kWeld - "Next level" DIY battery spot welder

Question for Kweld's designer:

Do you have any plans for adding a function in a future release that would switch the polarity half way through the welding cycle?
As I am sure you know, DC spot welders don't form a similar nugget under the anode and the cathode for reasons we do not need to get into now.

I purchased a couple of Kwelds in February and as advanced as they are, this function is the only thing missing to make this welder the holy grail of non-industrial battery spot welders.

Given the inconsistencies of hand holding the electrodes to weld, I purchased a pneumatic "weld head" from the far East, and I am not able to get equal welds under the electrodes when the pressure is identical.
By increasing the pressure on one side it is possible to reduce this difference but not eliminate it.
In checking the specs of the big welders (DC type) made by the industry leaders, one can see they address this phenomenon by reversing the polarity half way through the welding cycle, thus making each electrode both, positive and negative for each weld.

I have tested several spot welders and they all suffer from the same condition.
You have the know-how to make this happen and I wanted to know if there is a chance you, one day, surprise us with this high-end feature if at all possible in this non-production version of a spot welder.

reversing polarity would mean adding 3x more mosfets and a much bigger board.
i dont see that being economical.

NetPro said:

Question for Kweld's designer:

Do you have any plans for adding a function in a future release that would switch the polarity half way through the welding cycle?
As I am sure you know, DC spot welders don't form a similar nugget under the anode and the cathode for reasons we do not need to get into now.

I purchased a couple of Kwelds in February and as advanced as they are, this function is the only thing missing to make this welder the holy grail of non-industrial battery spot welders.

Given the inconsistencies of hand holding the electrodes to weld, I purchased a pneumatic "weld head" from the far East, and I am not able to get equal welds under the electrodes when the pressure is identical.
By increasing the pressure on one side it is possible to reduce this difference but not eliminate it.
In checking the specs of the big welders (DC type) made by the industry leaders, one can see they address this phenomenon by reversing the polarity half way through the welding cycle, thus making each electrode both, positive and negative for each weld.

I have tested several spot welders and they all suffer from the same condition.
You have the know-how to make this happen and I wanted to know if there is a chance you, one day, surprise us with this high-end feature if at all possible in this non-production version of a spot welder.

Click to expand...

Flippy basically nails it. Would be little more than a blink of an eye for me to implement this, but kWeld is targeted for the DIY community and this would rise the cost beyond economical. Still that's on my list for a future model, but integrated ultracaps and charger are on that as well. But if I find the time to investigate this phenomenon in more detail (not done yet), then I might find a different solution maybe.

@tatus1969.
Perfect. Thanks for info!
This is a well documented thing and there is a paper (among several) from a Chinese searcher in a university I can't remember, detailing scientifically his findings. I will post a link to it soon.
Traveling overseas now with only my phone and missing my desktop with big display.

NetPro said:

@tatus1969.
Perfect. Thanks for info!
This is a well documented thing and there is a paper (among several) from a Chinese searcher in a university I can't remember, detailing scientifically his findings. I will post a link to it soon.
Traveling overseas now with only my phone and missing my desktop with big display.

Click to expand...

Much appreciated!

tatus1969 said:

john61ct said:

Can you figure out what might be wrong?

Click to expand...

The voltage is absolutely okay, and if the welder makes consistent welds as well then I suspect an entirely different problem. Do you have the firmware updater so that I can send you two different firmware images for testing? I'm varying the 'short' detection parameter so that one firmware should always fail, and the other should always pass. If my theory proves true, then there will be an easy fix for this.

Click to expand...

I do not have the firmware update tool, and it seems out of stock.
When it becames available I can order it together with some electrode set, and then I will be able to check your testing firmwares.
Thanks.

frnandu said:

tatus1969 said:

john61ct said:

Can you figure out what might be wrong?

Click to expand...

