How to SETUP "Purple" Mini Spot Welder Machine kit DIY

"Purple" Mini Spot Welder Machine PCB Spot Welding for DIY 18650 Battery Spot Welder Kit #spotwelder #minispotwelder
Mini Spot Welding Machine for DIY 18650 Battery 99 Gears Digital Spot Welder Kit
Please like, share, comment & SUBSCRIBE to help the channel

[@simplyfocus9937]

Please like, share, comment & SUBSCRIBE to help the channel

[@PerryAlexander-il5xd]

ahh the power of editing....hmm...wait a minute..!!!!!!! thx for eventually showing me how to set this damm thing up. 2 thumbs up

[@Cdswjp]

Thank you

[@hooperwille]

My suggestion would be to at least double the wires. I also put more solder on the traces and soldered copper brake pipes along the original brass bars. However the main weakest point in this spot welder is the mosfets gate driver (which is actually just a optocoupler). I put a scope to it during welding and voltage drops from 15V to like 4-5V. This will burn mosfets eventually because 5V is too low. You need to cut 1 little trace from positive terminal and solder 1 diode and capacitor. This trick will protect your board from blowing up.

[@cookytrix]

show a photo of what you are saying? or a video?

[@hooperwille]

@@cookytrix basically what I was saying is to beef every copper trace with additional pure copper on top of it. However I burned this board and decided to go the DIY transformer spot welder route. Microwave transformers are almost bomb proof and cheap option. I bought 3 transformers ($5-10 each) and put them in series for maximum power output and voltage. Still working on the control board, now I just use push button to make a spot weld.

[@Sopheris]

I saw one guy on yt connected only (-) of the battery through the board. Plus was directly on the electrode. He added dc donnector to + and - and powered whole thing with separate 12V power supply. That way you can use any battery and not burn your mosfet.

[@jimcoleman52]

The supply (battery) does seem to make a massive difference. I started on a lead acid battery with slightly disappointing results, even with the thing turned up as far as it goes. I swapped the battery for a 430 Farad bank of Maxwell ultracapacitors that I have on hand, and even on the lowest setting it welds significantly more aggressively than the battery did on the highest setting. The wires visibly jump when the MOSFETs fire. The problem is that the limitations of the PCB start showing themselves. My first failure was across the main positive trace on the board, between one of the ends and the reinforcing copper strip they put on. If you look at the board, you'll see that the positive side is one trace connecting the in and out, so you can save the board from having to conduct all of that current by connecting the positive probe directly to the supply. You will still need to give a positive supply to the board to run the logic and fire the FETs, but at least the positive side current won't be running through the board as well. After I patched up my positive side of the board, I turned it up more and had a simultaneous failure at each end of the negative side, between the terminal pads and the big copper piece soldered on to reinforce the current path. So my solution has been to bend 12 gauge solid copper wire into loops, screw them to the terminals, and form the copper to lay along the strips the factory had on and solder them together. I also laid one deirectly across the drains of the FETs and soldered from them to the output terminal. I added a wire on the back side of the board along the trace from the positive input to the sources of the FETs, where there was a clearing in the solder mask but no copper bar soldered on. With the loops of 12 gauge wire on both sides of the PCB and soldered to it, I had to get longer screws to secure the wires. I haven't gotten to the next round of testing after the massive patching / reinforcement around the MOSFETs, but I do say with confidence that the PCB between the terminals and the FETs is a much weaker link in the chain than the FETs themselves are. I haven't yet put an oscilloscope on it to measure the voltage dropped across various portions of the circuit, but judging by heat the FETs are about the least resistance in the circuit, and reducing the already tiny Rds(on) by half will have minimal impact on total circuit resistance with consideration to the amount of heating everywhere else in the circuit compared to the FETs. I know I've wrote alot here, I hope to make a video exploring these things in the near future. I subscribed so I can see what fun you get into being able to spot weld packs. By the way, the "nickel" strip (appears to be nickel plated ferrous alloy) that came with mine measures 0.2mm thick.

