Microwave Oven Transformer Spot Welder

Thread Starter

scootye

Joined Sep 18, 2019
6
Recently I followed an instructables on building my own MOT spot welder, and the build went great. I am getting power from a normal home outlet, which is 120V and 15A, and I am getting out around 2 volts, so around 900A for the output of my second winding. I used a large gauged stranded wire for the second winding, and at the end of it, tied so smaller gauge solid wire that is around 2 mm thick as the soldering tips. I started out using it and it was going alright with decent and clean welds, and an occasional spark, with the nickel plate burned with a hole through by one of the copper tips. Now, everytime I try and weld with it at least one of the tips makes a spark, the surrounding weld has a black burn mark, and the nickel strip is burned straight through making a hole, so nothing ends up sticking to the battery well. I have tried reshaping the tips I am using and re-positioning them in so many different ways, but cant get a clean weld anymore. I have also tried changing the pulse time from 10-30 ms with the arduino im using as a timer, but shorter or longer pulses does not prevent the sparks and burns from happening. Any one have any idea why this could be happening and if I just got lucky with the weld I did on roughly 30 batteries?
 
1. the tips need to be clean
2. they should mate
3. Exert enough pressure tip-material-tip
4. Don;t move work and externally trigger the weld.

Commercial capacitive discharge welders I used had the ability to position and if you continued, the welder would activate.

Bigger spot welders same stuff applies.
 

Thread Starter

scootye

Joined Sep 18, 2019
6
1. the tips need to be clean
2. they should mate
3. Exert enough pressure tip-material-tip
4. Don;t move work and externally trigger the weld.

Commercial capacitive discharge welders I used had the ability to position and if you continued, the welder would activate.

Bigger spot welders same stuff applies.
i had cut the tips each time to keep them clean and was exerting pressure on each weld. What do you mean by they should mate. For number 4, I had an external button to trigger the weld. What should I use for a tip that may make it work better? Preferably something I could get from a hardware store, or would I have to order specific welding tips.
 

oz93666

Joined Sep 7, 2010
673
I don't think soldering tips are a good idea , the ones I know are plated , so will change throughout use .
I use copper tips , just the thickest single strand wire used for house wiring about 1mm dia ...
If burning holes the current is too high , try using longer wire to the tips to reduce it ...
Sounds like you're welding 18650s?? , when you have enough of these in parallel they work fine as a power source for spot welding , more relaxing than using mains , I use 100p 3s , but these are reused cells ... new cells 20p 2s or 3 s should work ...
 

Thread Starter

scootye

Joined Sep 18, 2019
6
I don't think soldering tips are a good idea , the ones I know are plated , so will change throughout use .
I use copper tips , just the thickest single strand wire used for house wiring about 1mm dia ...
If burning holes the current is too high , try using longer wire to the tips to reduce it ...
Sounds like you're welding 18650s?? , when you have enough of these in parallel they work fine as a power source for spot welding , more relaxing than using mains , I use 100p 3s , but these are reused cells ... new cells 20p 2s or 3 s should work ...

I dont have enough 18650 batteries to use as a source of power for the spot welder, and the batteries I am welding at the moment are being used to make a battery pack for a electric scooter system. If the current is too high, what else can I do besides making the wire longer to reduce the current? I was already using single strand wire for house wiring as tips and that is what is not working and is burning holes through the nickel plates. Thanks
 

LesJones

Joined Jan 8, 2017
2,361
Some designs use a short pre pulse to soften the metal so the two parts are in better contact before the main pulse. Are you using this method or does you control system just generate a single pulse ? Have you tried reducing the pulse duration ?

Les.
 

oz93666

Joined Sep 7, 2010
673
I dont have enough 18650 batteries to use as a source of power for the spot welder, and the batteries I am welding at the moment are being used to make a battery pack for a electric scooter system. If the current is too high, what else can I do besides making the wire longer to reduce the current? I was already using single strand wire for house wiring as tips and that is what is not working and is burning holes through the nickel plates. Thanks
Copper is what commercial battery producers use for spot welding their 18650s , it must be the best choice ...

Holes means too much energy per weld , either pulse is too long , or current too high , or both ...lengthening the wires to the two tips is the best and easiest way to reduce current .
 

jpanhalt

Joined Jan 18, 2008
8,118
It sounds like you are welding battery tabs. Several years ago, I built an MOT welder for doing that. Instead of arduino, I used a simple 555 timer and zero crossing solid-state switch. My pulse time was similar to yours. Open circuit voltage was about 3 volts with 110V AC input. I controlled voltage to the MOT with a Variac.

It worked well for ordinary spot welds, but I was unhappy with the amount of heat it put into battery tabs. I then went to a capacitor discharge welder for the tabs. That has worked fine now for about 10 years. My CD welder gives very little smut without shielding gas. Here are my homemade electrodes:
upload_2019-9-18_5-53-44.png

I wrapped the upper connector (silver) with insulating tape too. My tips stay clean and rarely need reshaping. They are about 1/4" (6 mm) in diameter.

The smut you are getting is probably due to some contamination; although, you may always get just a little. A shielding gas (Ar) may help.

