What's new
What's new

Problems with TIG on Mild Steel


Nov 8, 2008
I'm hoping that you all might be able to share some suggestions for what might be going wrong with my attempts to just 'get to first base' with TIG welding on mild steel or stainless (I've tried both). The basic problem I have is that as soon as the arc starts, everything goes 'fizzy' where the puddle should be. The puddle is more like a volcano. Lot's of spatter, out of control puddle, and black soot around the weld area. It's a total mess! I'm including a picture in this post that shows various attempts to start an arc on 1/4 in. mild steel. The rust and scale on both sides of the plate and the edge were flap disked to shiny steel and cleaned with acetone. Here are the facts about the rig and setup:
  • Miller Multimatic 220 AC/DC rig, running on 240 AC and using High Frequency start (with foot pedal). I started with default settings for 1/4 in. material, but tweaked some settings as described below for the various attempts
  • Midwest Tungsten Service 3/32 Lanthanated tungsten electrodes ground carefully to about 42° (grind marks are axial) and cleaned with acetone
  • #7 Miller cup that came with the rig and all connections were verified to be tight
  • Tungsten stickout about 1/4 in.
  • Argon flowing at about 6-8 LPM. I can hear it and feel it at the cup tip
The tungsten did not melt and it stayed sharp, so I'm pretty sure I didn't dip it.

When I MIG with Argon-CO2 on the Miller rig, everything is fine (nice, clean welds, good control), so I'm pretty sure that at least some of the electronics are working properly. The rig is relatively new. It has only a few hours of run time on it. My tanks of Argon and Argon-CO2 are both basically full (2 kpsi). The Multimatic allows one to simultaneously, yet independently connect two tanks of shield gas, one for TIG (argon, or so my bottle clearly says), and one for MIG (Argon-CO2). I'm pretty sure that the plumbing for both gases is entirely separate. The attempts shown below were done with only the argon bottle sending gas to the TIG torch. The Ar-CO2 tank line had zero pressure, since its bottle was closed, so I'm confident there is no cross contamination going on between the two bottles.

#1. First attempt with settings as described above

#2. Several days later, same settings as with #1 to see if anything would be different on a different day :\ I verified that the shielding gas supply hose was screwed tightly to the TIG port on the rig's front panel and other fittings were all tight.

#3. 9 LPM flow. The HF start did not seem to work, so this one was done with Lift Arc, which had no hesitation initiating.

#4. 12 LPM flow. HF start worked this time (??). Tungsten seemed to begin balling

#5. 15 LPM flow. New, sharp tungsten. Changed material setting to 3/16 (150A)

#6. 15 LPM flow. Material setting changed to 1/8 (120A)

#7. 5 LPM flow. Material setting at 1/8 (120A)

#8. ZERO LPM argon flow. I wanted to see if anything would change with no shielding gas. Material setting changed to 3/16 (150A)

Looking at starts 1 - 7, there is not a lot of difference, save for #3 with the Lift Arc setting, probably because slightly longer on time and a longer arc due to lifting away further from the surface. The result was even more volcanic cratering.

#8 however is significantly different than the others. Instead of heavy black soot, there is the brown crud that, as I understand, typically shows up with poor shielding gas coverage (which makes sense in this case, since there was NO gas coverage).

My conclusion seeing these attempts is that the argon bottle is contaminated, because with all flow rates of the supposed pure argon (low to high), there is significant black soot and an out of control puddle. The black soot did not occur when there was no shielding gas, so it seems to me that the argon bottle has a contaminant that is creating the black soot.

Would you agree? I welcome your feedback and if there are other things I should investigate. I'd really like to get to the place where I can start learning how to TIG weld. I'm stuck right now!