The voltage is absolutely okay, and if the welder makes consistent welds as well then I suspect an entirely different problem. Do you have the firmware updater so that I can send you two different firmware images for testing? I'm varying the 'short' detection parameter so that one firmware should always fail, and the other should always pass. If my theory proves true, then there will be an easy fix for this.

Click to expand...

I do not have the firmware update tool, and it seems out of stock.
When it becames available I can order it together with some electrode set, and then I will be able to check your testing firmwares.
Thanks.

Click to expand...

A new batch of these is manufactured and about to arrive in a few days

A year ago I was messing with some Chinese super caps and got several of them to explode. They too were 2.7v caps. I thought maybe it's Chinese caps. As we all know China can make decent stuff, but they rarely ever match the level of abuse that legit electronics can deal with.

WARNING: Do this at your own risk and frankly I recommend that you don't!!!

I wonder if I voided my warranty? I have had the welder assembled and working just 3 days now. This thing is brand new!
I have to say that it made very nice welds...far better than my Chinese Sunnko!

Those Chinese caps got me thinking...what can these boards/caps really handle? What can legit Maxwell caps handle?
People post they run them on 3S or 11.1v or a car battery...which based on what I did a year ago is way too high.
As it turns out. Maxwell caps can handle a fair bit before they explode.

I knew it was possibly going to cost me some money in caps when I killed them, but worst case, I need to buy more caps.
I got more on Mouser for $17 each. My board will be up and running again in a few days.
I'm an experimenter and this would not be the first time I deliberately blew up something just to see where it fails.
I plan to get another 6 super caps to put in parallel with the 6 already on the board.

Just my opinion, but I'd add what I post below in the manual! MAKE these results VERY clear to people.

So here is my results from actual over volting earlier today:


WARNING: If you do this, expect to make a giant mess and blow up caps!!!
WARNING: The released chemical will etch paints and plastics!!!
WARNING: Have good ventilation, the released chemicals are BAD!!!
WARNING: You will be buying more caps!!!


Note: My tests were static, just let the caps sit there at X voltage and see what happens.

1. 9v is not a big deal. No fan...the caps get slightly warm. I've done probably 300 welds at this voltage off my Meanwell 12A PSU set to 9v.

2. 12v they got warm and the fan ran most of the time. I saw 4-6 amps draw that settled down to around .4-.6 amps continuous after they reached voltage.

3. 14.6 volts was catastrophic as you can see below. The caps got VERY hot, the fan could not keep up. I saw continuous 10 amp draw. 3 of the 6 caps exploded. They sprayed nasty chemicals every where. My plastic box is melted together by whatever the chemicals are. My eyes and nose burned. Like the Chinese caps I messed with last year...this was super nasty! DON'T DO THIS!!!

I have my own battery building thread where I posted this already, but figured this is the K-weld thread.

There are a couple details about the K-weld that I thought needed changing. I don't build packs all the time. I'm a hobbyist pack builder and so there are times when the welder will get put on a shelf in a box for storage. I also wanted to run it on a PSU as I may or may not have LIPOs or a car battery handy.

Once you bolt together the super cap module to the welder, it's bulky and not really storage worthy. I wanted to be able to take things apart without disassembling everything.

The fully assembled welder on my Meanwell PSU after I did a few mods. I already had the PSU from some other project. It works OK, but 12 amps from the PSU is slower than my ability to weld. After 10-15 welds, the voltage would go too low and I'd have to to stop for a couple of minutes to let the caps charge again. I'm getting a Meanwell SE-600-12. They do 12v and 60 amps. I bet that keeps up with my weld rate.




