kWeld - "Next level" DIY battery spot welder

[tatus1969]

G'day,

I've been watching this for a while now.  I had a look at your shop and find "out of stock".

Any plans to assemble more kits?

Ozwolf

I will definitely make new kits after having received feedback on this first batch, and eventually having included the one or other improvement or feature.

[tatus1969]

Ah yes and you solder that wire to the electrodes that will also need the insulation of the wire to withstand the higher temperature.

The electrodes receive a fraction of the heat that is generated at the weld spot, I don't see that we can do anything against this. When welding at higher energies like 100J, they, the electrode holder, and the connecting cable will indeed get quite hot. I alreday have this on my improvement list, because I think that some form of cooling and/or a better heat insulation for your hands is advisable.

available if needed to lower the resistance even further

Be careful when further lowering the resistance. In this application you are basically short circuiting a battery, and there is nothing except battery ESR and the cabling that limits the current. (The power switch is negligible with <0.2mOhm) The welder will go into overcurrent as soon as 2kA is exceeded. And I haven't done robustness testing at this current level yet, only at 1.5kA. The stored inductive energy will almost double from that (E = L * I^2 / 2).

jumpstart his wifes car ... some people that read this will burst in tears.

If I get this correctly if you have 4 supercaps in series and charge it with a controlled Voltage of 10V, still one of the caps might get more than 2.6V due to imbalances in the capacitors?

Exactly. There are two contributing factors to this. a) of course the capacity variation between them, b) variations in leakage current that causes them to self discharge.

[Kjelt]

Thanks for the info I have to dig somewhat deeper in that balancing circuit because 4 or 5 caps are in series you can only charge them with the 40Amps when there is no or little resistance between them, if a balancing circuit uses a fet to switch a resistor across one capacitor the resistor should be very low ohms not to influence the charge current ?
If the resistor is too low it drains the charged capacitor, so not sure how this exactly works then. But that is offtopic so probably open another topic about that 

[tatus1969]

Thanks for the info I have to dig somewhat deeper in that balancing circuit because 4 or 5 caps are in series you can only charge them with the 40Amps when there is no or little resistance between them, if a balancing circuit uses a fet to switch a resistor across one capacitor the resistor should be very low ohms not to influence the charge current ?
If the resistor is too low it drains the charged capacitor, so not sure how this exactly works then. But that is offtopic so probably open another topic about that 

I don't understand that current limit? Is that maybe the current carrying capability of the balancer's PCB track that connects the cells? We need special treatment for them in this case anyway, because they will probably immediately melt in this application. A few strips of solder wick will solve that.

The FET / resistor combination is in parallel to the cell, so it does not influence regular charging. It's only purpose is to reduce charging current when the cell voltage gets too high. For example, if we want to charge the cells to 2.6V, then the balancer can kick in at just this voltage and make sure that the steady state voltage will not exceed this. It does not need to hurry here, and it can "work" with the remaining 0.1V.

Additionally, the balancer should have a beefy shottky diode in series with another resistor to remove any possible negative charge. This can always happen when deeply discharging them.

[Kjelt]

Ah yes I understand now, I am a kind of person that likes to see it (in a schematic) 
Ofcourse the normal charge "chain" is not broken so there is still current flowing through the capacitors. Thanks 

[tatus1969]

The other capacitor option that I want to investigate, has arrived: 30x 100F far east cheapo capacitors (http://www.ebay.de/itm/263097739655). I almost would have bet that they are fake or at least overrated, but I measured 113F capacitance and 13mOhm ESR per cell. Not too bad!

But for now I have to concentrate on the user manual... 

[SvanGool]

13mOhm ESR per cell. Not too bad!

Don't forget that the ESR multiplies with the number of cells, so you are going to get app. max 200 Amps @ 13 mOhm.

[Kjelt]

13mOhm ESR per cell. Not too bad!

Don't forget that the ESR multiplies with the number of cells, so you are going to get app. max 200 Amps @ 13 mOhm.

per string, so five strings would be 1000 Amps. 

[anishkgt]

Yea close to an MOT. Out of curiosity how much max amps does the exiting give out ?


Sent from my iPhone using Tapatalk

[tatus1969]

13mOhm ESR per cell. Not too bad!

Don't forget that the ESR multiplies with the number of cells, so you are going to get app. max 200 Amps @ 13 mOhm.

don't forget that the voltage also multiplies. So if one cell delivers 2.7/0.013 = 207A, then six in series will also do that. And I have five strings available that I can parallel.

