I have same thing mine came from a scrap yard
Tim, All of the technique stated above is good. Here is a little more. First a question for you: Have you ever created a cold solder joint? All of my leaks look like that. When you make a deposit that looks like a raindrop on the hood of your freshly waxed car, stop welding and grab your grinder. Grind it down as close as you dare to the parent metal. Resist the temptation to just add more and more deposit over it. Wetting was mentioned above. This is a critical factor, no, it is THE factor. Know this- MIG welding thin metal is impossible, but we do it anyway. You are making a continuous casting supported by the same material with the same melting point, and to make a gas-tight joint you have to melt the surface of the parent metal, daring it to melt through. There are a few secrets that they probably won’t tell you in 3,000 dollars worth of classes. They are never going to get a penny of my money, nor hours upon hours of that which is in most short supply- my time. In addition to those clues revealed above - review now- clean all the rust and paint off 1/2 inch wide at the joint. Orient the work so you are welding within 30 degrees of horizontal. Use an auto-darkening helmet. Close any gaps in the joint. Shine a bright light or good old sunshine on it. Clamp things in place. Grind paint and rust off the place where you will clamp the ground cable. Make sure the rollers are gripping the wire well. Set the wire speed and voltage per the chart. I find that running a little lower than the chart on both the voltage and wire speed works best for me. Set the gas flow high when you are a newbie, especially if there is a breeze. High is 30. Low is 10. One old-timer here likes to run 8, but he’s been doing this for many decades. The gas just pushes air out of the way so your molten steel does not oxidize. Those are the basics everyone will tell you, some of it is even in the manual that comes with the welder. Now here are the insider secrets that make this impossible task possible. Obviously, wear leather gloves. Hold the torch with your dominant hand and hold the copper nozzle with the thumb and forefinger of your other hand. Tilt it so you can see the wire protruding from the nozzle. Position your eyeballs like you are standing on railroad tracks watching a very slow oncoming locomotive. Visualize a horse shoe in place of the cowcatcher, with the lucky side facing you. Lead the puddle to one leg of the horse shoe. Do not let the arc jump off the puddle. Sometimes it does this on its own. Immediately back up the torch to put the lighting back on the original puddle. Bring it toward you and slope down to the parent metal. Watch for the moment the molten steel wets the parent metal and immediately head back up the leg of the horse shoe, round the “U”, slow down at the apex of the “U”, keeping the arc on the puddle. Go along the other leg of the horse shoe, moving toward you, speed up some, stop and wait to see wetting - about 1/4 second- and repeat back and forth, making each new horse shoe a tad closer to you than the one that you made on the previous pass. Watch the color of the puddle. Orange is good. A yellow “snake” writhing on top means it is on the verge of getting too hot. Release the trigger. Hover over the “U”, backed up just a tad. Watch for the cool-down and when there is just a little dull red light coming off the deposit, pull the trigger again. This takes less time than the gas valve timeout, so gas coverage is continuous. Half-seconds count here. Do your welding alone, so you can listen to the arc. Snap, Crackle, and Pop are bad. It’s a sign of either too much current or contamination by oil or rust, or you forgot to open the valve on the top of your bottle. When you are laying down a beautiful professional-looking caterpillar, there is more tape hiss than crackle. And you will hear a melt-through before you see it. Release the trigger immediately. Scrub the area with your stainless steel toothbrush. If the hole is tiny, stick a piece of clipped-off MIG wire in it. If it is large, stick a bright common nail in it. Not coated or galvanized or rusty. Plain steel. Begin the arc on the center of the last good horse shoe and proceed down one leg along-side the hole, watch for it to wet the parent metal, and back off extra fast; complete the other side of the horse shoe and release the trigger to let it cool. Watch the color for the right time to squeeze the trigger again at the top of the “U”. As you make horse shoes over the hole-filling wire, it will become wetted and contribute to the deposit and importantly absorb some of the arc heat that otherwise would have expanded the hole. Resist the natural urge to pull the locomotive toward yourself faster in order to prevent making the blow-through hole larger. If you do, it will at first seem to have worked, but when you leak test it the cold laps and dirty voids will show. If the hole was large you may have to make little horse shoes from the salvaging nail to one side of the hole, then after a cooling pause, do it on the other side of the hole. The cooling pause should only be long enough, approximately one second for your reference, to make the visible radiant heat almost gone when observed through your helmet’s window. Always re-start the arc on a known good point on the existing bead, in the center of the horse shoe. More often than not there is a small amount of contaminant in the bottom of the “U”. The hot puddle will push it ahead (toward you) if you don’t allow the arc to jump ahead of it. A jump-ahead will become a pinhole leak. Immediately move the torch back, on the