Things I wish the welder kit had in it already: (IE: My mods)

Note: I am using for these mods what I call 8mm long bullets, NOT the cheap shorter version. With the long connectors I can't tell they are there vs directly wired together. Castle Creations makes these long bullets and you can get Chinese versions too. I am using the Chinese ones. Watch out for the 8mm shorts! They are under powered for this application. You need massive dump current and minimal resistance.

https://www.ebay.com/itm/Castle-Creations-CC-Bullet-High-Current-8mm-Connectors-3-sets-CCBUL83/201296662275?epid=1200150920&hash=item2ede374f03:g8wAAOSwRiFch9Io:sc:USPSFirstClass!80021!US!-1

https://www.ebay.com/itm/5-Male-5-Female-HXT-8MM-Sprung-Gold-Bullet-Connector-Plugs-170-Amps/231536147191?epid=1185772157&hash=item35e8a13af7:g:8Z4AAOxy039TKjuV:sc:USPSFirstClass!80021!US!-1



1. The 8 awg cables between the super cap board and the welder are about 6" long. They bolt down to both boards. I won't be using the welder day in and day out so there are times when I'll stick it on the shelf for a while. I'll need to take it apart for storage. I cut the wires at the welder screw lugs and soldered on 8mm bullets to the screw lugs. This lets me disconnect the super cap module from the welder module.





2. The foot switch wires end at that green screw terminal block on the welder board. Again...I'll need to take things apart for storage. I used a 4 pin JST connector and used 2 pins in parallel for each wire from the foot switch. There's no real current load here. 2 pins each is just for long term reliability.





3. I haven't done it yet since I ran out of the 8mm long bullets. I'm going to add 8mm bullets at the welding cables and do what I did at the power input busses. This will let me disconnect the welding cables from the welder easily.

4. I think I will get rid of the longer wires between the super cap module and welder module. I'll work out how to mount the bullets from the super cap module so the two modules just plug together. Some solid copper with bullets soldered to them will do the trick. The whole thing will be a good bit more compact too.

5. This would be useful for PSU use or running from a battery. There is no graceful precharge for the super cap module. Just hook it up and KAPOW! What about when you want to put it away for a while? You can't gracefully discharge the caps except by doing welds. I've already got 8mm bullets on both sides of the super cap module, might as well use them to add graceful precharge and discharge. I've ordered some 20A rocker switches so I can put together a precharge/discharge module to use inline with my PSU. For people direct connecting to a battery pack, you'll need much beefier switches since your current load from the pack will be a lot higher than what I see from my PSU. The precharger/discharge is essentially a bunch of parallel resistors so you limit current flow and make it graceful. The switches let you bypass or disconnect the resistor banks.

Since I had just destructively tested my super cap board, I figured I'd mess with LIPOs direct connected to the k-weld. This is 6 10Ah Multistar 4S packs in parallel fully charged. I could get a weld, but it was very weak. I have another 6 packs, but I ran out of male XT60's so I can't parallel more of them. I think the C rating of these packs or maybe that they are older or the higher Ir they have is the issue.





I read that people were doing this with high C rate LIPO packs and using the dump current capabilities of LIPO. These are 750mAH packs for a qaud copter I built some time back. I don't know the amp draw I'm making, but I get nice welds. Who knows how long the charge will last.



I have 3 4S 10Ah Graphene packs. They are 15C. They have a seen a fair bit of use so their Ir will be higher. They did make a weld that was somewhat better than the Multistar packs, but it was still very weak. I was hoping they would do the trick.

I tried it with a 60Ah lifepo4 battery (a good one) and it would not even callibrate properly because the amps were way beyond 2000A.

So dont bother, you will only kill your kweld if you go for broke.

Always very well documented and challenging input Electricgod! Doing your best to live up to that name ...thanks...

ElectricGod said:

1. 9v is not a big deal. No fan...the caps get slightly warm. I've done probably 300 welds at this voltage off my Meanwell 12A PSU set to 9v.

2. 12v they got warm and the fan ran most of the time. I saw 4-6 amps draw that settled down to around .4-.6 amps continuous after they reached voltage.

3. 14.6 volts was catastrophic as you can see below. The caps got VERY hot, the fan could not keep up. I saw continuous 10 amp draw. 3 of the 6 caps exploded. They sprayed nasty chemicals every where. My plastic box is melted together by whatever the chemicals are. My eyes and nose burned. Like the Chinese caps I messed with last year...this was super nasty! DON'T DO THIS!!!

