kWeld - "Next level" DIY battery spot welder

How can you compare something that has already a good reputation (kweld) with some chinese stuff that you know nothing about?

You can set in the welder the default startup value for the joules.

Anyway why do you need so many joules? What are you welding?

My suggestion is first optimize your power source and the connection to it.

When all is optimized you will discover that you will need a lot less joules to weld and everything will be much cooler.

Don’t hurry to complain, instead think a bit what you did different that the others that get good results and are happy with the welder.


Edit. I just seen that you weld 0.2mm nickel. You don’t need 60j for that but more like 20-25j.

Edit2. With 25-30j I weld 0.15mm pure nickel and 0.1mm copper (with the actual pen) with the power source(capacitors) giving 1400-1500a

"How can you compare something that has already a good reputation (kweld) with some chinese stuff that you know nothing about?"

because i know by experience something can have good réputation without to be true, it's absolutely not an reference for me, i consider something good if i decide and validate it's good by myself and or by concret study on the subject, the advice of group of ppl who never touch any spot welder before except the bad first price one on china are not reference for me, same if that person doing less than 200 spots per month or if he use 3 days to make one battery ... Trust me for that

for the second point, the Chinese things, because i compare the Current capacity, but i precise i dont know if the current on the description are accurate, but i will told you that we will get it soon, and in addition, for the Kweld, you dont know as well ... who here measured it themselves ? Me yes, i get 100A less than what the Kweld displayed, but i'm not sure of the précision of my mesure with that current it's hard, i use 0.0002 Ohms 7W resistor 1% of accuracy and i mesure the voltage on my résistor with my Oscilloscope and i Spot on the both side of the resistor of course, i mesure that voltage and calculate the result by Ohms law, i will do the same with the 6000A Chinese one and we will see

"You can set in the welder the default startup value for the joules."

yes i say it, thx

"Anyway why do you need so many joules? What are you welding?"

First, take care, Joules and Joules, it's what i explained,

Exemple

1000A * 50ms = 50 Joules
2000A * 25ms = 50 Joules

you think you will have the same Energy injected and so the same result, NO !

both have the same energy but dont have the same result at all, depending of the dissipation of the material if you inject the same Joules but on more long time, you will heat around your point but not weld, so the automatic adjustment of the timing to maintain the same energy in spot i a good idea but only for very small range i think, the most important it's to have high current, short time as possible, Ideal it's to have current adjustment, and Timing adjustment (or Energy in place of timing) like that i ajust the current, i Select for exemple 1800A and 50J, Fortunately the Kweld have timeout protection, but fixed at 200mS it's already too much, you can't weld well if you need wait even 150ms you will overheat the part but not welding it, but if you had the possibility to inject more current, but on shot time to maintain the same energy you could solder the thing you can't with the 150ms and lower current
so, the amazing Kweld are not so amazing, it miss the cleaning pulse, miss a power switch, miss the timeout adjust, and miss the capacity of more current

and for answer at your question, i need to weld 0.2mm pure nickel to 18650 and 21700 with good repeatability and rapidity, and actually it's not the case, Even if i calibrate the Kweld, and i get something like 1400A, the repeatability is bad, and the power is not enough, if my tips is shaped yes but during the spotting the tips wear out, flatten and the contact surface becomes thinner and the power is no longer sufficient given the increase in contact surface, I also notice that the more the cables heat up, the worse the spot is good, which makes sense, and adaptive timing power compensation obviously doesn't help

it's me now have question, how can you compare Kweld who get alarm at 1800A with 6000A spot welder when you need to reduce the contact surface to weld comparing with Serious spot welder who Weld with 1.5 or 3mm contact surface ??? you see i can return the situation also, nothing is so easy as "good réputation = good" world are more complexe

"My suggestion is first optimize your power source and the connection to it."

it's on the way for, but you see the OEM set you have vidéo who make think it's powerful and get very good repetability without many heat, Yes ... on a shitty can, not on a real application, that yes i can complaining when precisely when you are made to believe that something is extraordinary, that you buy it and find that it is false and that it is far from it

"When all is optimized you will discover that you will need a lot less joules to weld and everything will be much cooler."

