TNP #14 - Teardown & Analysis of a Somewhat Dangerous LiPo-Based Mini Spot Welder 

I like that idea. Now I have to find a short somewhere...

and hope it burns the component and not the PCB traces :)

@@tmmtmm Lowest resistance loading will win the firework award. :)

Good for testing the 1st mixer in your analyzer.

Best idea was that of EYE PROTECTION.

1 or 2 short pulses before the main pulse is called 'prewelds' this provides a current path for the main pulse, without it, the main pulse could eject molten metal as 'weld splatter' or even burns.

How did your mod turn out ??

The battery swole. But I took a much bigger cell from a Lifepos4 motorcycle battery and it seems to handle the welds longer. It was a little tricky cause the cell tabs are on opposite ends. I haven't finished the 555timer version yet. Work. 😂

They're assuming you tell the truth. I just bought one of these of eBay, the delivery docket said it was a Tobacco Grinder.

I don't know if you're aware, but you can use salt water to reduce the current, when welding with a car battery. See this video for one example: ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-zkdsq1WdK9g.html

Conversely, a (regular) welding machine also makes a good starter pack for pretty much all diesel equipment with weak batteries. Just gotta make sure you have the voltage and contacts right...

​@@2xKTfcand have to make sure it is a DC welder instead of AC

Actually the performance is crap. The capacitor-based ones running on mains have an order of magnitude shorter pulses. And thanks to Ali, these aren't that expensive either ("HT-SW01" for example).

@@graealex Good to know, I'll have to look into those. To be fair this was the first time I've spot-welded battery packs so I didn't have a frame of reference to compare against.

@@graealex Google and aliexpress search fails to find any welders with HT-SW01 symbol. Is this symbol correct?

@@arekx , Hmmmmmm

@@arekx Yeah, I noticed. Actually needs HT-SW01A or HT-SW01B C D etc and then you'll find the spot welders with 60J to 240J energy.

Did you mean “the ubiquitous red PCB ones”?

@@tookitogo maybe lol

Solder a TVS diode across the MOSFET(s), SMCJ5.0A for example, anode to source, cathode to drain. Clamps the inductive kick. The MOSFETs probably die of repetitive avalanche (causes progressive internal damage). Greatly slowing gate turn-off is also an option; this increases turn-off losses, but it need not be much energy in comparison to the bulk of the pulse. Also, the losses include the inductive kick energy, handling that with less need for TVS.

@@T3sl4 I considered it. Just putting a big fat reverse-biased diode across the MOSFETs turns it into sort of a one-shot buck converter, which should clamp the output spike. A TVS would attempt to dissipate the energy rather than recycle it, not sure which one would be better. I ended up getting a Kweld instead, much nicer welder, if a bit pricey.

@@tookitogo , Probably.

OLED screen?

I think iphone and any other phone batteries are soldered to PCB. Only batteries with steel case need these welder.

@@seldompopup7442 they are mostly all soldered. Usually you would just buy a new or referb battery and install it without welding by clipping it in. However in the case with Apple they embed a serial number and other data that may require the pcb be transfers to a new battery.

@@seldompopup7442 Maybe on the current gens, but the 5/5S/5C I've worked on have tiny connectors.

People have had need of "battery welders" for many years now. Building your own battery pack has been a common DIY thing for years. Started out with RC toys, then moved up to solar systems ("battery wall"), RV's, and even traction packs for cars.

At the very least we know it has a CAT rating so good for a few hundred volts anyways.

Only when wet.

pure nickel or nickel plated steel? Because I own both versions and the older one actually has a battery with lower IR than the newer one (0.54mΩ vs 0.67mΩ). I found it to be roughly at the same level as the old one except my pens heat up very quickly and I suspect they were assembled poorly. Both welders could never really do a real weld with 0.20mm pure nickel even 0.15mm nickel requires being the kind with the slits in the middle to get a proper nugget left on the cell terminal. The board on the newer unit is definitely much more refined and the way they attach the lipo is far superior. If you haven't taken yours apart I would carefully slide the board and battery out just to check to make sure your bolts are tight. Be super careful not to short anything if you need to tighten them.

In china is made the good and the bad, you get what you pay,.in any country

That burglar will never get me alive!!!

I have the dx10 too ...have you been able to fix it??

The case looks anodized, which creates a thin insulating layer. Which is obviously not a good thing to rely on, but would probably reduce the actual incidence of issues enormously.

at 3:34 it looks like he touches them at the same time with his fingers, so my guess is they're insulated

@@dfgoui It's only a few volts. He can touch both terminals at the same time perfectly safely. It the terminals were shorted out by contacting metal, the result would be a huge arc flash or battery explosion. If you can't understand that the previous two sentences can both be true at the same time, you need to learn about the difference between voltage and current. Also, I don't think Shariah would have made such a big deal out of the issue if "they're insulated"...

They are not insulated. They only sit at 4.2V. But they can deliver 1000A.