Click to expand...

What's the use of deliberately killing the components?


Burned eyes - at least are your eyes ok now?

Megavvolt said:

ElectricGod said:

1. 9v is not a big deal. No fan...the caps get slightly warm. I've done probably 300 welds at this voltage off my Meanwell 12A PSU set to 9v.

2. 12v they got warm and the fan ran most of the time. I saw 4-6 amps draw that settled down to around .4-.6 amps continuous after they reached voltage.

3. 14.6 volts was catastrophic as you can see below. The caps got VERY hot, the fan could not keep up. I saw continuous 10 amp draw. 3 of the 6 caps exploded. They sprayed nasty chemicals every where. My plastic box is melted together by whatever the chemicals are. My eyes and nose burned. Like the Chinese caps I messed with last year...this was super nasty! DON'T DO THIS!!!

Click to expand...

What's the use of deliberately killing the components?


Burned eyes - at least are your eyes ok now?

Click to expand...

What's the point?

People ask me that fairly often...
1. Flat out curiosity.
2. I've been me my entire life...just being me.
3. I did what you'll probably never do...test to destruction.
4. Most people play it safe...I rarely do.
5. I buy lots of stuff, just to see how it works.
6. In this case I want to actually use the welder long term, but while I'm at it, lets see what those caps can deal with?
7. Most anything I own, I take it apart, look inside and see how it works. Last night I was installing a new hub motor and didn't take it apart first...how UNLIKE me!


The burning lasted seconds. Just took some ventilation to blow out the fumes.

ElectricGod said:

WARNING: Do this at your own risk and frankly I recommend that you don't!!!

Click to expand...

WARNING: Please do not do this at all, this is very dangerous and can even lead to permanent injury! These caps can break down violently when being charged beyond their rated voltage!

ElectricGod said:

What can legit Maxwell caps handle? [...] As it turns out. Maxwell caps can handle a fair bit before they explode.

Click to expand...

No they can't. Ultracaps are very sensitive to overvoltage, similar to LiIon/Lipo batteries. They don't contain Lithium and have significantly less power density, and therefore won't burn down when failing, but overvolting leads to excessive temperatures and they contain a liquid electrolyte that will boil and cause internal pressure that may eventually cause real problems. You wouldn't dare charging Lipo cells beyond their rated 4.2V, and for the same reason shoulnd't do this with Ultracaps.

ElectricGod said:

Just my opinion, but I'd add what I post below in the manual! MAKE these results VERY clear to people.

Click to expand...

Both the manual and also the kCap product page state the rated 8.1V, and the manual also has lots of warnings in big red letters on the very first page. In addtion, we include a leaflet with more warnings that asks for following the supplied instructions closely.

tatus1969 said:

You wouldn't dare charging Lipo cells beyond their rated 4.2V, and for the same reason shoulnd't do this with Ultracaps.

Click to expand...

but it is fun....

tatus1969 said:

ElectricGod said:

WARNING: Do this at your own risk and frankly I recommend that you don't!!!

Click to expand...

WARNING: Please do not do this at all, this is very dangerous and can even lead to permanent injury! These caps can break down violently when being charged beyond their rated voltage!

ElectricGod said:

What can legit Maxwell caps handle? [...] As it turns out. Maxwell caps can handle a fair bit before they explode.

Click to expand...

No they can't. Ultracaps are very sensitive to overvoltage, similar to LiIon/Lipo batteries. They don't contain Lithium and have significantly less power density, and therefore won't burn down when failing, but overvolting leads to excessive temperatures and they contain a liquid electrolyte that will boil and cause internal pressure that may eventually cause real problems. You wouldn't dare charging Lipo cells beyond their rated 4.2V, and for the same reason shoulnd't do this with Ultracaps.