yes, when you add lots of money and arrive to the price of serious things haha stupid, but so, i will can validate if i'm true or false, the 6000A Capacitor based spot welder we bough cost around 450€ and the Kweld Cost acutally 210€ + 130€ of handle+Wires = 340€ and actually no powerSource for that price, let's see the price of the Caps Enough to go at 2000A during ~100ms ?? and also the power supply to charge it, the circuit to balance the Capacitors ... we will arrive at the same or more than something can get 6000A but for less than 2000, it is really interesting ? i'm not sure actually, let's the result IRL because i will comparing both, and if i need to put 20 Lipo in parallel to get enough current to do good comparaison i will do it, or maybe i will use the Caps inside the 6000A chinese spot welder on the Kweld to do it xD


" Don’t hurry to complain, instead think a bit what you did different that the others that get good results and are happy with the welder. "

that is why i am here but unfortunately I often see the same thing, my expectations compared to those of an amateur community is not at all the same, if you were next to me and you didn't have your day to waste to do 10 correct spots you would understand what I'm talking about

if someone here get ~1200 spots in one time, with good repeatability and in half-hour ~40mn let's me know and i will get interest of your advice, for all the rest who get 8hours to spot ~100 cells, sorry but we are not on the same way



"Edit. I just seen that you weld 0.2mm nickel. You don’t need 60j for that but more like 20-25j."

Yes probably with more current less time ... and with a sewing pin as electrode (comeback to all my explanation)

"Edit2. With 25-30j I weld 0.15mm pure nickel and 0.1mm copper (with the actual pen) with the power source(capacitors) giving 1400-1500a"

that the point, with 1500A, and probably very sharp electrode which will wear out and therefore expand and therefore no longer weld after a few repetitions...



50 joules * 2400 spots = 33.3Wh* of energy, it's mean if we make 1h to weld that, it's need to have constant 33W* in average, if in half hour need 66w* constant power consumption, for me it's not that hard to do, my Soldering Iron is 170W, 150W const so

* without considering the energy lost

PS all the professional spot weld for battery for the industrie or battery assembler, it's start at 5000A to 10000A, it's probably not for nothing, and the tips is 1.5mm, 3mm, and more
 
Last edited:
First of all let me mention that I am not affiliated in anyway with Frank, I just appreciate what he did and also have my respect for the ''amateur community''(as you call us) here. I shared my discoveries to help the others just like their past posts helped me!


because i compare the Current capacity
did you ever think that maybe in that video the copper foil is actually copper plated nickel? 6000A going through those lousy connectors(that are present also on you new pen) means rapid heat up(because of the resistance added) and possibility to catch fire. Just think about how much they heat at 1300A in your system.

Me yes, i get 100A less
100A from 1300A is under 8% error which for a device at 200euro is good enough. 100A doesn`t change your life when welding but the connectors on the input and also the double(yes double! you use that sort of xt plug and also the terminals for crimping) connections on the output will rule practically the result.

Exemple

1000A * 50ms = 50 Joules
2000A * 25ms = 50 Joules
When you have bad connections(yes, I consider those xt connectors that you use very bad when you want to pull out 60J) you get low currents and longer welding times and in the end no welds.

you think you will have the same Energy injected and so the same result, NO !
Certainly not, but it`s not me that doesn`t understand, it`s actually you that uses 100 connections between the power source(I wouldn`t call that a power source with those thin wires for the series/parallel connections but this is another story) and the pen and then you expect good results, sorry but you are funny!

the most important it's to have high current, short time as possible, Ideal
I can assure you that my system fulfils these conditions(also the one that Allex has so we are 2 already and many others that posted here in the past) because we both weld at low voltages. Me, I weld at 5.5v with 1500a and he welds a bit higher at ~7V. You need more voltage when you have losses in your system. If your system heats up too much means there are losses and that`s what you experience, isn`t it?

the amazing Kweld are not so amazing
there is a lot of people here that uses the kweld to weld 0.2mm pure nickel and nobody complains about it, instead all are happy with the results and suddenly you came here talking about pro stuff and that the kweld is not up to the task. Everybody welds 0.2mm nickel at 25-30j, you instead need 60j. Oh man it`s clear that your power source is not up to the task. When will you admit it?

