There is also frequent reference (particularly with radiography) to 'centre-line shrinkage' where there is a small amount of concavity in the centre of the root bead area, due to differential cooling / 'shrinkage'.
Not a problem as long as concavity is not out of spec and compensating cap / re-inforcement results in a weld being thicker than parent material.
Quadswift
I think the term is somewhat misued in that most of the time it is not defined as to location. The last I saw it used was by a Radiographer on 22 6" Sch 10 S/S welds that had been butted and single passed with TIG. He knew it was LP but couldn't believe it was in every weld.
It can be used for Lack of Penetration. Lack of Fusion on one side of the butt especially on horizontal welds, either butt/groove with land. The usage is quite common with autogenous tube/pipe welds.
Definitely centerline
Why is all the powder hanging around the toe of the weld? I would take a 4" grinder and make a light pass and then make another pass with the powder.
That is what we used to call a Ray Charles Crack (no disrespect intended).
I did grind the weld, the crack extended toward the root about 75% through the weld.
The crack was casued by a weld metal issue coupled with very high heat input. (I can say too much about the weld metal issue but 0.25% of an alloy makes a big difference on solidification segregation.)
Just a thought from an old welder. Flux-core,looks like the weld was "pulled". You may want to try "pushing". Maybe just alittle uphill if possible.
The weld was pushed however the gun angle was not optimum.
The weld had unmelted flux in the center of the weld.
With this lot of filler, I can reproduce the centerline cracks by either pushing or pulling. Once the weld metal chemistry and flux composition were slightly changed (still within AWS specifications), I could not produce the crack.
A little off thread
Being an older Welder and Inspector the reason for mentioning the edge is that in early times of MIG welding we had 44 C/S filer bodies fabricated using Pipe and 12" 1500 # WNF. All passed 100% RT and spot MT. After about 10 Yrs service we noticed cracking at the toe (both sides) of the Grove welds at the flange. All filter bodies in service had the same cracks. In fact all MIG welds on the filter bodies had this type cracking. The toe wasn't rolled over as much as you show but you could see where considerable grinding had been done, these areas were cracked also.
The service was 6 weeks at 600°F then 1000°F for cleaning. 2 to 3 months off then another cycle. The internal pressure at this weld was 20 psig but the hydraulic end force on the flange was 2000 psig.
We had two vessels that had never been in service, only stressed relieved, and upon planishing the weld we found a considerable amount of thumbnail cracking at the toe, but also several continuos cracks 1" or better.
Toe dressing with a grinder is a pet peeve of mine. I have seen more damage done than would have been done it left alone. If the toes need to be dressed I perfer that they be dressed with GTAW (no filler), if possible. The welder should be trained to avoid this rework.
Having a good inspector on the shop floor when welding is occurring is the key. Budget cutting indirect labor (QC/QA/Inspectors etc.)to the minimum, very few shop have enough qualified inspectors. Looking at a weld after the fact, you do not know, current, voltages, speed , preheat teemperature, electrode angle etc. One needs to stopping the problem before it starts.
Thanks for the links--very interesting. I also understand were you are coming from. Living in the pressure vessel world I am very cautious.
ASME does allow for deletion of filler metal-However it is an Essential Variable and your procedure and welder must be qualified for it.
My concern(esp.in your example)would be a slag inclusion under the toe of the weld. Your links mention a 1/16"t remelt. I believe this toe would need to be ground away first then blended with the remelt.
How would you qualifiy your procedure and welders for this?
Rich2001,
You didn't read my post very closely. If you had you would have noticed in the last post I used the word planished even though I did use the word grinding also. I assumed you would have realized never to grind with the axis of the weld except on straps. One should always grind with the centerline axis of the grinder parallel to the axis of the weld if you have to grind. Or better yet don't put notches in a weldment by grinding, filing or otherwise.
The overlap still bothers me and along with all the people at this end who have looked at the picture. All here agree that if this were a real weld the overlap would require investigating.
We normally see centerline crack as you have produced where there is considerable restraint and not enough material was deposited.
As you are into structurals’ look at the Northridge Earthquake Report and check some of their observations.
We make and use a considerable number of stub shafts that require machining during manufacture and tried every type of welding available in house and outside contractors (world renown) and we still haven't been able to completely overcome the thumbnail problem with GMAW. Analysis say they aren’t bad but in service observations prove otherwise.
The overlap, had this been a production weld, it would bother me a lot. It must be removed and repaired and welder must be retrained, if not let go, period! There is no excuse this this type of workmanship from experienced welders.
The Northridge earthquake report, along the Fema 350, 351, 352 & 355A through F all contain good information.
The photo - I keep a library of photographs and samples that I use to train NDE technicians and welders. In the case of this weld I was attempting to create a centerline crack after have production problems. Porosity, undercut, overlap and lack of penetration are easy defect to create, cracks are harder, so when I stumble on a technique I take full advantage of it.
The unmelted flux in the weld is the key to solving this problem. By being unmelted we did not get the welding alloy transfer from the flux to the weld pool.
BTW the crack does not break the weld surface. There are no PT indications, weak indications with AC MT and much stronger with DC and Pulse DC MT.
