kWeld - "Next level" DIY battery spot welder

Caps heating and exploding is caused primarily from reversed polarity; if the caps were installed backwards this can happen even if the wiring is correct.

It's possible, though unlikely, for a batch of caps to have the shrinkwrap on them installed "off position", possibly by enough to cause the installer to put them in backwards without knowing it.

THey can also explode from too great a ripple current, but that's unlikely in this situation--even feeding halfwave rectified unfiltered AC to the caps shouldn't do that, and you're feeding them DC (presumably).
 
skeetab5780 said:
I would just get the Ultra Cap module and use a power supply, its easier than having to monitor batteries at high current and charge

obviously it cost a little bit more but it will last longer

Recommended power supplies?

Thanks!
 
tomjasz said:
skeetab5780 said:
I would just get the Ultra Cap module and use a power supply, its easier than having to monitor batteries at high current and charge

obviously it cost a little bit more but it will last longer

Recommended power supplies?

Thanks!

-Any current limited lead acid battery charger with 13.8V output voltage
-Any bench power supply

I bought this POS PSU off Ebay, it does work but it is junk IMO and the adjustments are very off. I just got it tuned to where the Kweld is happy with it and do not touch the pots now since they are iffy
https://www.ebay.com/itm/133989733448
psu.jpg
 
As already said in the past, 8.2V is too low to get the Caps completely full, if it is a newer kCap with different capacitors on it.
 
Eraser-1 said:
As already said in the past, 8.2V is too low to get the Caps completely full, if it is a newer kCap with different capacitors on it.

Good to know, maybe you should email Kweld since they apparently still list it on the website! You think they would remove it if what your saying is true. I do not know since my PSU goes up to 30v but in my picture it runs at 9.5v
 
9.5V is also OK on my side.

Eraser-1 said:
On my kCap with the SkelCaps I need to use more than 8,1V, otherwise the indication LED's on the PCB of the kCap would indicate that it is not fully charged. I increased the input voltage to around 9-10V and that is fine.
...
Unfortuantely this information is not listed anywhere and on the homepage of Keenlab and there are still the old caps listed and also in the manual.
 
Wouldn't you get more oomph high power bursts from a LI battery pack?

The PSU / charger could continuously top up its SoC over time

What voltage range is ideal for the pack?

Capable of how much peak amps output?
 
john61ct said:
Wouldn't you get more oomph high power bursts from a LI battery pack?

Nothing oomphs quite like supercaps. They can dump 100s of amps without issue and they can last up to 250,000 cycles. Not so sure about their calendar life though, I think its 20 years or more.

Lithium packs are certainly up to the task and just doing brief pulses will take a very long time to exhaust the 300 to 600ish cycles you could expect from them.
 
Sorry I was assuming the caps in place, question is there an advantage to feeding them from battery as opposed to straight off a PSU.

So you are saying a LiPo pack can feed the Kweld without the cap kit?

Would the same be true for a big AGM starter cranking battery?

Is 12V too high?
 
You cannot feed the ultra caps off a battery i did this and blew my first set up…it has a max amp inrush of 70a i fried my board completely hooking it to a 26ah 12v car battery

Bought a new one and a psu now making battery packs is simple as pie
 
john61ct said:
Sorry I was assuming the caps in place, question is there an advantage to feeding them from battery as opposed to straight off a PSU.

So you are saying a LiPo pack can feed the Kweld without the cap kit?

Would the same be true for a big AGM starter cranking battery?

Is 12V too high?
Caps need cc/cv to not damage itself or input device. I quess you can rig something up with battery + something + caps but I see no advantage other than off the grid and posibly more portable.

Lipo and 12v starter batteries can be used directly with welder. Depending how big/high end you go you get more weld amps than with caps and that matters if you are going for copper welding.
 
Yes MOAR is better, would love to get copper thicker than 0.1mm

Also trying to figure the best fixing method, copper-to copper transition from plate/foil to fine-stranded tinned marine wire for the series connections, looking for structurally strong connections stand up to big shock/vibration stresses, not just ampacity.

Thinking getting the strands all flat-splayed out over the sheet, that then folded over, try to spot-weld everything together

without using solder or other less-conductive stuff involved.

Silver-filled epoxy being another path to try, concerned about brittleness though

really like the physical flexibility of a pure-copper join
 
The new version of mine uses SkelCap 0300 with a rated voltage of 2.85V, so 0.15V above the old ones.
 
Here is the datasheet.

Much lower ESR and therefore much higher short circuit current.
 