and for answer at your question, i need to weld 0.2mm pure nickel to 18650 and 21700 with good repeatability and rapidity, and actually it's not the case
I am sorry for you but mine works like a champ, to mention that this is with the new pen. With the stock one (because of me like I said earlier) once in a while gave bad welds.

it's me now have question, how can you compare Kweld who get alarm at 1800A with 6000A
When you say 6000A do you mean the welder you shown on aliexpress? When setup right the kweld doesn`t need more than 1500A and 20-30ms to weld 0.2mm pure nickel at 25J.

i will can validate if i'm true or false, the 6000A
First validate then turn back here and praise that thing and make comparisons. This if you have a bit of respect for us the community, not to mention Frank.

my expectations compared to those of an amateur community
Your implementation and what you try to do suggests that you are not even close to the amateurs here who get good results and you don`t!
With what you do and what you say you just make a shame of yourself.

if someone here get ~1200 spots in one time, with good repeatability and in half-hour
in this case I guess we will leave you by yourself, sorry but with your attitude you ask for this!

that the point, with 1500A, and probably very sharp electrode
the head of my electrodes is rounded and not sharp, if you had consulted closely the last image I attached in the earlier post you would already guessed. But seems you are blinded by your failure and can`t see over it.

PS all the professional spot weld for battery for the industrie or battery assembler, it's start at 5000A to 10000A
Nobody advertised here the kweld as a pro welder, it`s only you that keeps mentioning about these PRO WELDERS and expect the kweld to be one and to be honest it is if you know how to implement it in your system.
A complete kweld get`s you to spend 500euro, for a pro welder you have to spend 10x more. Now let`s be serious, I hope that you don`t consider a pro welder the one you linked from aliexpress. That is a toy with the classic chinese good marketing that many fall for and then come here to complain they were scammed.

I am attaching an older photo where the welds were done with the stock pen. You see there 0.2mm pure nickel welded at 20J on 21700 cells. Aren`t they gorgeous? Seems an amateur like me is more pro than a pro like you :))))

So having these said good luck finding a pro welder as you wish for cheap or maybe for free!


PS: I remember that in the past I experienced something similar to your problem and it was my power source(3s/250ah battery at that time) that wasn`t charged enough. You live and you learn but you can`t blame the others when it`s your fault.


At this point being an amateur I can`t give no more advices to a pro like you because "I don`t want to make a shame of my self' so...


GOOD LUCK!
 

Attachments

  • IMG_3558.jpeg
    IMG_3558.jpeg
    1.8 MB · Views: 24
First of all let me mention that I am not affiliated in anyway with Frank, I just appreciate what he did and also have my respect for the ''amateur community''(as you call us) here. I shared my discoveries to help the others just like their past posts helped me!



did you ever think that maybe in that video the copper foil is actually copper plated nickel? 6000A going through those lousy connectors(that are present also on you new pen) means rapid heat up(because of the resistance added) and possibility to catch fire. Just think about how much they heat at 1300A in your system.


100A from 1300A is under 8% error which for a device at 200euro is good enough. 100A doesn`t change your life when welding but the connectors on the input and also the double(yes double! you use that sort of xt plug and also the terminals for crimping) connections on the output will rule practically the result.


When you have bad connections(yes, I consider those xt connectors that you use very bad when you want to pull out 60J) you get low currents and longer welding times and in the end no welds.


Certainly not, but it`s not me that doesn`t understand, it`s actually you that uses 100 connections between the power source(I wouldn`t call that a power source with those thin wires for the series/parallel connections but this is another story) and the pen and then you expect good results, sorry but you are funny!


I can assure you that my system fulfils these conditions(also the one that Allex has so we are 2 already and many others that posted here in the past) because we both weld at low voltages. Me, I weld at 5.5v with 1500a and he welds a bit higher at ~7V. You need more voltage when you have losses in your system. If your system heats up too much means there are losses and that`s what you experience, isn`t it?


there is a lot of people here that uses the kweld to weld 0.2mm pure nickel and nobody complains about it, instead all are happy with the results and suddenly you came here talking about pro stuff and that the kweld is not up to the task. Everybody welds 0.2mm nickel at 25-30j, you instead need 60j. Oh man it`s clear that your power source is not up to the task. When will you admit it?