ElectricGod said:

Just my opinion, but I'd add what I post below in the manual! MAKE these results VERY clear to people.

Click to expand...

Both the manual and also the kCap product page state the rated 8.1V, and the manual also has lots of warnings in big red letters on the very first page. In addtion, we include a leaflet with more warnings that asks for following the supplied instructions closely.

Click to expand...

I can vouch for capacitors doing exciting things when abused. Kind of an irrelevant anecdote but; I was once standing in a room directly above our high current testing where we had a big bank of caps each about 3ft tall and a ft square. Max controlled discharge was about 200,000A and a *lot* of joules (the rig was for simulating lightning strikes) but someone left a ground lead near one of the terminals while charging, oops...

Resulting uncontrolled discharge vaporized & ignited the oil & the resulting explosion blew a wall out.

flippy said:

tatus1969 said:

You wouldn't dare charging Lipo cells beyond their rated 4.2V, and for the same reason shoulnd't do this with Ultracaps.

Click to expand...

but it is fun....

Click to expand...

How about Tatus wouldn't do this, but obviously I would. =)
And yes...it was fun to find out.
No major harm done...at least not to me there wasn't.
The replacement super caps will arrive from Mouser today so no big deal.

However, for other folks (not me), please stay within safe voltages for the super caps so they don't explode on you. It is messy and the chemicals are nasty. I imagine getting the chemicals sprayed directly in your eyes would be quite unpleasant and perhaps damaging to them.

Taswegian said:

Resulting uncontrolled discharge vaporized & ignited the oil & the resulting explosion blew a wall out.

Click to expand...

still sounds like fun.

flippy said:

Taswegian said:

Resulting uncontrolled discharge vaporized & ignited the oil & the resulting explosion blew a wall out.

Click to expand...

still sounds like fun.

Click to expand...

Exactly! I've been blowing stuff up since I was old enough to ignite gun powder as a 9 year old. I still have all 10 fingers and both eyes.

I have purchased a second super cap board so I can run them in series for more voltage. I want to mess with using my many LIPO packs as a power source, but they are all 4S or larger which isn't compatible with a single super cap board...as I demonstrated previously.

My existing super cap board will be running again tonight. Replacement caps arrived...I think I'll be OK.

ElectricGod said:

I have purchased a second super cap board so I can run them in series for more voltage. I want to mess with using my many LIPO packs as a power source, but they are all 4S or larger which isn't compatible with a single super cap board...as I demonstrated previously.

My existing super cap board will be running again tonight. Replacement caps arrived...I think I'll be OK.

Click to expand...

They are not designed to put in series, because the balancers act in an equilization fashion unlike most other modules, which only limit the cell voltages. The latter approach is not very suitable for such high current applications, as this would require a beast of a limiter otherwise.

To put them in series, you'll at least need equalization resistances across them to mitigate possible unbalance between the packs. I've made a corresponding note in the kCap user manual.

What you want is to get more current for the welder, and decreasing resistances helps better with this than increasing voltage. I have explained the relationships earlier in this thread. Putting two kCap modules in parallel should yield ~ 1700A of current.

ElectricGod said:

How about Tatus wouldn't do this, but obviously I would. =)
And yes...it was fun to find out.

Click to expand...

Sure, I didn't want to stop you from having fun (myself had fun in young years by connecting electrolytic caps directly to mains AC ), but as the kWeld designer and seller I felt urged to raise that warning for others.

tatus1969 said:

ElectricGod said:

I have purchased a second super cap board so I can run them in series for more voltage. I want to mess with using my many LIPO packs as a power source, but they are all 4S or larger which isn't compatible with a single super cap board...as I demonstrated previously.

My existing super cap board will be running again tonight. Replacement caps arrived...I think I'll be OK.

Click to expand...

They are not designed to put in series, because the balancers act in an equilization fashion unlike most other modules, which only limit the cell voltages. The latter approach is not very suitable for such high current applications, as this would require a beast of a limiter otherwise.

