Schaali1,

Stainless steel weld can respond drastically different than carbon steel.  Stainless steel is an anisotropic material, which makes it very dificult to penetrate with ultrasound.  The ASME B&PV Code acknowledges this in Section V paragraph T-451 which states "it is usually necessary to modify and/or supplement the general provisions in order to test the material."  It is typical that shear waves cannot be used and angle beam inspections must be carried out using longitudinal waves.

We welded a 2" thick stainless steel shell and found an additional 34 dB of attenuation in the weld versus base material.  Since your piping is thinner the affect would not be as great, but significant undertesting can occur if welded stainless steel calibration blocks are not used to establish the test sensitivity.

JR97

I would agree that UT is a poor choice for SS. A calibration on CS for SS would be useless.

I've heard many times that SS welds require "special" considerations when inspecting by UT.  My question now relates to thickness measurements.  Is there much difference between UT thickness readings for CS and SS components when calibration is on a CS block.  The components are typically tanks, vessels and piping with thicknesses in the range of 1/8" to 1".

Joe Tank

JoeTank;
Yes, there is a difference. You should have a set of stepped blocks to calibrate your thickness meter or with UT using an L-wave for austenitic stainless steel. There are differences in sound attenuation between austenitic stainless steel and carbon steel that could result in thickness measurement errors.
 

Joe Tank,

Errors in thickness measurements are solely a result of the test piece acoustic velocity being different than the velocity of the sample used to calibrate the instrumentation.  According to the NDT Handbook, Volume 7 longitudinal velocities of steel range from 5,830 to 5,900 m/sec.  Austenitic stainless steel velocities range from 5,660 to 5,740 m/sec.  If an average of both ranges are used; inspecting a stainless steel component based on a calibration performed on mild steel will result in readings approximately 2.9% thicker than the actual component thickness.

Another concern is velocity variations within thick (several inches)stainless steel forgings and castings.  I have seen velocity variations approaching 10% when the center of the piece is compared to the extremity.  Even calibrating the instrument on the same piece then measuring an area with a different velocity will result in inaccurate readings.

In our shop ultrasonic thickness gaging is a method of lst resort, only used when direct measurement is not possible.

JR97

metengr and JR97,
Thanks for the feedback.  That's pretty much what I had heard from less knowledgeable folks than y'all (2 to 4 percent error on high side for SS when using CS block).   

Joe Tank

Shear wave inspection of stainless steel welds requires use of high angle low frequency longitudinal waves to minimise the beam-skewing effect from the columnar cast structure of the weld  metal. Appropriate welded test blocks with artificial reflectors are also required to demonstrate the path of the sound beams is giving complete coverage.

 

Nigel Armstrong
Lloyds Register
Independent Verification Body Surveyor
 

I recently ran a test for a client. I calibrated to S/S and took readings on a C/S pipe. Then calibrated to C/S and took readings at the same locations. My thickness readings varied .030" on same. That's dealing with materials 3/8" to .5" thick.

API-510 cert # 32890

There are different types of Stainless Steel.  Austenitic is the difficult one to test with UT.

I think some are easier and more similar to carbon steel.


 

The longitudinal velocity of carbon steel and stainless steel are the same.  The shear velocity of the CS and SS are significantly different and will impact the test results. This has been done in the industry for many years, most step blocks are carbon steel with a nickel chrome protective coating.  When you get a chance look a the corners of the step block, if used for a while the coating will be chipped at one or more of the corners.

WBH,

I must disagree with your comment that the longitudinal velocities of carbon and stainless steels are the same.  As I posted earlier, referring to the NDT Handbook, VOlume 7, Table 1, the velocity for 300 series stainlesses are about 3% slower than typical carbon steels.

JR97

JR97

http://www.k2technde.com/downloads/Angle%20Bm%20General.pdf

Follow the link above youmay have to cut and paste into your browser.
The chart used in the AWS training course is typical of those used by Krautkramer. Etc. The longitudinal velocitys are the same. Take in mind this chart covers groups of carbon and stainless steels. It may vary for specifc types and grades of cs and ss. Keep in mind this is a guide used to teach others, not for a technical paper.  I challenge you to calibrate on a carbon steel block at .500" and measure the .500" step on a stainless block and report the measurements. The calibration standards will measure the same. I have done this and know the results. If done on actual calibration blocks it's the same.