I am sorry for you but mine works like a champ, to mention that this is with the new pen. With the stock one (because of me like I said earlier) once in a while gave bad welds.


When you say 6000A do you mean the welder you shown on aliexpress? When setup right the kweld doesn`t need more than 1500A and 20-30ms to weld 0.2mm pure nickel at 25J.


First validate then turn back here and praise that thing and make comparisons. This if you have a bit of respect for us the community, not to mention Frank.


Your implementation and what you try to do suggests that you are not even close to the amateurs here who get good results and you don`t!
With what you do and what you say you just make a shame of yourself.


in this case I guess we will leave you by yourself, sorry but with your attitude you ask for this!


the head of my electrodes is rounded and not sharp, if you had consulted closely the last image I attached in the earlier post you would already guessed. But seems you are blinded by your failure and can`t see over it.


Nobody advertised here the kweld as a pro welder, it`s only you that keeps mentioning about these PRO WELDERS and expect the kweld to be one and to be honest it is if you know how to implement it in your system.
A complete kweld get`s you to spend 500euro, for a pro welder you have to spend 10x more. Now let`s be serious, I hope that you don`t consider a pro welder the one you linked from aliexpress. That is a toy with the classic chinese good marketing that many fall for and then come here to complain they were scammed.

I am attaching an older photo where the welds were done with the stock pen. You see there 0.2mm pure nickel welded at 20J on 21700 cells. Aren`t they gorgeous? Seems an amateur like me is more pro than a pro like you :))))

So having these said good luck finding a pro welder as you wish for cheap or maybe for free!


PS: I remember that in the past I experienced something similar to your problem and it was my power source(3s/250ah battery at that time) that wasn`t charged enough. You live and you learn but you can`t blame the others when it`s your fault.


At this point being an amateur I can`t give no more advices to a pro like you because "I don`t want to make a shame of my self' so...


GOOD LUCK!
Thx for your help and to take time for answer me Schultz, also take into consideration that I am not an English speaker, and that it is not my native language, so communicating in this language is in itself a difficulty for me just to transcribe what I think and feel correctly, but for some point you maybe have true, for the bad connections, and the connectors that you call "XT" but which are actually "SQ8" so Bullet 8mm, originally I didn't have them, with the completely original kit, and the most contacts possible, and the result was even worse, so for me this is a non-debate, only a factor to discredit my opinion, but as I said, I will soon have an answer to my questions, once again thanks to myself and not to the amateur community, only able to criticize what the other is doing without being able to point the finger at the real problem , that said, you are the most likely person to do it, I recognize that my power source is too limited, and also that the 8 Awg cable section is too low, but it is the cable section that is supplied with the kit so supposed to be sufficient, this is not the case! it's a lie, and after buying it i have to reseach by myself why it doesn't work well and manage it, that is frocking annoying because it's always the same, at the end i finish everytime to have what i want, but i need lost time to search and solve it my byself, when i hope i will not the case cuz i trust of competent ppl who make complet kit to do it for me ... do you understand that point ? Ash
 
I took already into consideration that english is not your language neither mine so sorry if I got you wrong.

It doesn`t matter what connectors are those what matters is they are not designed for these current levels, keep this in mind, and they will add resistance that like I said simply stays in your way when trying to get a good weld.
Try to eliminate all extra contacts/connectors and also wire length that aren`t needed by all means.
Between the power source and the pen you should have 8 VERY STRONG CONNECTIONS and not more. With strong connections I mean wires directly crimped to the ring lugs and 50mm2 cables or even more like 100mm2(the welders you talk about have 100mm2 i think, now guess why? they weld at even lower voltages than 5.5v because of less losses in the wires). For good flexibility I recommend starting with very short wires(20cm max both wires) of 50-100mm2 between the power source and the welder and not less of 50mm2 cable between then welder and pen with a max length for both wires being 80cm.
Now you could try thicker wire like 100mm2(between the welder and the pen) but then it won`t be so flexible to move around. I went with 50mm2 for all of them and the above lengths and the outcome is awesome.
If you want to get the best electrical contact first of joining 2 pieces use some gloves and clean both contacts with isopropyl alcohol, don`t touch it after with your bare fingers. Wait a bit that they dry up and then mount them. Do a calibration test first and after doing this and you will see only from this what difference you get in the overall resistance. I suppose you didn`t clean nothing.