To put them in series, you'll at least need equalization resistances across them to mitigate possible unbalance between the packs. I've made a corresponding note in the kCap user manual.

What you want is to get more current for the welder, and decreasing resistances helps better with this than increasing voltage. I have explained the relationships earlier in this thread. Putting two kCap modules in parallel should yield ~ 1700A of current.

ElectricGod said:

How about Tatus wouldn't do this, but obviously I would. =)
And yes...it was fun to find out.

Click to expand...

Sure, I didn't want to stop you from having fun (myself had fun in young years by connecting electrolytic caps directly to mains AC ), but as the kWeld designer and seller I felt urged to raise that warning for others.

Click to expand...

I've put lots of stuff in series and they work OK. In the case of the super cap boards, each board acts like a voltage divider. It is possible to get a bit more voltage at one board vs the other as it is unlikely both boards will produce the exact same resistance or load. However, as long as each board is more or less close to a reasonable voltage, they should both be OK. 8.1v per board is pretty ideal, but lots of people run them at 11.1v. As long as I stay under that voltage per board it should be OK.

However, you are right, I really want more current, not more voltage. In parallel is the better approach. Also, in series adds resistance which is likely to be counter productive.

When I posted about testing to destruction, I did post several warning messages in big red letters. I think I was pretty clear about NOT doing what I did. Most people are going to see that post and already be unwilling to try this out. The people that aren't paying attention, don't know what they are doing, etc...well maybe they will see my post and be more careful before they have an accident.

I'm 52 and still having fun. No point in growing up now!

My blown up board is running again. It took a lot of heat to melt the solder on the 3 blown caps. Most people won't be able to repair a blown super cap board. I have a Hakko solder station that gets up to 1000F at 60 watts.

I have added another MOD. I have several LED volt meter modules. They cost around $1 each bought on ebay. There's an unused diode location next to the input power LED. It's right across the entire super cap bank so I see the voltage of the entire module there. I have soldered down the leads from the volt meter here and then mounted the meter on the plastic case. Now I see exactly what the voltage is across the super caps. I'll do this to both super cap boards. On ebay, search for "LED volt meter 2 wire" and you will find them in abundance. I'll post a picture of it later.

The new Maanwell PSU arrived yesterday so that too got installed. I haven't adjusted my precharge resistor bank yet to account for the greater current capability of the PSU. I also haven't done any welds to see the improved recharge rates I should get. The theory is the PSU can now keep up with my spot welding cadence so that the super caps don't run down faster than I can charge them.

I won't be deliberately blowing up super caps again, but the chemicals in them corroded the solder on the small components. It turned all the solder black. I've tried scrubbing the small electronics with a tooth brush to clean up the solder, but that didn't work. I have a couple of chemicals that are used for electronics cleaning I have yet to try. Anyway. the new board got all the small electronics coated in conformal. This will protect them from water, chemical or metal bits. I think I will take apart the welder and do the same for it too.

what mean well did you get?

I wanted to try out the welder on a single super cap module like I was doing before on the smaller Meanwell PSU that wasnt keeping up. That's what you see below. The idea was to see if the Meanwell SE-600-12 could keep up with my weld cadence or not. It is doing great. I see momentary voltage drop on the volt meter for a second and then it's right back up again. I think it will do fine.



This is the precharger. Close the switch and you bypass the resistor bank. The switch us 2 pole and I have both poles in parallel just to be sure I have lots of current handling. This is 24 3 ohm resistors in parallel. It's what I had on hand. I could have used any resistor value in parallel. like 1 ohm and 100 watts. I'll make up another 10 resistors in parallel for graceful discharge if I still want it.



This is the lowest the Meanwell PSU will go. Still not high enough to be a significant problem.



I wonder if I was starving the welder previously? This weld was at 60 joules and it popped a hole in the cell...grrr. I've reduced power to 40 joules so it doesn't happen again. Last time this happened I was running at 70 joules.