  • TIG_problems_part2_numbered_2024-02-16.jpg
    2.1 MB · Views: 69
AC power maybe but I dont think he reversed the DC, if he did the tungsten would ball up and he said it is still sharp.
He did mention in #4 the tungsten began to ball. I mean he could have the back cap loose and its pulling in air. I've seen that happen to someone before. But I'd definitely say it's between the gas and the polarity...unless he just got blessed with some real crappy material.
Two different gasses to the one machine seems like a fail waiting to happen...........might also be draught blowing the gas away ......doesnt take much ,in fact ideal tig situation of still air is opressive to some people.
Looks like chicken poop, all of them. Possibly funky gas connection or cup angle or cup distance wrong. Not to worry. Takes several big bottles of argon to get it down. For most of us. Some of us need many more bottles to get it down, and big friggin box or abrasives to look good. All starts look a bit sooty. Cleaning procedure?
Disagree there ......Id never tig or mig welded ......bought a 1960s miller tig with bottle of argon at an auction ,and was doing steel and stainless welds from the first bead ......sold the miller as seen working ,and never returned to tig as too expensive on gas for steel or stainless.
I have never had a tungsten ball at the end when welding steel. I've dipped the tungsten, or touched it with the filler rod about half a million times - and got a ball of filler stuck to it, but never balled the tungsten itself.
I'm with plastikdreams, I'd check the direction of flow for the magic pixies first.

Trying to diagnose welding issues really sucks. I was working on a weld job last year; I kept getting sporadic porosity and just couldn't figure out why...

Well, my Lincoln Precision Tig 225 has an air cooled torch with one of those 'gas hose and cable in one' arrangements. Come to find out, the gas hose cracked right behind the torch handle. When I held it up right to inspect the torch and check gas flow - the crack sealed up and gas would flow through the torch. When I held the torch horizontal to weld, the crack opened up and leaked the majority of my argon out behind my hand. 🙄 once I figured it out, I replaced the torch assembly, hose and all, haven't had an issue since.
I have never had a tungsten ball at the end when welding steel.
You have to reverse the polarity to ball a 2% tungsten. That used to be a way to get more heat out of a tungsten when doing aluminum. I dont recommend it since you need to keep switching back to the DC to maintain a good ball.
But your are correct that it should never ball when welding on DC.
Yeah, DCEN, not DCEP. And I'm with the guys saying likely a shielding gas issue also. Tungstens can definitely still ball up on DCEN with bad or no shielding gas. OP I'm sure you've checked this already, but are you positive that you've got the lines from the shielding gas cylinders connected to the correct ports on the machine?
I doubt this to be the problem but....

Just two days ago,
I attempted to tig a steel part off of a Japanese machine, the puddle smoked, sparked and occasionally exploded. After checking the basics I decided to check the weld, it was as hard as carbide to my file.
My guess is that there were a lot of iffy elements in the steel.

Have you tried welding another piece of known origin and alloy?

Just a possibility.

Good luck
Two things to add, just in case. One, I have seen DCEP cause black, sooty welds...'cause I've done it by accident. I don't remember much other than it was obvious something was wrong as soon as I hit the pedal and the results were awful.

Two, can you show us a picture of your torch disassembled, but the collet on the tungsten as it would be during use? I ask this because I had a friend buy a new welder and he was getting black, sooty results and suspected a contaminated bottle. He brought his entire rig over to my house. I set up the machine and pulled the obviously contaminated tungsten to sharpen it, and immediately noticed the collet was in backwards. If the collet is in backwards you'll get some flow, but not much...you can hear and feel it, but it's nowhere near what it should be. It's easy enough to check, just in case.

Lastly, I have had several contaminated bottles over the years, so it does happen. When I get a new bottle back the very first thing I do is try it so it's a much easier discussion when I take it back to the shop if there's an issue. I can say I was welding fine, swapped bottles and suddenly I get black soot. My bottle supplier is only 15 minutes away and knows me, so they don't give me any grief. They told me this happens mostly when a bottle gets changed from something like a C02/Argon blend to straight argon and they don't purge it properly.
Thanks, everybody for your input! I will try out some of the things you mentioned and report back. A few things first to answer some of the issues raised:

1. The Multimatic 220 AC/DC, as I understand, is supposed to handle within the rig the swap of DCEN and DCEP automatically. Maybe it isn't doing that(?)
2. The bottle sticker says clear and plain, Argon
3. Attempts were made in my garage in still air.
4. I will take a picture of the torch and collet and report back. I'm pretty sure that the collet was in properly and everything was screwed on snugly
5. Torch was near vertical to about 10 degrees or so. Distance from tip of tungsten was held at about the diameter of the tungsten as best I could.
6. I'll see if I can try out a known material and even a cold-rolled piece to reduce any mill scale possible contamination. I'm pretty sure though that I was down to bare, shiny metal, that also came up clean when I used acetone on it several times.

More soon. Thanks for your help!