If you will check closely the user manual here https://www.keenlab.de/wp-content/uploads/2018/07/kWeld-operation-manual-r3.0.pdf you will find on page 3 that the kweld is able to weld 0.3mm nickel WITH THE RIGHT POWER SOURCE!
Now this is true but there should be something extra mentioned(I guess Frank assumed this) that your strip or plate should have those cuts in it between the welding points, like my strips have in the last post.
Without those the welding energy is not forced to go entirely through the cell tab this resulting in the need of higher weld power that will burn the nickel strip/plate and in most of the cases the weld wont happen.
I think you could try to put more pressure on the electrodes but still not a guaranteed success neither like this.
You need those cuts by all means. There are a lot of videos online with cnc pro welders, if you check those you will see that the strip or plate they use have these cuts like we(all the satisfied guys about kweld) have here.
Maybe we are amateurs here like you say but most of the times when you put many small minds together they are able to find a better solution than a greater mind, a pro in this case.

once again thanks to myself and not to the amateur community, only able to criticize what the other is doing without being able to point the finger at the real problem
Yes that one would be thanks to yourself, but other things you found out here(like the pen you just bought) were thanks to these amateurs and until now you offered nothing in exchange more than calling us amateurs which from now I START TO CONSIDER AS AN INSULT! . Now you didn`t expect that these amateurs will buy that aliexpress welder because you want to and will compare it to kweld? We don`t gain money from youtube or other kind of publicity around here so you got it wrong if thinking this.
Maybe the way we speak around this side of the world is different than the one on your side and maybe because of that you get us wrong and the same happens to us too.
Oh boy now we start to talk about criticizing the others, but look who started it first? To me seems it was you. Where you made the wrong affirmations you were criticized, the rest was not critic but advices what to do to be one of the happy users around here.
You don`t know to who you talk to here, you need to be kind and listen to the suggestions you get, instead you act every time like you know everything thing that pushes away the help that you can get. Nowadays the attitude makes the difference, actually it always did.
If you don`t want help and you know already everything what is your purpose here?

Once again if you use the proper strip/plate with the cutouts and also optimize your power source it will be a night and day difference.


Now let`s start with the right foot and do a nice thing. Let`s do a simple thing and try to optimize your power source.
I see you have a total of about ~40cm(which is a lot) of wire between the welder and the power source and also a connector. Is it possible to change this part with 2 wires of 10cm each and same thickness as the one you use there even thicker if you can and no connector.
Do a calibration test and then try again some welds starting with 30J and report back what you get? Don`t forget to charge properly your battery just to be sure it gives enough current for the cal test.
I attached a few photos of my welder so you can make an idea of the wire length that I use and also about the connections.


PS. Did you setup the total cable length of your setup in the firmware of kweld?
 

Attachments

  • IMG_3565.jpeg
    IMG_3565.jpeg
    2 MB · Views: 19
Last edited:
I'm interested in having only the base board of kWeld Pro, since I already have capacitors like the pictured ones.

But here, I also have a question regarding kCap:
Is it possible to add capacitors to existing kCap, or replace the existing ones?

Main concern is whether the balancing circuit will be able to balance the new or added capacitors. The other thing is, that only at 2 kCap modules the limit of 2000 A is already very close. But 2 kCap modules have only 2x 3S2P of 310 F, which is essentially equivalent to having a 3S system of a 1240 F capacitors. But the smallest capacitors I have are 2000 F each, and they are not used, but brand new. So, it's reasonable to expect, that if I do a modification on kCap, and even if the balancing circuit can balance them, now I will have the problem of an excessive current. Which I can only see solving by using some thinner wires, or lowering the voltage. Thinner wires will create excessive loss and heat. And I suspect, lowering the voltage is limited. To what limit than is possible to lower the voltage? I have also kSupply and an HP server power supply attached, which is currently set at 8,1 V.
I plan to sell assembled and tested units (with EMC and safety certification later on) as well as kits. I already have a laser-cut sheet metal enclosure here on my desk, form example. There will be a kit that includes the capacitor module, and also one without. It will be possible to mount a kCap alternatively, although that will provide less power of course.