Follow up, I couldn't get a new board for this so I bought the "upgraded" unit with the OLED screen and the XT90 probe connector....yes it is a slightly smaller battery with two 5Ah cells in parallel vs two 5300mAh cells in parallel in the Seesii model shown here...but the Seesii battery pack gives me an AC IR of 0.540mΩ and the new one gives me 0.670mΩ. Trying to find a way to parallel the two packs together and hopefully it won't be too powerful for the new spot welder which does have a much better looking board. They got rid of the stupid slots the tabs go through and went with a large pad that the tabs are bolted down to with a hole through the tabs and the board pad. Seems like a much better connection (and obviously safer) so maybe the difference in pack IR is a wash after you take the connection on the SeeSii into consideration. I didn't care for the cable and electrodes on the SeeSii unit but the new unit pens heat up VERY fast. I have a feeling its related to how it is assembled but I haven't taken the pens apart because its my current only working welder. I finally got my LSUC caps in from Korea after they did a tour across the Pacific with stops in Mexico, Panama, up to NYC and then getting off the boat in Virginia. Jesus that was brutal to watch. I also bought a CNHL 5000mAh "120C" Lipo from their racing line since it was cheap and I want to try some of the PCB boards. Perhaps I'll end up finding a 1S Lithium spot welder circuit I like and I could turn my Lipo into 3 parallel cells and then combine the four 5/5.3Ah cells from the other welders with it....or create a second 3S lipo out of the best cells from the spot welders if I stick with 12V who knows. Going to take a little bit of money to put the parts together for my cap welder and honestly I now believe 2S isn't enough and I need to go 6S but most of the control boards I've seen are for 2S and 3S configurations so 3S2P might be the way to go?

Nope. Super Caps would not give enough current. People have tried it, and it doesn't work.

Why do you think this? In fact, the thing super capacitors are really good at is extremely small output resistance and therefor high current capability. This system would work only better with super capacitors.

Mine copy is powered by two 3000F super caps in series. Each one is rated 0.5m ohm ESR. It's very easy to reach 1000 amps. The welder consists of four 300A mosfets in parallel.

Very low voltage and lots of resistance of the skin vs. super low resistance of the copper probes. I still would be scared haha.

^ As Axel said. TL;DR: Current and Voltage are two very different, orthogonal things (and Power is sorta another thing).

@@Aexorzist it's not about skin resistance, he was shorting copper to copper. The point of contact between the leads could potentially get very hot and he was holding it with his bare fingers.

@@alexsuykov The point of contact is very small, and there's a large amount of mass to sink that heat. Unless your finger was right on the contact point, you won't even notice it.

You won't feel anything through your bare hands at such low voltages, unless you decide to unwisely try it on your tongue. Even then the above unit (not all brands) only power up and fire the pulse, once the unit detects a direct short across it's probes. I personally do not trust the electrical integrity of anything manufactured in China.

Or of an up-to-date control circuit for spot welder similar to his?

In Dan's new "Workshop Tour" video at 32:08, he refers to a "miniature spot welder" that "works exactly as the big one" you reference. The miniature one is made by Weltek. It looks like a WELTEK 410E PRECISION WELDING HEAD, with a separate Weltek solid state power supply.

@@d.a.nordquest3082 Good find, although, I'm not sure if "works exactly as the big one" means that the internals are the same architecture or just that the principles of spot welding are the same, just in a smaller package.

But it may make you stronger in the process :P

In my case lasted the first test, it was ok, but stored it 6moths and it wont weld nevermore, no even 0.1 strips (pure nickel)

Minimal pulse length might be a problem...

And you have to use a big one so that the contact dont Weld themselves shut. I have tried it with a old 20A 3ph ABB contactor and a car battery, with all contact in parallel, it did work for a few times and then the contacts diddnt open up and it wrecked the cell I Was welding on and gut the cabels nice and hot before I could react. I would not reccomend unless you go with one rated for 40A or higher, and even then I would Still recommend being care full.

im guessing best way would be to take one cell and estimate what its short circuit currrent pulse capability was , maybe using one of these spot welder control boards for that even. You put in 1 cell, test the short circuit current. Then another cell in parallel to see by how much it went up ... hopefully that will let you null out the internal resistance and calculate the ESR for short pulses per cell.

You need to find out the "C" rating of the batteries you get. Search for "battery C rating" for more info.

@@Mueller3D jup, youd wanna get very high C rating batteries, the ones they put in those crazy RC packs that can do 200C or something silly.

You need to consult the data sheet of the specific battery cells that you intend to use. Note that they have lower voltage as well.

I would wager that the LiPos are cheaper than a supercap setup that would require more than of the shelf components on the power management side.

@@madinatore The problem is lithium batteries are not designed to repeatedly be short circuited so eventually they will short internally and catch fire.

@@cda32 - Those cells are more than capable of coping with that sort of current.

Ha!

LiPo stands for Lithium Polymer

New nuclear/radioisotope battery? Nice

It is alright. Why do you ask?