Edit: I use Ni for my tabs. Have never seen pure copper used on any cell package I have opened.
 
Last edited:

MaxHeadRoom

Joined Jul 18, 2013
19,197
Now, everytime I try and weld with it at least one of the tips makes a spark, the surrounding weld has a black burn mark, and the nickel strip is burned straight through making a hole,
For a number of years I was involved in maintaining factory resistance welders of all kinds.
Generally the reason for the symptoms stated were due to lack of electrode pressure.
Max.
 

Thread Starter

scootye

Joined Sep 18, 2019
6
Some designs use a short pre pulse to soften the metal so the two parts are in better contact before the main pulse. Are you using this method or does you control system just generate a single pulse ? Have you tried reducing the pulse duration ?

Les.
Copper is what commercial battery producers use for spot welding their 18650s , it must be the best choice ...

Holes means too much energy per weld , either pulse is too long , or current too high , or both ...lengthening the wires to the two tips is the best and easiest way to reduce current .

Ok, I was thinking of switching out the current wire I used for my secondary coils to maybe something thicker to reduce current. Also, I have seen on many MOT builds that people first connect they secondary coil wires to solid copper tubes or bars that then have smaller copper tips joined to them. Would using something like a hollow copper bar or copper pipe also reduce the current somewhat and help? In terms of the pulses, I have not been using a pre pulse to get closer contact, but I had varied my single pulse length from 15ms to around 30ms, and the results were that with the longer pulse, the likelihood of a hole through the tabbing was greater, and with the shorter pulse usually the nickel tab would just stick to the welder tips but not adhere to the battery terminal. I would like to attempt to use the pre pulse method as well as making some modifications for reducing current, so what do you suggest should be the time for the pre pulse and then the main pulse in milliseconds? thanks
 
Ok, I was thinking of switching out the current wire I used for my secondary coils to maybe something thicker to reduce current. A
All that matters is the cross-sectional area at the point of contact. R=ρL/A
R= resistance
ρ = a constant for copper called resistivity in ohms-length units
L = length. The length the current has to travel
A= cross-sectional area

At least you can see the effects of changing stuff.

The Total R is going to be a sum of the parts at the various diameters. e.g. the wire, the big part of a probe, the connical section, the mating section.

To do the connical section. It would require a derived formula using integration.

The nickle plated tabs are a different media, so ρ is different.

Total amps will depend on R and contact resistance.

The current density Amp/area depends on the tip diameter. Smaller tip, higher density and thus higher temperature.

Hollow copper will lower the current. It has less cross sectional area than a rod.

Aside: In high power RF transmitters, the output wiring is usually silver plated tubes, but for a very different reason.
 

Thread Starter

scootye

Joined Sep 18, 2019
6
1D21DCD9-4F99-460C-89AF-2B0F2279A553.jpeg E50FABE6-DC52-4749-BEBF-87C3F9ED2456.jpeg Quick update, so I decided to take everyone’s advice and partially redid my setup, now I have a longer cable which I am looping backward for added length, and then I’m connecting it to the two copper tubes that have the thicker electrodes at the end. I haven’t tested it yet but will be posting results once I do. I had one more question though, I was originally planning on attaching the terminal lugs from the secondary coil to the copper pipe with a nut and bolt, but do you think there is a better way to screw it to the copper pipe. I want it secure enough to ensure good contact, and the only reason I thought a nut was necessary was so if it was accidentally yanked on, the screw wouldn’t be stripped out of the copper pipe, let me know what you all think
 
How are you switching the primary? A relay.? Are you running into minimum cycle time problems, or relay bounce? (meaning you are not reducing the closing time when you are think you are)
An ssr may be better.
 

jpanhalt

Joined Jan 18, 2008
8,118
This looks like an SSR in his box:
upload_2019-9-25_5-53-51.png

However, one might be limited, as was my design, by the zero crossing aspect. At 60 Hz, that occurs at approximately 8 ms intervals.
 

Janis59

Joined Aug 21, 2017
819
Never had understood why to smear the fingers with spot welding if batteries nickel are going to be well soldered with normal lead-tin soldering tin by means of 100W powerful soldering hammer applying the one of three fluxes - a) acidum ortophosphoricum concentratum, half droplet b) solution teaspoon of ZnCl2 powder into teaspoon of aqua destilatum, c) acidum aspirinum medicinalis purum (acetylsalicil acid), powder form 1 mm3 under each joint. After solder the wash under running water is obligatory for each of these..
 

jpanhalt

Joined Jan 18, 2008
8,118
1) Spot welding generally requires no cleaning before or after. It is a 1-step process.
2) Less heat is introduced to the cell. NiCd's were routinely soldered. Lithium cells are more sensitive. The heat affected zone is much smaller.
3) Soldering requires a solderable cell surface. Spot welding much less so (like tabs on LiPo cells).
 

chucknnc

Joined Feb 12, 2015
4
I had the same problem and found out the import strips were varying quality. I bought some high quality rolls from these people https://sunstonewelders.com/ . Their welders are for pros and prices as such but the rolls of nickel strip are worth the expense. I can even use my decade old MOT welder that only as a foot pedelDIY quick welder.JPG
 
Top