where expertise comes together - since 1996

Web's Largest Portal of Nondestructive Testing (NDT)
Open Access Database (Conference Proceedings, Articles, News), Exhibition, Forum, Network

All Forum Boards
Technical Discussions >
Allusion of Transverse Cracking
Career Discussions
Job Offers
Job Seeks
Classified Ads
About NDT.net
Articles & News

13:02 Jan-22-2010

Deston Henson

NDT Inspector,
Joined Nov 2009
Allusion of Transverse Cracking

Came across what appeared visibly like transverse cracking that were on circular welds on stainless steel vessels to a compressor. These crack-like indications were visible to the naked eye but didn't show up on a liquid penetrate test (because they are superficial?). Any idea of what would cause these false indications?

14:37 Jan-22-2010

J. Mark Davis

Teacher, And Consultant
University of Ultrasonics, Birmingham, Alabama,
Joined Mar 2000
Re: Allusion of Transverse Cracking In Reply to Deston Henson at 13:02 Jan-22-2010 (Opening).

These may in fact be cracks. IGSCC???

Is this an inservice incpection or a fabrication inspection.

If the cracks are shallow or very tight your PT method may not show them. Maybe leave the PT on for about 45 minutes.

You might try an OD creeping Wave, or a refracted L-wave at about 80 to 85 degrees, (Not a surface wave) as a verification. OD Creeping waves have been demonstarted to detect a crack like indication at a depth .001 with a 5 to 1 S/N ratio.


J. Mark Davis
University of Ultrasonics

15:21 Jan-22-2010

Juan Amado

Engineering, Inspection
Arco Industrial, S.A.,
Joined Nov 2001
Re: Allusion of Transverse Cracking In Reply to Deston Henson at 13:02 Jan-22-2010 (Opening).


If these crack-like indications are visible with the naked eye, why do you need to do PT on them, or any oher method for that sake. What information are you trying to gather?

Juan Amado

04:19 Jan-23-2010

chen zhiqiang

Teacher, - - engineering for
Joined Oct 2009
Re: Allusion of Transverse Cracking In Reply to Deston Henson at 13:02 Jan-22-2010 (Opening).

I had the same experience and visual inspection can find transverse crack-like indication on the circufrencial weld of stainless steel,but after PT ,no indication.so I think maybe the indication is too shalow or dwell time is not enough,but I still suggest to repair and re-PT again!

16:20 Jan-23-2010

Deston Henson

NDT Inspector,
Joined Nov 2009
Re: Allusion of Transverse Cracking In Reply to Juan Amado at 15:21 Jan-22-2010 .

zoom image[zoom] Armadillo Cap

I think that we would normally use magnetic particle (because they are surface breaking) but since the vessel is a non-magnetic stainless we tried liquid penetrate. And, of course, we did use shear waves to look for transverse cracking but there weren't any signals that corresponded with each of the visible crack-like lines on the surface of the welds. Two of my (much more experienced) colleagues are pondering the possibility that these surface indications are left over from the submerged welding process where the slag cracked upon cooling and left their imprints upon the weld caps. Has anyone heard of that? Oh, the liquid penetrate test was done twice... with different dwell times.

It would be nice to know with more certainty... in spots, the weld caps look like the backs of armadillos with the quasi-cracks spaced apart (photo uploaded).

16:31 Jan-23-2010

Deston Henson

NDT Inspector,
Joined Nov 2009
Re: Allusion of Transverse Cracking In Reply to chen zhiqiang at 04:19 Jan-23-2010 .

These vessels have been in service for a few years and a repair would need some strong justifications.

In Reply to J. Mark Davis on 14:37 Jan-22-2010

If we can give the creeping wave a try Monday, I'll keep you posted.

Many thanks

17:01 Jan-23-2010

Michel Couture

NDT Inspector,
Joined Sep 2006
Re: Allusion of Transverse Cracking In Reply to Deston Henson at 16:31 Jan-23-2010 .


From what you are telling us, I conclude that it is the first time you and your colleagues are inspecting this vessel. One way to help in this matter is to review previous inspection reports. From the photo you have supplied, I am more incline to agree with your colleagues. I know its hard to draw a conclusion from a pictures but I can't really see anything that resembles cracks. These areas of discolouration resemble what you would see on the back side of "T" weld when two pieces of very thin plates are joined.

