When is Lamellar Tearing a concern?

[MSL93]

I have seen an uptick in customer specifications that state the need to do ultrasonic testing to check for lamellar defects in steel plates. The specifications will generally be vague on the details for what constitutes what needs to be tested. For example, they will say things like "Plates subject to high, localized through-thickness strain" but not list any criteria for what constitutes what is considered "high" or "thick plates subjected to bending stresses" but not list any criteria of what a thick vs. non-thick plate is (greater than 1" thick, greater than 2" thick, etc.).

My question is does anyone know of any good references that I can look into to fill in the knowledge gap here to better understand how to apply these customer specifications?

 

[ANE91]

I ran into this once before and was satisfied by the AISC Commentary on the subject. Ended up dog-earing a triaxial weld to avoid the issue.

 

[David3399]

See also CSA W59 - Annex Q

 

[BAretired]

I thought the problem was found with very large thicknesses, say 80 mm, but the attached diagram suggests it could be much less.

 

[ANE91]

https://www.aisc.org/steel-solutions-center/engineering-faqs/5.3.-other-general-information/#9570 cites DG 21 as a source of knowledge for lamellar tearing.

 

[DrZoidberWoop]

As a parallel, I think "cracking" was thrust into the limelight relatively recently due the the San Francisco Transit Center fiasco. The plate-girder/built-up node had cracks propagating from welded joints at thick plates. I've noticed greatly increased Charpy V-notch testing requirements lately on contract documents from the engineering corporation that designed the failed connection.

San Francisco closes new $2.2bn transit hub after crack found in support beam

Salesforce Transit Center, which opened just last month, is shut down as engineers investigate

www.theguardian.com

 

[JoshPlumSE]

This is the summary research paper that I read when I was in college. It was published in 1973, but my professor still thought it was relevant in 1995-ish when I took that class. In particular it shows a number of examples of details that are susceptible to lamellar tearing versus ones that are "improved".

Commentary on Highly Restrained Welded Connections | American Institute of Steel Construction

www.aisc.org

I should point out that a lot of this is essentially written into the our pre-approved moment details.... At least since the Northridge based revisions to increase ductility. I'm talking about the rat holes, back gouging and such. A good welder that knows how to properly sequencing his (or her) weld passes or buttering the joint before.

It's all about how through-thickness strains can be brittle when that strain is overly restrained. This should be more common with thicker plates because of the thicker welds they require.

 

[David3399]

This paper is definitely still relevant....still in the bibliography of CSA W59 Annex Q along with other valuable references.

 

[LittleWheels]

Defects - Lamellar Tearing

The characteristic features, principal causes and best practice in minimising the risk of lamellar tearing are described

www.twi-global.com

The Welding Institute has a good basic reference and guide.

The threshold for 'problematic thickness' will vary depending on the degree of restraint, the type and size of weld and even whether preheat is used.

EN 1993-1-10 Material toughness and through-thickness properties is the relevant Eurocode standard.
PD 6695-1-10 Recommendations for the design of structures to BS EN 1993-1-10 is linked from the National Annex and has a simple checklist to determine whether a higher 'Z-quality' steel is required (more ductility in the Z-direction).