Adding capacitors to kCap is not recommended, the balancers are not very strong. Replacing them would be possible, but in that case wouldn't it be easier to add some other suitable balancer to the pack?

You can use kWeld down to 4 volts, which in most cases will be low enough to get below the 2000A limit.
 
Hey Frank,
was muss ich da entdecken :D

Biete mich gerne als Betatester an. mit den letzten 3 Kwelds hab ich aktuell ~75.000 Zellen hinter mir :D

Hab den Elektrodenverschleiß nun auch im Griff mit mein "Big Boys" :D
(english for the other guys :) Happy to hear that, I should have roughly 10 prototypes within the next 2 to 4 months (I need to stretch my time sometimes due to other projects) and plan to start a poll once ready.
 
1- the Source of the current should be enough to inject much more Current, Cuz a Spot at same "Joule" but at 1000A and 1500A will not make the same result, the first will burn the nickel, the second will weld it
The energy metering approach of kWeld (which eliminates the need of having 1 or 2 softening pulses) assumes that the power source doesn't vary that much. This is because 1500A means roughly double welding power than 1000A. Consequently you get the same variation in pulse duration. The longer the pulse, the more heat can travel away from the spot.

If you have varying results with kWeld, then this usually indicates that the calibration did not work and needs to be redone. Make sure to calibrate with the same power source as used later for welding, clean the tips, make sure all nuts are tight, and during the SHORT step push the electrode *tips* together with *nothing* inbetween them. If you get sparks, repeat.
 
for me it's not true, it good approach but it can't eliminates the need of a cleaning pulse, why you did not implemented the cleaning pulse ? it's just a bit of code lines
I had done that and tested, and found that there is no improvement. kWeld delivered perfect and repeatable results without the cleaning pulse, no matter what I do. Also the community gives me this feedback. If you keep having trouble then please contact us and we'll check your setup and guide you towards good results.
 
Hello Frank,

I use my kSupply to generate 5.5V and would like to replace the single turn trimmer with 2 fixed resistors.
How can I calculate their value?

Thanks.
 
With mass-assembled battery packs that are assembled and welded by robots, the cleaning/softening pulse is cheap insurance.

With hand-assembled packs, the kWeld does not need a pre-pulse
 
I use my kSupply to generate 5.5V and would like to replace the single turn trimmer with 2 fixed resistors.
How can I calculate their value?
One simple way to determine the values is to adjust the trimmer to the point it does what you want. then take it off the board and measure it from wiper pin to outer pins, and those are your two resistor values.
 
You are right but this implies having the resistor values in house else I have to solder back the trimmer to have the welder working until I get the resistors and disolder it again when getting the resistors.
I don’t want to heat up the pcb 2 times but if I have to..
 
First of all let me mention that I am not affiliated in anyway with Frank, I just appreciate what he did and also have my respect for the ''amateur community''(as you call us) here. I shared my discoveries to help the others just like their past posts helped me!



did you ever think that maybe in that video the copper foil is actually copper plated nickel? 6000A going through those lousy connectors(that are present also on you new pen) means rapid heat up(because of the resistance added) and possibility to catch fire. Just think about how much they heat at 1300A in your system.


100A from 1300A is under 8% error which for a device at 200euro is good enough. 100A doesn`t change your life when welding but the connectors on the input and also the double(yes double! you use that sort of xt plug and also the terminals for crimping) connections on the output will rule practically the result.


When you have bad connections(yes, I consider those xt connectors that you use very bad when you want to pull out 60J) you get low currents and longer welding times and in the end no welds.


Certainly not, but it`s not me that doesn`t understand, it`s actually you that uses 100 connections between the power source(I wouldn`t call that a power source with those thin wires for the series/parallel connections but this is another story) and the pen and then you expect good results, sorry but you are funny!


