PIPE FLATTENING TEST ACCEPTANCE PER ASME A530 1

[yamoffathoo]

I am looking for testimonials from test labs (or literature) on the frequency that Clause 19.3.4 acceptance criteria is involked for small diameter A312 TP304L pipe. This will help form a risk argument for the probability of cracks forming during an emergency crimp on a 3/4"nps sch40 pipe.

Here is the quote from A530 Clause 19.3.4:

"When low D-to-t ratio tubular products are tested, because the strain imposed due to geometry is unreasonably high on the inside surface at the six and twelve o'clock locations, cracks at these locations shall not be cause for rejection if the D to t ratio is less than 10."

 

[salmon2]

If you calculate the engineering strains at OD and ID, both at 6 or 12 o'clock positions, you can see the ID strain begins to jump aggressively when D/t is less than 10. And the onset strain at ID surface when D/t = 10 is 12.5%. You can compare this value to your cold-worked 304L's elongation at fracture and get a feel how severe is the deformation from flatenning test. Of course the best way is to do several tests.

 

[yamoffathoo]

salmon2, I have received testimonials from the SANDVIK testing labs in Scranton PA and Arnprior Ont. that cracking is extremely rare.

Would you care to provide a reference for the strain calculation metholology?

 

[salmon2]

yamoffathoo, I didn't use any book, but just consider a long half tube being completely flattenned. The neutral plane is the half radius. Original length for OD, half radius and ID are 0.5*pi*OD, 0.5*pi*(OD-t), and 0.5*pi*(OD-t), respectively. Therefore, the engineering ratio at ID is: 1/(D/t-2). Similarly you can set up formula for OD strain. Plot them with varying D/t and you can see what I was talking about.

 

[salmon2]

Sorry, Repost with Correction:

yamoffathoo, I didn't use any book, but just consider a long half tube being completely flattenned. The neutral plane is the half radius. Original lengths for OD, half radius and ID are 0.5*pi*OD, 0.5*pi*(OD-t), and 0.5*pi*(OD-2*t), respectively. Therefore, the engineering strain at ID surface is: 1/(D/t-2), which yields 12.5% when D/t=10. Similarly you can set up formula for OD strain. Plot them with varying D/t and you can see what I was talking about.

 

[yamoffathoo]

Salmon2,

Thanks, fyi ASME B31.12 Appendix D contains strain calculations for dents in pipes.

 

[salmon2]

Thank you and I will check it out.