Complete failure of a TIG weld...

I tried to TIG weld a 1/2" pipe nipple, to a 16 gauge sheetmetal piece, so that the pipe nipple is perpendicular to the sheet metal.
The objective was to make a little stand on which I would mount a VFD.
I used about 100 amps of current, DCEN, and ER70 filler rod.
The temperature in the garage was about 40F.
The result was a decent looking TIG weld, nothing to brag about, but passable.
However, as soon as I stopped welding, the weld zone began to cool and started making little popping noises. After a minute or two, I whacked the 16 ga sheet to straghten it, as it was warped a little.
After just one or two whacks, the sheetmetal piece FELL OFF, basically cleanly, with the entire weld remaining attached to the pipe nipple.
I was completely shocked. After looking at the failure area, I had an impression that some cracks developed UNDER the weld and a part of the sheetmetal stuck to the weld, so the weld, sort of, tore off a part of this sheetmetal piece, as if it cracked in the middle.
I feel kind of fortunate that this was never meant to be a structural piece, but I get some jitters just thinking about what would happen if it was.
Would anyone have any clue as to what it could have been.
i
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
I'll suggest that the sheet metal was high carbon, cold rolled stock. Your weld set up a HAZ in the sheet metal that cooled very rapidly, made it extremely brittle. Since the pipe was much thicker, it doesn't get anywhere near the stress levels that the thin stock will see. The popping noises were the stress cracks starting to develop.
You could test this scenario by heating a chunk of the sheet metal to a high temp (yellow orange) and quenching hard in water. If it gets really brittle (snaps when bending it 90 degrees with a pliers), you have found the culprit.
We had a project that used 1/4"x2" bar stock that got welded and then cut through the weld. Worked fine for years, one day we tore up 3 bandsaw blades in an hour. I traced it down to ONE STICK of bar stock that had a carbon content well above the normal hot rolled.
Ignoramus8601 wrote:

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

The sheetmetal was definitely hot rolled, with typical mill scale.

Definitely.
Yep.
i
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Mon, 07 Dec 2009 21:31:38 -0600, Ignoramus8601

When welding thin stuff to thick stuff, preheat the thick stuff first. With Tig, it's easiest to preheat with Oxy-Acet.
A GOOD tig welder can get the base metal of the heavy peice up to temp with just the tig torch without burning the thin material through, but it's tricky and takes more than a little practice.
Also, shop should be above 60F with still air. At 40F the joint will cool fast.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

i remember seeing a photo of a failed weld like that. was the mounting bracket for a joystick on a (steve) wittman designed aircraft. was shocking to see. the article said the joystick fell off in his hand and the guy managed to fly it safely to a landing, i think it said just as night was falling. ugh huh?! scary. i'll let someone more expert than me try to answer your question. i'd think it would have something to do with the alloy of the sheetmetal.
b.w.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
I've stick welded using Nickel rods flanges to 1/4" sheet metal. So 1" pipe could screw in.
I did the welding on refractory bricks - and used USA made flanges. The made in China ones melted like butter and fluff.
I did four welds - doing it across from each other - should have done 5... I had a large weld to the sheet and it melted into a wide section of flange.
I could take a pipe and attach it - wailing it on the ground. I used a hammer on the close end - they all worked ok - and delivered two sets of four.
The flanges were something like cast iron. USA was cast iron. Others were junk metal.
Perhaps mine cooled slower being on brick. The squares (round corners) were 12 or 18" on a side. Been a while.
Martin
William Wixon wrote:

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

I have seen thin wall tubing do that when welded to a piece of plate. You need to keep the heat on the thick part almost entirely, and let the accumulation of the molten puddle wash down on to the thin metal until you "see" fusion. It is best done in a flat position. But then, at that moment the hotshortedness of the metal will occur in an instant, and if you're too slow, it will blow a hole. Hotshortedness (sp?) is when a metal collapses under its own weight when it reaches molten temperature. Either that, or maybe there was some coating on the sheet metal that created a barrier to fusion. As suggested, I'd try preheating, and doing it in a little warmer circumstances.
Hot-shortness on Google: ('ht 'shrtn?s) (metallurgy) Brittleness, usually of steel or wrought iron, when the metal is hot, due to a high sulfur conten
Steve
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Steve, not to contradict your instructions, which sound like what instructors have told me about welding thin to thick pieces, but that's not actually what hot-short means. Hot shortness occurs well below melting temperature, and it's a complete loss of strength in the metal at that lower temperature. It's taken advantage of, for example, when breaking up iron castings into smaller pieces for scrap, often for feedstock in casting. When the metal becomes hot-short, a whack with a hammer, that ordinarily wouldn't do a thing, suddenly is able to break the metal into small chunks.
It's brittleness, as the definition says, not a molten state.
--
Ed Huntress



Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

I had heard the term used when aluminum is there one instant and on the floor the next. Heated past the melting point, then turns to liquid and collapses. I had never heard that sulfur content had anything to do with it, tho. Either way, I'm sure there's a reason it didn't stick. Has to be. I'll ask my old welder friend tomorrow and see what he says. He's pretty sharp.
Steve
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

That's true, but it's because most aluminum alloys have distinct liquidus and solidus temperatures that are very close together; they have no "pasty" range, or none that you'd notice. You can equate this to various solders; some electrical solders behave just like that, but the alloys made for wiping plumbing joints or for automobile body solder are the opposite -- they get pasty and stiff, and then pasty and runny, and then they finally slump and run.
The other reason this is remarked about with aluminum in welding applications is that the liquidus temperature is below the temperature at which the metal glows red, so you have little warning when it's about to turn to liquid and collapse. People who weld it a lot become sensitive to other signs, mostly surface appearance. With steel, the red, then yellow glow help you to see where you stand.
Some aluminum alloys are indeed hot-short, but I don't think it has anything to do with sulfur in this case. It may be the copper in 2000-series alloys. It's one reason that 2024 is so difficult to weld successfully; as it cools and shrinks, it quickly passes through the hot-short range, and that's when it's likely to crack and tear. It's also a problem with some grades of steel; which ones, I don't recall.
Terminology in these trades is kind of a mess, and if someone is calling the sudden melting of aluminum "hot shortness," they're misapplying the term. It happens when the metal is very hot but still in the solid state. It's a tearing or cracking, rather than melting or slumping.

As for the sulfur, I don't remember. It's been years since I was really up on the details of metals behavior. And I'm a pretty crude and amateurish weldor, so I can't help there, either.
--
Ed Huntress



Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Not to worry. I've seen lots of darn good welders who couldn't carry an intelligent conversation about metallurgy or terms but could pass x-ray every time. Not to mean that you aren't conducting an intelligent conversation, but really, even "crude and amateurish" weldors can get stuff to stick, and stick for a long time. After that, it's aesthetics, unless it's for inspection. I've found some old stuff welded together that's been stuck for decades. Gorilla welds, I call them. But they're still stuck fast. My dad used to call them mud daubers.
Steve
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Steve B wrote:

In the steel industry, hot shortness is used to describe failure during hot rolling where a small amount of liquid forms at the grain boundaries and the material falls apart. It doesn't take much liquid and can be a couple hundred degrees below the melting point. I doubt that this is the case here as it is hot rolled stock, and mills don't like it when a several ton billet flies out in small bits. (Well ok, it's poetic licence.) Sulfur is usually the culprit in these cases, but there are oddball elements that can cause a similar effect.
The advice on heating and quenching is a good practical test. It is likely a higher carbon content. What was the result?
I've lurked here for years. I taught a course for a few years on the metallurgy of welding and it helped me to read the adventures of welders and get a sense of reality. I'm happy to finally be able to chime in....
Ken
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Ken, I have retained that piece and the test seems easy enough: get it red hot and plunge into cold water, then bend.
I will take some pictures, one of which would be of it prior to testing (but with evidence of a failed weld) and then the results of testing.
i
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Dec 8, 5:19pm, Ignoramus10187 <ignoramus10...@NOSPAM. 10187.invalid> wrote:

I posted a simple demo of hardness vs elastic modulus once that involved bending hard and annealed bugle head screws held by their points in a vise. The difference was very obvious between the two, but without an identical reference it's difficult to compare. In this case you could cut three pieces, harden two and anneal one of those, then bend all three. The annealed one represents what you can do to fix it after welding.
jsw
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
wrote

Good to see you here, Ken. You might be interested in this brief piece on hot-shortness. It's all a little fuzzy:
http://adsabs.harvard.edu/abs/1980RSPTA.295...89M
--
Ed Huntress



Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

SteveB, I thought about this overnight.
I think that this steel sheet was a bad specimen, and it had lamination issues and some sort of hot-shortedness or high carbon content, in the local place where I attempted welding.
SteveW, it was definitely hot rolled, with plenty of mill scale (which I wirebrushed away prior to welding).
This is very scary. I thought that a half decent looking TIG weld was guaranteed to have fusion. Now, I do not know how to tell a good one from a bad one.
i
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
wrote:

how about hit them with a hammer?
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Well, there is always doubt as to whether I hit it hard enough or not.
i
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
wrote:

Just give up then, testing stuff is too hard, don't even bother.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

would you consider doing exactly the same thing an inch or so away with some of your tig compatible bronze and tell us if you get the same failure? Seems there was a thread with that in it just lately around here.
not everything is weld able. There are ways to make sparks off metals and examine what you get but sometimes just the four and a half inch grinder and a piece of known structural mild steel as a control can give you an idea if what you have isn't mild steel. The first guy also told you how heat it quench it bend it.
Fran
not cross posted
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Polytechforum.com is a website by engineers for engineers. It is not affiliated with any of manufacturers or vendors discussed here. All logos and trade names are the property of their respective owners.