I can assure you that my system fulfils these conditions(also the one that Allex has so we are 2 already and many others that posted here in the past) because we both weld at low voltages. Me, I weld at 5.5v with 1500a and he welds a bit higher at ~7V. You need more voltage when you have losses in your system. If your system heats up too much means there are losses and that`s what you experience, isn`t it?


there is a lot of people here that uses the kweld to weld 0.2mm pure nickel and nobody complains about it, instead all are happy with the results and suddenly you came here talking about pro stuff and that the kweld is not up to the task. Everybody welds 0.2mm nickel at 25-30j, you instead need 60j. Oh man it`s clear that your power source is not up to the task. When will you admit it?


I am sorry for you but mine works like a champ, to mention that this is with the new pen. With the stock one (because of me like I said earlier) once in a while gave bad welds.


When you say 6000A do you mean the welder you shown on aliexpress? When setup right the kweld doesn`t need more than 1500A and 20-30ms to weld 0.2mm pure nickel at 25J.


First validate then turn back here and praise that thing and make comparisons. This if you have a bit of respect for us the community, not to mention Frank.


Your implementation and what you try to do suggests that you are not even close to the amateurs here who get good results and you don`t!
With what you do and what you say you just make a shame of yourself.


in this case I guess we will leave you by yourself, sorry but with your attitude you ask for this!


the head of my electrodes is rounded and not sharp, if you had consulted closely the last image I attached in the earlier post you would already guessed. But seems you are blinded by your failure and can`t see over it.


Nobody advertised here the kweld as a pro welder, it`s only you that keeps mentioning about these PRO WELDERS and expect the kweld to be one and to be honest it is if you know how to implement it in your system.
A complete kweld get`s you to spend 500euro, for a pro welder you have to spend 10x more. Now let`s be serious, I hope that you don`t consider a pro welder the one you linked from aliexpress. That is a toy with the classic chinese good marketing that many fall for and then come here to complain they were scammed.

I am attaching an older photo where the welds were done with the stock pen. You see there 0.2mm pure nickel welded at 20J on 21700 cells. Aren`t they gorgeous? Seems an amateur like me is more pro than a pro like you :))))

So having these said good luck finding a pro welder as you wish for cheap or maybe for free!


PS: I remember that in the past I experienced something similar to your problem and it was my power source(3s/250ah battery at that time) that wasn`t charged enough. You live and you learn but you can`t blame the others when it`s your fault.


At this point being an amateur I can`t give no more advices to a pro like you because "I don`t want to make a shame of my self' so...


GOOD LUCK!
Now thats how welds should look like, no burned stains around the middle spot. Shows that the pwr spurce can give enough juice so the pulse is very brief.
 
I use my kSupply to generate 5.5V and would like to replace the single turn trimmer with 2 fixed resistors.
How can I calculate their value?
Attached is the voltage regulator. The trimmer sets the reference voltage, and V+ is the kSupply's output voltage. From that you can calculate the values. You can use R3 and R7 to mount the fixed resistors.
 

Attachments

  • Unbenannt.png
    Unbenannt.png
    11.3 KB · Views: 10
I mentioned in an earlier post that I charge my 4s1p (3000f 3V caps) power source to 5.5V. This got me 1750A in the calibration process and 1450A when welding.

I’ve been thinking lately to remove 2 caps and see how it goes. I wanted to do this because it’s easier to move around 700g than 1400g and it also takes less space on the workbench.

So today I did it.
Now with only 2 caps(2s1p) instead of going to 5.5V I go to only 4.8V and I get the same results as with the 4s1p config.

So the system became more efficient! Less caps, less power losses. One shot two hits.

All the tests were done using 4awg(25mm2) wire with a total length of ~30cm between the power source and the welder and ~90cm between the welder and the pen.
When calibrating the system I always got 1.45mR.

Because these wires(between the welder and the pen)are not as flexible as I wanted them to be(flexible as the 8awg ones)I thought to replace the wires between the welder and the pen with 8awg wires with 90cm total length. The wires between the power source and the welder remained the same.
With the 8awg I get 2.1mR when calibrating the system and ~300A less so 1450A when I calibrate and 1150A when I weld.
I can compensate this loss by charging the 2s1p caps to 5.5V

If I weld 0.2mm Ni the 8 awg wires are ok and I need ~25J for good consistent welds. I welded a 14s10p pack and everything went flawlessly.
At the end I tried to weld 0.15Ni + 0.1Cu but those 300A less do make a difference and the welds are not consistent.
So back to 4awg wires. I guess I will order 4awg silicone wire to see how that goes.