Attachments

  • 02-DS-211228-SKELCAP-PCB-1B.pdf
    1.8 MB · Views: 53
Eraser-1 said:
Here is the datasheet.

Much lower ESR and therefore much higher short circuit current.

Thanks, what's the max current output on the kWeld with a single new kCap. Mine tops out at 1100A. From datasheet this supercap is looking impressive. Better in all aspects.
 
Eraser-1 said:
The measured current on the kWeld I get from the kCap is around 1650A with the original cables used for the connection between kCap and kWeld.

In the future I will maybe get bigger cables to reduce the resistance to get even more current, but for now it is enough for me.
 
Eraser-1 said:
Eraser-1 said:
The measured current on the kWeld I get from the kCap is around 1650A with the original cables used for the connection between kCap and kWeld.

In the future I will maybe get bigger cables to reduce the resistance to get even more current, but for now it is enough for me.
That's at 8.2V or 8.55V?
 
Charged until the kCap shows that it is full, so more than 8.2V. I didn't remeber the correct voltage, this happened months ago.

But this little higher voltage alone can't explain the rise of the current for 50% more.
 
Thicker and shorter the better but obviously its got to be usable no good having 3cm of 16mm2 what ever probes you use will have a maximum cable size that fits so 6awg is likely the limit if not im sure the hole could be opened out to 4mm if need be but for a 10cm run or so thats plenty thick enough for 1000amp burst current no issue at all every few seconds repeatedly 8awg will work fine too but 6awg is for the difficult.
 
the stock 8awg does get hot when welding with alot of power you need to take a break every cell group or it will even burn your hands. But I have not found it to be much of an issue to wait 30sec every group
 
Tatus1969, do you repair your units? I accidentaly touched one of pins on LCD board with positive electrode and my old kweld died, I have spare and use it, but pity to throw away maybe not so hard to repair (lcd board?price?) unit? I can send it to Germany.
Second thing, I finished new power supply battry, 5s12p A123 anr26650m1b, as expected, even with connectors on power cables I reached overcurrent with 1980A, will I kill unit when using it with that power supply? Longer power leads will make conductance too or only electrode cables can't be long? Adding connectors at electrodes should help?
 
Sztef89 said:
tatus1969 said:
Sztef89 said:
Hi! Today is good day to burn... kWeld ;(
I find that I burn Littelfuse SMDJ28A diode and one place. I want replace by myself...
I can't find SMDJ28A to buy local, but I find only (Transil 33V 3kW SMDJ33A DO214AB / LITTELFUSE SMDJ33A-LF).
Can you tell me if this diode will be OK?
The burnt track suggests that the large TVS diodes at the right might be bad as well. These are Littelfuse SLD8S28A. Your replacement parts can't be used as their voltage ratings are too high. Depending on your battery voltage, you can go to lower voltage devices such as SMDJ24A (Digikey has a few of those).

You're right. There is a high chance that the TVS SLD8S28A is also damaged. I think SMDJ24A will be ok, my battery full chage voltage is 6 x 4.2V = 25.2V, this model can support 26.7V, there will be ok :) Either way I'm not going to use the kWeld on a full charged battery. I don't want to cause more damage... There is too much power on this Leaf modules...
EDIT paragraph: Today I found diodes to buy locally, model SM8S26A

In the meantime, I did an experiment. I discharged the battery to about 23V (about 3.8V per cell). I installed the SMDJ33A diode and started the calibration. Unfortunately, one SLD8S28A diode has exploded... Can you tell me if I put everything together well? (Direction and place) Image is below.

I was tired then... maybe I did something wrong during calibration. I still wonder if I held the footswitch long enough... I dont remember offset number... When "Short!" come on LCD then I touched the electrodes and the TVS diode exploded... electrodes slightly melted :shock: I felt like I drank a very strong coffee :roll:

I must admit that I can replace burnt components, but I don't know much about electronics and how it all works :)

jBvn2LD.jpg


Thanks to you I know that the SMDJ33A diode works with too high voltage (maybe it caused the TVS explosion?). But what if the battery voltage is around 23V, will it work properly?

Thank you for help! I am grateful for your time :bolt: :thumb: :bolt:

Hello

I got similar problem to described above. 3 diodes which burnt because of some wiring problem I noticed too late. I bought smdj28a and 2 diodes- sld8s43a as there was no sld8s28a:/ I read datasheet of it but am not an engineer…. Are these diodes proper for replacement? They will fit instead of sld8s28a without causing another damage in some time? Thanks in advance
 
Back
Top