kWeld - "Next level" DIY battery spot welder

[mrpackethead]

Im sure lots of folks have probably said. "eww thats expensive".. I had a quick look at the costs of components, and its not that expensive..   Its definately not worth me trying to make my own! All the hard work is done.

[p.larner]

better off getting the cheap red chinese ones and fitting a cap to the mcu/driver feed!.1/4 of the price of the kweld!.

[mrpackethead]

Which cheap chinese one are you refering to.

[Mr.B]

Any news on progress with "kWeldPro" Frank?
Last post in Feb had me drooling...
I have an application coming up early next year.

[p.larner]

Which cheap chinese one are you refering to.

theres one called maple leaf look here

[tatus1969]

Its the lack of power from the mosfet driver that kills the transistors ,this is caused by voltage sag when welding,this causes the fets to work in linear mode(BAD),just fit a 470uf cap like shown on yt vids,after that they work well,a double pulse would be good for thicker nickel strip tho.

Your post doesn't relate to kWeld... Wrong thread? 

[tatus1969]

Any news on progress with "kWeldPro" Frank?
Last post in Feb had me drooling...
I have an application coming up early next year.

I wasnt able to progress as fast as I wanted, because we had to do firefighting for weird firmware bugs in a customer project, literally for several months in a row. Conclusion: Wireshark does not see everything bad happening on USB, but thanks to the makers of the Beagle USB sniffer we managed to find a solution (also, thanks to Microsoft for making USB things different since Win11... ).

At the moment, I am close to ordering a first batch of maybe 30 preproduction boards. Firmware development will take quite a bit more time, and therefore it is my idea to do the most important stuff first and then publish a number of firmware updates. For example, remote control via USB will come later, among other not so mission critical features. The GUI development is going nicely, attached a short clip of what this looks like at the moment. Last week I have started the enclosure design, which will be done by a fellow mechanical engineer who is mindblowing in SolidWorks. The enclosure will be punched, embossed, bent steel with powder coating, so there will be a sampling batch followed by a larger one. These processes require tool manufacturing, so that might become the bottleneck towards my goal of having the first smaller batch of units ready before this Xmas :-)

https://www.keenlab.de/wp-content/uploads/2023/09/VID_20230818_194152340.mp4

[Southerner]

A Last week I have started the enclosure design, which will be done by a fellow mechanical engineer who is mindblowing in SolidWorks. The enclosure will be punched, embossed, bent steel with powder coating, so there will be a sampling batch followed by a larger one. These processes require tool manufacturing, so that might become the bottleneck towards my goal of having the first smaller batch of units ready before this Xmas :-)

https://www.keenlab.de/wp-content/uploads/2023/09/VID_20230818_194152340.mp4

I am glad to see a metal enclosure.  I hate the plastic and 3D cases. I know they are easier to make and cheaper but metal is so much more rugged.
Thank you.

[tatus1969]

A Last week I have started the enclosure design, which will be done by a fellow mechanical engineer who is mindblowing in SolidWorks. The enclosure will be punched, embossed, bent steel with powder coating, so there will be a sampling batch followed by a larger one. These processes require tool manufacturing, so that might become the bottleneck towards my goal of having the first smaller batch of units ready before this Xmas :-)

https://www.keenlab.de/wp-content/uploads/2023/09/VID_20230818_194152340.mp4

I am glad to see a metal enclosure.  I hate the plastic and 3D cases. I know they are easier to make and cheaper but metal is so much more rugged.
Thank you.

That's the current enclosure status. This is laser-cut and bent, and I'm intending to publish the design files for people wanting to go that way. The downside, as can be seen in the picture, is that spray paint isn't as rugged as powder coating.

[mrpackethead]

Is there a thread to talk about Spot welding in general thats not on top of this kweld one.  While the kweld certainly seems like a great product, its closed source/design,  and very limited distribution, and expensive shipping make it a option i have to discard now.

https://www.eevblog.com/forum/projects/opensource-high-current-super-cap-welder/new/#new   

[ltwin8]

Is there a schematic for the kWeld available? I would like to rebuild one with better FETs and use is with supercaps.

I would like to use weldable slug caps and maybe laserweld them together in a 2S or 4S configuration using caps rated for pulsecurrent that are actually designed for the „abuse“ they are subject to, I plan on having the welder charge from USB-C or DC.

[mrpackethead]

There is one in the se

Is there a schematic for the kWeld available? I would like to rebuild one with better FETs and use is with supercaps.

I would like to use weldable slug caps and maybe laserweld them together in a 2S or 4S configuration using caps rated for pulsecurrent that are actually designed for the „abuse“ they are subject to, I plan on having the welder charge from USB-C or DC.

There is a schematic on page 2, which fets did you have in mind, and which caps?

[ltwin8]

That seems like a REALLY old schematic, I myself use a kWeld from 2020 (rev3 I think) and remember seeing Inductors on it? I thought it had a DCDC for driving gates at stable higher voltages?

Caps I thought of 60x138 3400F 0,230mOhm ones.

FETs I first have to search again bit it was a part with WELL below 1mOhm RDS_on and 60V V_DS, I found a 100V version also but it had something like 0.1mOhm RDS_on

So there is no schematic for the new kWeld version? I need to reverse it myself?