I don’t use no balancing and because how the caps charge(20-30mv difference between them at 2.4v) I don’t think I will use it in the future.

My conclusion to all of this, if you optimize the wiring and the contacts you don’t need 6x 3000f caps neither 4x 3000f caps, you will do with only 2.
If I charge 2x 3000f at more than 4.8V the system gives overcurrent when you do the calibration.

A big thanks to Frank for the kweld and ksupply and to the community for all the info around here.

If there is someone interested in the 2 extra caps and the wires that I have between the power source and the welder and also the wires between the welder and the pen I would gladly sell them.
 

Attachments

  • IMG_3613.jpeg
    IMG_3613.jpeg
    3.7 MB · Views: 9
There is something that bothered me since I had the original electrodes and I still encounter with the actual pen.
When I weld there is one spot that is stronger and the other one is a bit weaker. The stronger ones is where I have the negative electrode and the weaker is on the positive electrode.
Now when I weld nickel strips the weaker spot(the one on the positive electrode)is strong and requires enough force to rip it off.
When I weld 0.15Ni+0.1Cu the spot on the negative electrode comes always very strong and difficult to rip off but the spot on the positive it’s not so strong and sometimes it’s ripped off easily.
I tried to use bigger pressure on the positive electrode and less on the negative one and also the contrary but it doesn’t change much.
Is there something that I do wrong or is normal that one spot is stronger than the other?

When I say positive/negative electrode I mean the copper-tellurium parts that come in contact with the battery terminals. The positive electrode is the one that carries the positive from the power source through the fuse and the negative electrode the one that comes from the power source and goes through the kweld mosfet switch.
 
Here is an example, to make it more obvious I increased the energy to 60J for the first two welds from top to bottom and 70J for the bottom one. On the left is the negative electrode and on the right the positive one.
 

Attachments

  • 04373ABB-5546-4546-BBFF-A514474C2AF1.jpeg
    04373ABB-5546-4546-BBFF-A514474C2AF1.jpeg
    1.1 MB · Views: 10
After searching a bit around I think I got the answer, I needed 100-110J for this job.
So I charged the 2 caps at 5.4V(above this voltage the welder goes into overcurrent protection when I try to weld, when I try to calibrate it goes into overcurrent protection above 4.8V ) to get 1700a. At 4.8V I got 1400a.
With 5.4V and 1700a I manage to weld 0.2mm Ni + 0.2mm Cu at 100-110J. The required time for this is 100-110ms.

At 120J I managed to blow a whole in bottom of the cell. I am not sure if the reason was because of too much energy or too much pressure or because I did many test welds on that cell.
Anyway like I said earlier many thanks to all!
 
0.10mm thick copper is equal to the current rating of 0.40mm nickel. If you believe that you need more than 0.10mm copper, try 0.15mm.

Even if you do not burn a hole in the cell case, this high of current can create damage inside the cell.

I suggest you also add the "infinite slot" which will help reduce the current needed, because it forces the all of the welding current to take only the path you want it to.

0.15 copper is rated for the same series current as 0.60mm thick nickel
 
It’s not that I need(for now) 0.2mm Cu neither it’s current capability but I wanted to know how far I can go reliably. I had the 0.2mm Cu strip at hand and because of that I used it.
Lately I had a few requests for high power packs so from there the curiosity.
The packs that I weld daily are 14s10p that are discharged at 50A max(usually 30A max)so 0.2mm Ni does a pretty good job and the welder with its actual config do a very good job too, I could say awesome.

How much energy do you recommend as a starting point for welding 0.1-0.15mm Cu + 0.2mm Ni?
 
Here is the complete document, where it is described why one side is making a getter connection.
Also a lot of other interesting things there to read.
This link was already posted a few posts above.

 
This link was already posted a few posts above.

Hmm a few posts earlier I shared an image from this article.
Seems that I forgot more than I know but here comes into play the great community around here!
It’s great having you around!
 
Back
Top