power supply up to 8.1v to the kCap?
However 2s 18650 wouldn't work->4.2x2=8.4v?

Because cheap powersuplies dont have current limiting system and will destroy thenselves because over load.(capasitors draw thousands of amps when they are at 0v and charge voltage is aplied.) And usually have low current for repeated welding.

2x 18650 works for some time like cheap psus and have same problems. Current limiting and low output. Dont try its dangerous..

Like li-ion chargers you need cc/cv (constant current constant voltage) but alot more amps than usually.. And not that common voltage..

There are cheap supplies with current limiting but currents are so low that you might need to wait like 60s between welds :roll:
 
Yes, it is very important to NOT charge a 2S cell bank to 4.2V, 4.05V-4.1V would work.

I do not recommend using an 18650 2S cell bank. I only mention them because they are an option that "would" work. Many DIY home powerbank builders have hundreds of low amp cells laying around.
 
ossivirt said:
eikido said:
Will someone please be kind and explain why something like kSupply exists?
Why can't i just connect an appropriate/cheap power supply up to 8.1v to the kCap?
However 2s 18650 wouldn't work->4.2x2=8.4v?

Because cheap powersuplies dont have current limiting system and will destroy thenselves because over load.(capasitors draw thousands of amps when they are at 0v and charge voltage is aplied.) And usually have low current for repeated welding.

2x 18650 works for some time like cheap psus and have same problems. Current limiting and low output. Dont try its dangerous..

Like li-ion chargers you need cc/cv (constant current constant voltage) but alot more amps than usually.. And not that common voltage..

There are cheap supplies with current limiting but currents are so low that you might need to wait like 60s between welds :roll:

I was thinking something like this
https://www.amazon.com/Variable-Precision-Adjustable-Regulated-Alligator/dp/B077GVMP5X/ref=mp_s_a_1_72?keywords=power+supply&qid=1563811275&s=gateway&sprefix=Power+s&sr=8-72

I can dial in the voltage and limit the current.
How would these work?
 
john61ct said:
30V and 10A max?

really?

Sorry I just found a link and pasted it without reading the specs.
The 30v spec is good but what rated max current should I look for?
Or Maybe thats just a bad idea?
 
I think you need to buy a welder.

If you can't afford it, wait scrimp and save.

Maybe a kit.

Or perhaps you can just follow some detailed HowTo with a complete MoM.

I do not think you can just slap something effective together by winging it like this.
 
I'm getting a kweld.
I've made up my mind.
I'm just considering a kcap but I'm not really sure.
And the kcap needs a power supply.
 
eikido said:
Will someone please be kind and explain why something like kSupply exists?
Why can't i just connect an appropriate/cheap power supply up to 8.1v to the kCap?
However 2s 18650 wouldn't work->4.2x2=8.4v?
The main reason why I developed that module was that I realized that the welder not only requires lots of current during the short pulses, but also significant average power during heavy use. When you are in a good flow, then you can achieve close to one weld per second. Doing this with thicker nickel strips like 0.2mm easily requires 500W (average!) input power. At 8.1V, this is 500/8.1 = 62A. A bench supply with 50+ amps is out of reach for most of us, and I couldn't find off-the-shelf current limited buck converters with enough power. The kSupply module delivers even more current when being used with the recommended server power supply, and that at a much lower cost.

You can of course also use the kCap/[edit]kWeld system with a smaller 8.1V DC power supply, but the tradeoff will be that the system needs much more time to get ready for the next pulse. The kCap manual has a graph at its end that shows this relationship. That power supply must implement current limiting mode instead of overcurrent shutdown behavior, just like a charger or bench power supply. Virtually all standard DC adapters have overcurrent shutdown and cannot ramp up voltage of a supercapacitor.
 
tatus1969 said:
eikido said:
Will someone please be kind and explain why something like kSupply exists?
Why can't i just connect an appropriate/cheap power supply up to 8.1v to the kCap?
However 2s 18650 wouldn't work->4.2x2=8.4v?
The main reason why I developed that module was that I realized that the welder not only requires lots of current during the short pulses, but also significant average power during heavy use. When you are in a good flow, then you can achieve close to one weld per second. Doing this with thicker nickel strips like 0.2mm easily requires 500W (average!) input power. At 8.1V, this is 500/8.1 = 62A. A bench supply with 50+ amps is out of reach for most of us, and I couldn't find off-the-shelf current limited buck converters with enough power. The kSupply module delivers even more current when being used with the recommended server power supply, and that at a much lower cost.

You can of course also use the kCap/kSupply system with a smaller 8.1V DC power supply, but the tradeoff will be that the system needs much more time to get ready for the next pulse. The kCap manual has a graph at its end that shows this relationship. That power supply must implement current limiting mode instead of overcurrent shutdown behavior, just like a charger or bench power supply. Virtually all standard DC adapters have overcurrent shutdown and cannot ramp up voltage of a supercapacitor.

Thank you.
I understand everything now.
I also completely missed that bench power supplies were so expensive at higher currents.
 
How does kweld weld 0.2mm pure nickel for you guys? I understand it depends on the power supply but i'd love to hear how it's going for you guys.
This is the last detail i'm unsure of. Order 0.15 mm or 0.20 mm.
:D
I'm learning a lot. :lol:
 
Back
Top