Use of Ultrasonic Phased Arrays for Examination of Austenitic Steel Welds Santanu Saha Technical Manager, Non-Destructive Testing Intertek INSPEC

Introduction Air cooled heat exchangers (stationary & floating header boxes) manufactured in accordance with ASME Sec VIII Div. 1 2013. The base metal weld of the material (SS316L) contain large grains and coarse columnar grains in the weld. Weld geometry (Corner welds shown below) prohibits satisfactory use of Radiography. Conventional Ultrasonic Testing do not provide satisfactory results due to severe scattering and attenuation of sound waves in the coarse grain austenite. Corner Weld geometry

Stainless Steel Weld Characteristics Large elongated anisotropic grains (dendrites), often forming an ordered columnar structure, which are characteristic of the such welds. This anisotropic grain structure can lead to scattering and attenuation of ultrasonic waves, contrasting with the isotropic behavior of homogenous welds made in carbon or low alloy steels. Weld Microstructure showing coarse grains

Conventional Ultrasonic Testing (UT) Manual UT was not successful due to the following: Austenitic Stainless Steel has highly coarse grain microstructure. The weld is made of Stainless Steel electrode, which provided even coarser grains and preferential orientation of grains (dendritic) having cast like micro structure. Inadequate access for scanning

Ultrasonic wave propagation in Austenitic Microstructure – Detailed (1) The average grain sizes of the Austenitic welds are about 1/10th or higher the ultrasonic wave length deployed for such testing The ultrasound propagation is sensitive to the angle of the wave- front with respect to the grain axes. This angle changes dramatically as the wave mode under consideration is changed from the common vertically polarized shear waves, to longitudinal waves and to horizontally polarized shear waves (SH). When traveling through this material, the wave-fronts are not generally at right angles to the beam axes, causing the effective direction of the beam in anisotropic weld to differ from the nominal beam direction.

Ultrasonic wave propagation in Austenitic Microstructure – Detailed (2) In practice, for weld inspection it is not possible to predict the beam width except for simplified structures. This means that defect size estimation by techniques which rely on knowledge of the beam shape (e.g. 20 dB drop method) will not be satisfactory in situations where the beam is distorted in its path through anisotropic weld metal. Due to variations in beam shape, amplitude methods for defect evaluation are less reliable for austenitic welds Reflection at the base metal – weld interface, refraction at the interface and mode conversion at the interface can occur and produce spurious indications which may hinder proper interpretation of indications.

Ultrasonic wave propagation in Austenitic Microstructure - General Depends on grain size and preferred orientation of the grains with respect to fusion faces of the welds. Coarse grain cast structures have marked effects, leading to increased scatter, attenuation, variations in sound velocity, and often to beam-distortion. Many of the ultrasonic characteristics of austenitic welds derive ultimately from the anisotropic elastic properties of the columnar grains which form the weld.

Procedure for Ultrasonic Phased Array Ultrasonic scan plans have been generated to understand the covered volume of the welds with longitudinal wave, shear wave angle and normal beam probes at different positions. Scan plan below is for a 10mm to 20mm Tube sheet to top and bottom plate weld geometry:

Full Scan Plan

Weld Geometry

LW Wave Scan

Results of Ultrasonic Phased Array Tests Flaw detection by LW Wave (sectorial scan 350 - 600)

Results of Ultrasonic Phased Array Tests Flaw detection by LW Wave (sectorial scan 350 - 600)

Results of Ultrasonic Phased Array Tests Flaw detection by LW Wave (normal -300 - +300)

Advantages of Phased Array (1) Longitudinal sound beam used in this welds can travel through the large austenite grains and detect the flaws which are at the opposite surface of the weld. Beam focusing capability at pre-determined depth enables high intensity beam reflection from the flaw and thus a good response from the flaw can be obtained. Unique capability of beam steering and right focusing to the intended flaw depth which enhances the detectability of flaws at various unfavorable orientation. Accurate positioning of flaw by encoding. Automated data acquisition controlled by powerful software which makes it possible the reliability, repeatability and reproducibility of the test.

Conclusion Convectional Ultrasonic examination of austenitic welds is difficult, if at all possible. The weld type is mandatorily required to be examined by NDT volumetrically and weld geometry excludes use of Radiography. A successful demonstration and qualification of the procedure was possible to meet the mandatory requirements of construction code (ASME Sec VIII Div. 1) Use of semi-automated Phased Arrays was successfully implemented throughout the project

Contact Us Santanu Saha Tehcnical Manager - NDT Intertek INSPEC Santanu.saha@intertek.com +971 50 481 4962 Catalin Tomescu Director of NDT & AIM , MENAP Region Intertek Catalin.tomescu@intertek.com +971 56 401 5392

Thank you!