[@simplyfocus9937]

Thanks for the reply. I will respond later

[@cookytrix]

still maybe gonna make that vid :) i was thinking to use a server psu i had for mining crypto way back when. i really dont know shit about these spot welders. cheers

[@jimcoleman52]

@@cookytrix I have 2 new spot welders sitting on my desk waiting for me to find the time 😭

[@markb1911]

People sharing their experiences with the various available inexpensive spot weld pcb’s is valuable, as not everyone can afford a k-weld, and it’s a challenge to see if we can get a $15-$20 to actually do something somewhat tricky with high current. There’s not much info on certain pcb’s though. I was actually about to order the purple “99 gears” welder & do the gate mod, but your account along with another shows it does have that weak link in the pcb traces, as does the one I’m currently using so I’ll keep looking. Currently I’ve been using the BIFRC 9-16.8v board with 8 fets, and am currently attempting to reverse engineer for a schematic as there’s almost no technical info at all for it anywhere that I can find. It uses an STC86 microcontroller powered by a 5v regulator. The microcontroller signals what looks like 2 transistors with reversed polarity to drive all gates in parallel. I isolated the control circuit by connecting a 15v-2A switching supply for control, and have the positive lead directly off the battery but even the solder-reinforced negative trace on the fet drain eventually smoked & tore so I cut a copper busbar, removed the fets, (didn’t want to overheat them) and solder pooled them in place. The 15v supply seems to be working well with a 2200uf-25v holdup cap & across the welder input although because it’s isolated I may not need the cap, but, just to be safe. The fets stay room temperature for the most part with 15v driving the gates. However, before the 15v supply I used a 12v supply and promptly blew 2 of the 8 fets, which I removed and left blank as it’s unlikely I’ll find similar fets with matched Rds. The other 6 seem well matched however, and with the busbar reinforcements & 15v isolated fet gate supply, it’s been working for a few months now without issue. I just finished a pack with it. For the welding current I’m using 4 headway 38120 cells in 4s-1p, which could be the reason the traces gave out with only solder reinforcement. From my experience the headway cells push around 200A short-circuit. That may be pushing it, but the welder has been working great on #2 power level. (8 “gears” total) I haven’t had a chance to put an oscilloscope on the gates to see the rise/fall time yet but I’m going to do that in the very near future. Cheers!

[@jimcoleman52]

@@markb1911 the 2 transistors is an emitter-follower pair, a basic class B current amplifier because the controller's pin alone cannot pull the gates up and then back down fast enough alone

[@danburch9989]

I'm going out on a limb and going to recommend one lead directly on the battery terminal and the other lead on the nickel strip. That will force the weld current through the joint. Touching both leads to the nickel strip just melts the nickel strip because that's the path of least resistance..

[@simplyfocus9937]

This little device is working fine. It was a savior. It saved me a lot of money. I would recommend it 👌

[@danburch9989]

@@simplyfocus9937 I've got one one on order

[@simplyfocus9937]

@@danburch9989you will not regret it. It's the best. Get the 4s Lifepo4 battery 🔋, keep it between 12 to 15v when you're using it. You won't have any problems

[@antibrevity]

One mistake they make is to run the positive across the PCB when it's not being switched. To remedy this, it's best to connect both the battery positive and the welding tip positive to the "input" terminal on the PCB. Not only does running the positive current across the PCB reduce performance, but is also costs the manufacturer more as they must run a bus bar or thick trace across the board. That's just dumb. Swapping the positive output over to the input terminal on this board should also avoid voltage drop on the optocoupler, which might help the MOSFETs last longer. @hooperville's mod should also help. They almost never properly protect the drivers on these boards, even though it would likely cost less than running those unnecessary bus bars for the positive output.

[@simplyfocus9937]

I used the device a lot without any modification. It works 👍

[@rahmanocawni5618]

Good presentation. So 30 v pulse rate for 1 second that's what I understand ok

[@joshkelly3743]

I just got one of these but have yet to try it

[@simplyfocus9937]

It looks to me that it can't do 0.2mm. I just purchased 0.15mm they sent me 0.2mm. Unfortunately I will turn it back. I will test it when I get a 0.15mm

[@joshkelly3743]

@@simplyfocus9937 mine just did 3 layers of .12 on setting 11 . But I am using a newish car battery with 735cca . Also I am running it with a separate power supply to run the board .. the positive of the battery is NOT hooked up to the board

[@simplyfocus9937]

@@joshkelly3743 Nice, about 0.2mm

[@joshkelly3743]

It did .2 my battery is at 13v 735cca at setting 30 it kinda sticks .12 on top of .2 to my test cells . I will turn it up after I get one parallel strip done on each series of my pack

[@dd_ranchtexas4501]

@@joshkelly3743 " a separate power supply to run the board" Yep, that's another good way to keep the FETS in their most conductive state - ie channels fully saturated. The welding supply voltage can dip without damaging the FETS....