[tatus1969]

Yea close to an MOT. Out of curiosity how much max amps does the exiting give out ?

You mean the welder plus the current Lipo? That is around 1500A. I'm not sure if an MOT delivers that?

[SvanGool]

13mOhm ESR per cell. Not too bad!

Don't forget that the ESR multiplies with the number of cells, so you are going to get app. max 200 Amps @ 13 mOhm.

don't forget that the voltage also multiplies. So if one cell delivers 2.7/0.013 = 207A, then six in series will also do that. And I have five strings available that I can parallel.

That is what I said, but I didn't realize you were planning on using all the caps (5 strings) for one welder. 

[tatus1969]

13mOhm ESR per cell. Not too bad!

Don't forget that the ESR multiplies with the number of cells, so you are going to get app. max 200 Amps @ 13 mOhm.

don't forget that the voltage also multiplies. So if one cell delivers 2.7/0.013 = 207A, then six in series will also do that. And I have five strings available that I can parallel.

That is what I said, but I didn't realize you were planning on using all the caps (5 strings) for one welder. 

ok, got it 

[anishkgt]

Yea close to an MOT. Out of curiosity how much max amps does the exiting give out ?

You mean the welder plus the current Lipo? That is around 1500A. I'm not sure if an MOT delivers that?

Yea MOT (microwave oven transformer) delivers much more than that close 1800A. That is of course after rewinding the secondary with 2AWG copper wire.


Sent from my iPhone using Tapatalk

[SirJMD]

A rewired MOT could be a good power source. I don't know how AC will affect the weld though.

I've ordered a 5000mAh Lipo and some various stuff. Will see how this works and continue from there.

[anishkgt]

Using a MOT would not fit the purpose of the K-weld because can be very bulky and heavy. On the other hand it can deliver enormous amperes after rectification and proper rewiring.


Sent from my iPhone using Tapatalk

[Kjelt]

Well the six supercaps and supply will also be beefy as well as a car battery.

BTW received the kWeld today looks very nice  and will start this weekend.
Also the psu deal is going through although the specs are different it is a 3kW psu 0-45V 0-70A 
But it has some small issues so I also need to do some TLC there.

[SirJMD]

Using a MOT would not fit the purpose of the K-weld because can be very bulky and heavy. On the other hand it can deliver enormous amperes after rectification and proper rewiring.


Sent from my iPhone using Tapatalk

The purpose is exactly what the user want it to be. If I want a stationary battery welder, then I don't care about size or weight. If you want a small portable welder, that's your choice

I like the kWeld so far because of the build quality and functionality. Now I just need to find a good and realiable power source. I'm a bit worried that the continued hammering of the lipo packs might be asking for trouble.

[anishkgt]

Then the MOT would be a good choice here as a power source.


Sent from my iPhone using Tapatalk

[tatus1969]

Then the MOT would be a good choice here as a power source.

We would also need a bridge rectifier that can handle these currents, as the kWeld can only switch DC currents.

[anishkgt]

Yes that would be a enormous one to get the job done. Tatus1969 if you remember I had published a video on my channel showing a pcb for spot welding using a MOT https://youtu.be/WAoY9HTdF6o


Sent from my iPhone using Tapatalk

[tatus1969]

the psu deal is going through although the specs are different it is a 3kW psu 0-45V 0-70A 

But it has some small issues so I also need to do some TLC there.

I'm not a native English speaker and wonder what that abbreviation stands for?

[tautech]

But it has some small issues so I also need to do some TLC there.

I'm not a native English speaker and wonder what that abbreviation stands for?

Tender Love and Care. 

[tatus1969]

But it has some small issues so I also need to do some TLC there.

I'm not a native English speaker and wonder what that abbreviation stands for?

Tender Love and Care. 

Thanks 

[tatus1969]

Yes that would be a enormous one to get the job done. Tatus1969 if you remember I had published a video on my channel showing a pcb for spot welding using a MOT

Yes I had watched that video. You are switching the MOT with an SCR on the primary, that is a common approach. This way you can chop the weld pulse into multiples of 10 milliseconds. I didn't go for that because I wanted a welder that has full and fine control over the pulse time, and the idea of maintaining constant pulse energy cannot be realized with MOT/SCR in my opinion. Therefore I went for the harder way - DC supply and MOSFET switch.