gap in weld joints

I have lent my copy out but one must be careful about what welding process is being discussed. We could be talking about submerged arc where often one uses no gap and very high amperages to fuse deeply into a joint. A gap of an eighth of an inch in that situation could result in a very spectacular burn through. In my experience I have found that one cannot suppose that one has penetrated a joint. One must rely on proper fusion to the sides of the prepared edge. A simple gap can just as easily have cold lap on one side or both sides yet show metal on the back side. A square gap between two quarter inch plates is not a useful joint preparation for most processes. A vee preparation with a 1/16th inch gap and 1/16th inch land is something that pipeliners regularly see however. Check to see what process is being discussed. You will find that with the exception of piping open gaps on butt joints are avoided unless there is a backing bar in the bottom of the joint. I only assume penetration as deep as the physical preparation unless the joint is tested or sectioned in some manner to prove greater penetration. That is why on most structural work if 100 percent penetration is required the item is turned over, backgouged to sound metal and welded up. Additionally if you do not fuse completely through a gapped joint you have created a perfect notch for future cracking. Randy
Hi everyone My copy of "Design of weldments", 1963 edition, just arrived. Browsing trough it, on single pass welds it reads: "One fundamental rule to remember: About %60 penetration is all that can be safely achieved with one pass withot backing on a joint with no gap ; EVEN LESS WHEN GAP IS PRESENT (my capitals)". Now, every text I've read so far that having some gap in the join promotes penetration. Being an obedient novice I always make sure my joints have a gap and it really seems to work. What's the bottom line, if there is one?
Regards,
Camilo
-- Posted via a free Usenet account from

They're talking about a vee joint with no bottom, which will take multiple passes to fill the vee, not a flush butt joint. If a vee joint is closed at the bottom then more of the pass will go towards filling the vee. If it's open then your first pass has to be a keyhole weld just to fill the root.
If you are butting two pieces of flat bar together then for sure you'll get better penetration with some gap than without.
You're right and the book's right, you're just not completely communicating.
GWE

Bottom line is yes, no, definitely, and maybe.
It all depends on the shape of the joint you are welding, the rod, the position, and the direction of travel.
Steve

Basically, they are trying to insure that you don't burn through the joint. 60% penetration insures that the welding on the other side fills to the first weld without problems. The backing on a joint will insure that if you do penetrate to 110% then you won't blow through the joint at that point.
-- Yeppie, Bush is such an idiot that He usually outwits everybody else. How dumb!

That is why most structural code welds have backing bars if not "gouge to sound weld" instructions. It is much easier to fuse into a backing bar and you are likely to get consistent quality with this kind of joint. As soon as you have an open root full penetration weld requirement you are paying for a more highly skilled welder, more time preparing the joint, and often some sort of non-destructive testing afterward. If a backing bar or back gouging is not possible you better dig deep into your wallet. I often have to roll up backing rings to fit inside large pipe pilings. They drive a length of pipe into the ground then joint another on top before driving further. Imagine the fussing around and difficulty there would be if the joint was a full penetration open root between the two pipes. You have to contend with weather and fitting in the field conditions. With a backing bar ring in the pipe the joint can have a 1/4 to 3/8 or more gap between pipe ends. All the welder has to concern himself with is fusing into that backing ring. Imagine this on a pipe that is a metre in diameter and up to one inch thick! Keeping things simple and foolproof is a lot cheaper.... and the crew can be hung over :'))) Randy
So the key word is "safely" and one has to backgouge. But what if its not possible to backgouge?
Regards,
Camilo
-- Posted via a free Usenet account from

So the key word is "safely" and in any case one should backgouge and weld again. But what if its not possible to backgouge?
Regards,
Camilo Ramos

So the key word is "safely" and one has to backgouge. But what if its not possible to backgouge?
Regards,
Camilo

Would it be right that a backing bar weld is easiest to trust, especially if the crew is the cheapest and worst bunch that floated by your way? You'd only have to check the backing bar seems fully fused, taking a lump-hammer or crowbar to the backing bar and hearing / feeling what you get, to be sure the joint is fused. ???
With the "60% penetration" double-sided joint, you'd have no easy way to check that (i) the penetration was 60% (ii) even if it was, the crew bothered to gouge that deep before sealing
Thanks for helping the less experienced amongst us to think through the economic issues.
Richard Smith

. Nice trick and I'll be using it. However, and forgive my ingnorance of the real world, but doesnt the rings somehow disturb the flow inside the pipe?
Regards,
Camilo Ramos

Camilo - Randy is talking about structural tubes. He refers to pilings in particular. A tube ("pipe") is often chosen as a purely structural member bearing loads - look at steel buildings. The inside will be just still air. You are thinking of an issue in pipeline - which I have read of as being - have a teeny smooth penetration bead for full strength and no crevice for corrosion while don't cause turbulence in the internal fluid flow. As Randy also says - these pipeline "open-root full-penetration" joints are expensive!
Rich S

Backup rings in pipelines cause all sorts of problems unless it is a noncritical flow. They prevent the "pigging" of a line, that is, sending a cleaning wad down through there every so often.
Open-root full penetration joints are expensive? How so? I welded pipeline, and they are less expensive than with a backup ring. The setup and welding is critical, and not just any brother-in-law or illegal alien can do it, but other than that and the xray, what's so expensive about it?
Steve

Hi Steve
Expense is relative - all welds in steel are "cheap" (???). But steel Open root V butt vs. a backing-ring weld???
More expensive for open root - Set-up more critical - Higher skilled welder - Need hi-tech NDT to be sure it is good (?)
Cheaper for open root - ???
I'm visualising with backing ring you can cheerfully weld with plenty of power, every weld being a "simple" hot well-fused fillet weld. And makes faster wire-fed (flux-cored-wire) easier?
What am I missing? I'm very interested to learn practical issues.
Richard Smith

Me, too. An open root weld on two properly prepared pieces of pipe doesn't take a lot of rod. It does take proper fitup, but with lineup clamps and fitting wedges, it's not rocket science. Lots of pipeline welds are done on rollers, and lineup is pretty easy.
A backup ring isn't THAT expensive, although it would have to be decent quality steel. It would prevent pigging the line, though.
As per operator time, and the cost of that, by the time a company goes through the expense of the land, digging the ditch, getting the pipe there, getting it in the hole, etc, etc, etc, the $$$ a good welder gets isn't that big a factor in the equation.
Pipeline welders are pouring rod most of the time. They don't have a lot of down time. And if they blow too many welds or have too many cutouts, it's hit the shower and catch a chopper time. And don't call us, we'll call you for the next job.
Yes, pipeline welders get good bucks because their work has to pass NDT. But when you compare the labor cost against other costs, it's not the biggest piece of the pie.
One of the things, though, you are missing is the specs. What does the engineering call for? Sometimes, FCAW is not acceptable for engineering reasons. Is it a 5P or 5p+ that is required for the root? A 7018, 9018, or 11018 for the other passes? You can't always do what's easiest. You have to do what the job calls for, and what the inspector wants. Even on the weave pattern of the cover pass. A lot of them are reasonable men, and some are a bigger PITA than a woman going in for a beauty makeover.
Just some musings from what few brain cells I have left .........
Ahhhhhhhhh. Welding on the high seas ................
Kind of like getting to be a pirate, it was ............
Steve

If NDT is called for in the spec., then it will also be needed on a weld with backing strip. Ultrasonic inspection usually used for this type of weld around here.
Fit-up is also a problem when using a backing strip - lining up the 2 pipes without mismatch (hi-lo) being the biggest problem (pipes are never round). A backing strip won't stop this.