[@Mihai1174]

You are very smart

[@simplyfocus9937]

Thanks

[@sonnyson7381]

Your battery's current is inadequate. You need more amperes, not voltage.

[@tungtung5872]

if you use 12v 60ah battery or 12v 60c lipo battery you will see the difference. I'm from VietNam!

[@simplyfocus9937]

I was in Vietnam 2016. This battery started my SUV Honda PILOT 2009

[@tungtung5872]

@@simplyfocus9937 kzbin.info/www/bejne/rJLKkHyum9Fsaa8 .you can refer to this video

[@simplyfocus9937]

@@tungtung5872 I watched the video, it doesn't help. You didn't weld any 18650 cell. Why not?

[@tungtung5872]

@@simplyfocus9937 (I don't know your language so I have to use google translate, so I don't understand what you mean). that video is not mine, i don't have 1 video on my channel. but I also used a pcb welder like you and added 5 mosfets. it welds well 0.2mm with only 10ms. 2 supercapacitors LSUC 2.8v 3000F (2.8 + 2.8 = 5.6v/1500F). malectric arduino spot welder like yours, 0.2mm only needs 5ms (2.8 x 5= 14v).

[@simplyfocus9937]

@@tungtung5872 I speak English. It wasn't your video. I was responding to someone else

[@cookytrix]

what about using a 12v power supply? i have a basic old school bench supply and a few server supplies. they put out a lot of power.

[@minkorrh]

No. not enough current. This video is painful to watch and shows me how people get electrocuted daily. Not for the OP's lack of caution but for the simple lack of electrical knowledge.

[@cookytrix]

@@minkorrh ​ so a dps-1200sb-a 1200w hp server psu wont work? also, i didnt really watch this vid when i first posted. yes, its very painful to watch. christ.

[@minkorrh]

@@cookytrix the instructions that came with mine specifically said not to use switch mode power supplies,..... If I had to hesitated guess I would say it's simply because they cannot provide enough instantaneous current. Go to an auto wrecker and pick up a decent car battery. It might cost you thirty-five or forty bucks at the most..... Possibly something from a German car like a BMW as beamers have the battery in the trunk.

[@simplyfocus9937]

The power supply is not the issue, the device is not strong enough without any modification. So don't buy it

[@diyzona]

@@cookytrix No power supply can run welder. you need some 500-1000A of power available in 50ms.

[@derbar5035]

Hello, I bought the same board for spot welding, I am satisfied with the performance and function so far, Do you happen to know how I can turn off this “auto-off” function? It annoys me when I take a long break and have to switch the device on again and again Greetings

[@simplyfocus9937]

Sorry, I am not sure how to do that.

[@manvsbando]

I'm using a 4s 1300mah lipo to power the board and it behaves very strangely: for example I put the power to 1 and it sparked, completely piercing the nickel... Then he started welding very slowly he couldn't even weld the sheet metal on the battery... It's as if he arbitrarily decides the power every time I touch the nickel with the pens... It's unusable... Could anyone help me?

[@simplyfocus9937]

This is a Chinese product, nothing Chinese will last longer. It seams that uou have a bad mosfett.

[@manvsbando]

@@simplyfocus9937 I think that my problem was the incorrect use of the spot welder... now seems to work properly

[@simplyfocus9937]

I am glad it is working for you, I am currently using the MALECTRIC SPOT WELDER

[@petemiddleton64]

car battery works better say 40amp

[@simplyfocus9937]

Check out my other recent video It worked with a better battery

[@simplyfocus9937]

I just made 2400 Spot Welds with it. It is currently saving my life in Haiti 🇭🇹 my other Spot Welder is died.

[@petemiddleton64]

@@simplyfocus9937 best welder ive used other ones dont seem as good ive tryed

[@joshkelly3743]

Your battery is way to low amp .

[@simplyfocus9937]

How do you know the amp in my battery

[@diyzona]

It welds semi good on settings 99. China rubbish