The lack of indications from your PT inspection reinforce your colleagues' conclusion.

17:57 Jan-23-2010


NDT Inspector
Joined Aug 2009
Re: Allusion of Transverse Cracking In Reply to Michel Couture at 17:01 Jan-23-2010 .

I cant see any cacks either. weld looks ok.

23:13 Jan-23-2010


R & D, NDT tecniques metallurgy
Joined Jan 2003
Re: Allusion of Transverse Cracking In Reply to Deston Henson at 16:20 Jan-23-2010 .

Dear Sir,

These indications are not cracks. Your PT results are right.

Indeed your colleagues are right that these indications have formed as a result of slag cracking.

Austenitic stainless steels have higher thermal coefficient of expansion and much lower thermal conductivity compared to carbon and low alloy steels.

Sub merged arc welding (SA) is not a popular method for joining austenitic stainless steel because of the high heat input generated in this process that poses many problems in the control of the weld due to slow rate of heat dissipation. On the other hand SA of stainless steel cladding has more acceptance because of the unique fusion pattern without deep penetration.

Due to the higher coefficient of thermal expansion the weld contracts much faster as it cools from the solidification temperatures to about 700° C. The slag starts cracking around this temperature due to the higher differential rates of contraction (mainly in the transverse direction). For Carbon and low alloy steels this happens at much lower temperatures.

Where ever the cracking has taken place (in the slag cover) the atmospheric air enters the crevices and oxidizes the surface layers, giving a distinct hue from which the precise temperature can be surmised. (It can be noticed that the indications are not sharp and some what diffused)

In this case these indications are showing dark blue hues, indicating that the subjected temperature is about 600° C. (Same dark blue hue in carbon steels indicates the subjected temperature is about 350°C).

If you so wish you can remove this blue colour by a simple process. You may take an austenitic stainless steel piece/mesh wrapped in cotton and dip it in a solution of Phosphoric or Oxalic acid (8 to 10 %by volume). This should be connected to the end of a wire joined to the negative terminal of a dry cell giving about 3 volts (two flash light cells of 1.5 volts may be joined). The positive end of the wire should be pressed on to the surface adjacent to the indication. The cotton wrapped SS mesh shall be swabbed on the indication a few times and a bright surface emerges. By this basically the oxidized surface layer is electro chemically removed.

With Best wishes


02:20 Jan-24-2010

Deston Henson

NDT Inspector,
Joined Nov 2009
Re: Allusion of Transverse Cracking In Reply to P V SASTRY at 23:13 Jan-23-2010 .

Hello All,

Thanks to all of you for sharing your valuable knowledge on this... it is much appreciated.

After looking at the photos (under the bright flash and close-up) it does now seem to obviously be mere staining. Although we used a loop, it appears that the photos reveal much more detail / clarity than the naked eye and, in retrospect, I feel that I should have taken the photos earlier.

Nevertheless, it is still fascinating for me & I thank everyone for their input.


06:25 Jan-24-2010


R & D, ndt inspector,engineering
Hebei power,
Joined Jan 2010
Re: Allusion of Transverse Cracking In Reply to Deston Henson at 13:02 Jan-22-2010 (Opening).

hi, a helpful way to indentify whether the indications are cracks or not ,I think , is to observe some of them under a on-site microscope(just like metallographic examination), if the indications are characterized by zigzag propagation path or/and crack branching, they are probably cracks. Be sure that the right polishing agent is chosen to avoid the intergranular corrosion of SS if electro chemical method is adopted. Mechanical polishing method is recommended.
It’s should be noticed that the cracks (just assuming that the indications are cracks) may be destroyed when being polished, since they are too shallow to show up in PT. But if they are IGSCC or stress cracks, the will show up again days later because of residual stress.

jifeng zhao

07:16 Jan-24-2010

Nigel Armstrong

Engineering, - Specialist services
United Kingdom,
Joined Oct 2000
Re: Allusion of Transverse Cracking In Reply to jifengzhao at 06:25 Jan-24-2010 .

Weld ripple removal by mechanical means followed by ACFM or pre-penetrant etch and penetrant will define whether any surface indications are cracks or not.


© NDT.net - The Web's Largest Portal of Nondestructive Testing (NDT) ISSN 1435-4934

Open Access Database, |Conference Proceedings| |Articles| |News| |Exhibition| |Forum| |Professional Network|