center of the bead, so the heat floats the contaminant away from the locomotive. You can’t really see the contaminant, but you can see the puddle is misshapen. If all of this sounds difficult, try using stick-welding on sheet metal. In that case you are looking at the flux floating on top of the puddle, magnetism is pulling the end of the stick away from where you aim it, the only way to halt the arc is to pull it away, and you have to chip the flux off before re-starting if you need an air-tight weld. But as was said above, MIG makes poor tacks. It helps a little to pre-heat the area with a torch, but take it easy with that, because heating to the temperature where the shiny metal starts turning blue causes the parent metal to go potato-chip. We have no choice but to tack, and almost every tack becomes a leak. It helps to grind the poorest ones down to half their height. When you strike an arc on it, dwell there to re-cast (melt) it to start the wetting effect. Release the trigger when you see a yellow snake riding the little puddle. Start a new arc when the brightness subsides - about 1/2 to one second is all it takes. Re-start the arc right on top of the bead then mosey over to the parent metal just long enough to see the puddle wet it, then move back to the tack; this is 1/2 of your first horse shoe. You may have to pause at this point to avoid making a hole, or maybe not. You have to watch the color of the puddle. A welder with years of experience can do this on the back-side of a joint where he cannot see. That is not you or me. All of this describes the microcosm. Step back and look at the big picture often. Because this is thin metal, the whole assembly can turn into a potato chip due to contraction of the joint. So we move around, adding a half inch or an inch to each caterpillar. When joining 55 gallon drums you can get away with 6 inch caterpillars. And there you can dwell longer on the edge sticking up (using Wayne’s technique of taking advantage of the drum corrugation) , just long enough to melt it like candle wax, and it contributes to the puddle and leaves a smooth, leak-free joint. On the drum-wall side of the horse-shoe you dwell just long enough to see it wet the surface, then smoothly go back around the horse shoe. How fast? A little slower than would cause the arc to get ahead of the puddle. This metal is thick enough that you don’t need cool-down pauses once you establish a rythm. Nobody learns to play the piano in one day. Don’t be too hard on yourself when you blow a hole in the sheet metal. Could be you need to go down to .023 wire (0.6mm). Pause after each horse shoe until you get the muscle memory. Our work just has to stay together under a slight vacuum and most importantly not leak. -RedOak
12 Likes
This is true poetry about welding 
8 Likes
Hi Mark, I have found clamping my ground to the barrel that is on the inside of the two barrels works best when over laping the metals. I like the locomotive coming down the track and horse shoe with the U shape when welding the puddle. Thank you for the great descripition of welding the thin barrels together.
Bob
2 Likes
There are procedures that can make MIG welding a bit more successful. Mark did hit on many key issues with MIG welding.
As Mark stated. Typical MIG welding ALWAYS starts out cold. The process is squirting out wire to maintain an arc but the parts being welded are cold - thus the melted wire is just piled on the surface. Some quick reversal of feed direction when beginning the weld can help fuse this lump of melted wire. Always try to use small diameter wire when trying to get optimal results with thin material.
One thing that folks often do that tends to back fire is buying consumables based on price. The cost of the consumable is a minor aspect compared to the total project.
Consumables such as shielding gas and filler wire are really the last place you should be trying to save money. Shielding gas such as CO/2 is cheap and works well enough. However - it has more spatter and less wetting action. When trying to have leak resistant welds you want wetting action. So - blended gasses are preferred.
Here in the USA for retail customers we are limited on choices for shielding gasses. CO/2, Argon/Co/2 blends, and Argon+CO/2+Helium blends are pretty much all the consumer has available. I used to do quite a bit of stainless welding so kept a supply of tri-mix gas for MIG welding stainless. It actually works nice for thin carbon steel work but definitely costs more.
I had a gas blender that allowed me to vary the amount of argon/CO2. I found that higher percentages of argon helped provide better results with very low amperages using a transformer MIG welder. Hands down though were the improvements that a high frequency welder provided.
We had a local automotive seat frame manufacturer who was robotically welding the seat frames. They had 18 to 22% rework due to cold joints and failure to wet out. They switched to using dual shield electrode. This is a “flux core” electrode that also uses shielding gas. Rework dropped to 6-7%.
I was using my inverter MIG welder last spring and ran out of solid wire on a job. I put on a spool of flux core. I was using 75/25 Argon/CO2 shielding gas. WOW!!! The combination of electrically conductive argon gas, the metal oxides in the welding smoke, and the high frequency electrical power produced an amazingly hot arc. It was so radically different that I had to stop and get a spool of solid wire to finish the job. I want to try dual shield welding again with my inverter welder though when I have some time. Possibly buy dual shield rated electrode for that attempt.
5 Likes