[mrpackethead]

Yes, i probably is an old schematic. I guess sometime during kwelds evolution it went from being somewhat open to being a closed thing.

[ltwin8]

Hi tatus, is there a schematic available for the kWeld v3 somewhere?

I would like to improve it myself and was wondering if a reverse engineering would be necessary?
After all the „magic“ is in the software I guess.

Or is it necessary to reverseengineer?

Best regards

[mrpackethead]

THe major differen

Hi tatus, is there a schematic available for the kWeld v3 somewhere?

I would like to improve it myself and was wondering if a reverse engineering would be necessary?
After all the „magic“ is in the software I guess.

Or is it necessary to reverseengineer?

Best regards

The major difference i can see between the published schematics and the currnent version looks like the addition of of a buck/boost swither so it can cope with a variety of inputs..

The Fets and drive dont' look like they have changed much.   I've started ditching putting knobs/lcds/panels on things, and either network connecting them or usb connecting them..  I dont' need all the screens on every device.    A phone makes a great controller for a LOT of things.

[JohnG]

I would like to improve it myself and was wondering if a reverse engineering would be necessary?
After all the „magic“ is in the software I guess.

I humbly suggest that if you are doing repeated and well controlled pulses >1000A  into a set of cables and electrodes (essentially a variable load) in a small unit, and doing it reliably enough for a product, there is also a fair amount of "magic" in the hardware.

John

[mrpackethead]

I would like to improve it myself and was wondering if a reverse engineering would be necessary?
After all the „magic“ is in the software I guess.

I humbly suggest that if you are doing repeated and well controlled pulses >1000A  into a set of cables and electrodes (essentially a variable load) in a small unit, and doing it reliably enough for a product, there is also a fair amount of "magic" in the hardware.

John

Magic's just science that we don't understand yet.     N Channel Fets are probably something that we do understand :-)

[tatus1969]

That seems like a REALLY old schematic, I myself use a kWeld from 2020 (rev3 I think) and remember seeing Inductors on it? I thought it had a DCDC for driving gates at stable higher voltages?

Sorry, but kWeld is closed source. I wouldn't mind at all having your DIYers copy and improve stuff for your own, but as this is part of my business I need to at least try to protect me from copy piracy.

FETs I first have to search again bit it was a part with WELL below 1mOhm RDS_on and 60V V_DS, I found a 100V version also but it had something like 0.1mOhm RDS_on

The MOSFETs on kWeld have among the lowest Rdson on the market, there is not much in that chip package that is better. They are also part of the current measurement circuit (as are the brass bus bars) and therefore cannot be modified without breaking functionality.

[mrpackethead]

Those Fets are a pretty good choice, even after nearly 10 years.   There are a few fets that may have 'better' characteristics, but we dont' need to be paying $300 each for them.

 

[rhb]

This is  off topic, but it's also the only place I think I might find an answer.

I am looking for recent monographs on capacitor discharge welding as applicable to making  a 1/2 sq inch area weld of a 3/8" block between two 1/8" plates with minimal distortion of the center block.

I'm interested in doing this using milled copper electrodes and fixtures to hold things.  I have Carslaw and Jaeger, so I *can* do it from first principles, but that's pretty tedious.

Search has become so manipulated that it's useless.

Reg

[tatus1969]

capacitor discharge welding

In fact, neither kWeld+kCap, nor the new kWeldPro are capacitor discharge welders. Both keep the welding voltage constant (using oversized caps), because that keeps the pulse power constant. I'm afraid therefore I have little knowledge about this. You may have a look at the devices from SunStone, they have their manuals available for download online.

[rhb]

Thanks.  I bought this:

https://www.amazon.com/Resistance-Welding-Fundamentals-Applications-Second-dp-1439853711/dp/1439853711

It seems pretty comprehensive and covers many PS types from rectIfied AC to constant energy pulses, electrode design and wear, clamping pressure, etc.  I could find no other recent books.  The others were over 50 years old.


Have Fun!
Reg

[WarthogARJ]

Thanks.  I bought this:

https://www.amazon.com/Resistance-Welding-Fundamentals-Applications-Second-dp-1439853711/dp/1439853711

It seems pretty comprehensive and covers many PS types from rectIfied AC to constant energy pulses, electrode design and wear, clamping pressure, etc.  I could find no other recent books.  The others were over 50 years old.

BTW.
You can get this on Library Genesis for free, you know.
Plus a few other related textbooks.




Have Fun!
Reg

[FrankT]

Sorry if this is not the right thread to post this.  I couldn't a find a kweld forum, or a general spot welding forum...

I have kweld (with super caps) and it's been great so far for batteries.

Now, I need to "stick" wires to some hardened steel rods. These are short rods, 3mm diameter, 8mm long.  I need to attach the wire to the end, at 90 degrees to the rod.  The wires are used detect when a mating part is in contact, so minimal current. (kinematic coupling)

So a few questions -
Will spot welding copper wire to steel work?  Is copper too low resistance, or should I try a different material? 

Can I weld a thin wire to a larger rod?  Will the heat dissipate through the large rod without welding?

Should I be placing both probes next to each other when trying to make the weld, or can I place one probe on each end of the rod.  The standard probes are quite large to place both on the end of a 3mm rod.  I may need to make smaller ones.