Difficulties in Ultrasonic Testing and Phased Array Solutions of Austenitic Welds

Difficulties in Ultrasonic Testing and Phased Array Solutions of Austenitic Welds

Author: Yun Weirui, Lin Jiaqiong (Application Lab) 

1. Difficulties in Ultrasonic Testing of Austenitic Welds

Austenitic stainless steel refers to stainless steel with austenitic structure at room temperature. When the steel contains about 18% Cr, 8%~25% Ni and about 0.1% C, it has a stable austenite structure. Austenitic stainless steel includes the well-known 18Cr-8Ni steel and high Cr-Ni series steel developed on this basis by increasing the content of Cr and Ni and adding Mo, Cu, Si, Nb and Ti. Thanks to the characteristics of good room temperature, low temperature toughness, weldability, corrosion resistance and heat resistance of austenitic stainless steel, it has been widely used in petrochemical, marine and nuclear power industries, as well as deriving ultrasonic testing of austenitic materials, especially austenitic tubes and flat welds. Fig. 1 is a material structural diagram of austenitic stainless steel rod and welds.

Compared with ordinary steel welds, ultrasonic testing of austenitic stainless steel welds is difficult. The main difficulties are as follows:

1.1 Coarse grains cause beam scattering and acoustic energy attenuation.

1.2 The structure is anisotropic and in columnar grain arrangement, which has strong scattering and twisting effects, resulting in the change of velocity and the attenuation of acoustic energy.

1.3 There is an obvious heterogeneous interface between the weld and the base metal, which causes the interface reflection to generate cluttered signals and reduce the signal-to-noise ratio.

1.4 The structure of the weld is greatly affected by the welding process and specifications, so the material and welding process of the calibration and simulation of test pieces are required to be consistent with the actual test object.

2. Phased Array Ultrasonic Testing for Austenitic Weld Inspection Difficulties

In view of the influence of austenitic welds on ultrasonic testing, the phased array ultrasonic testing technique is introduced for the testing below.

2.1 The influence of coarse grains and the testing method in use.

Based on the microscopic observation by metallographic microscopy, the average diameter of austenitic weld is between 0.13 and 0.18 mm, as shown in Fig. 2.

The ultrasonic acoustic energy attenuation is subject to the ratio of the grain diameter and the ultrasonic wave length. When the grain diameter is close to 1/10 of the ultrasonic wavelength, there will be obvious sound scattering. When it reaches 1/2, the sound scattering increases significantly, and ultrasonic testing cannot be performed. That is to say, The use of austenitic weld testing shall try to ensure that the wavelength be greater than 1.8mm and at least not less than 0.9mm. Therefore, firstly the frequency of the probe shall be selected according to the material of the test object and the length of the weld to be penetrated. In addition, for ultrasonic waves of the same frequency, the wavelength of the longitudinal wave is nearly twice that of the transverse wave. Using the longitudinal wave is the most effective way to increase the wavelength. Fig. 3 is a phased array probe with a frequency of 5 MHz. The transverse wave (wavelength is about 0.64mm) and longitudinal wave (wavelength is about 1.18mm) are used to detect the austenitic weld comparison test block and the φ2 side-drilled hole opposite to the weld. It can be seen from the figure that the transverse wave detection effect is extremely poor, while the longitudinal wave can basically detect artificial defects. 

The tests have shown that only longitudinal waves can be used for ultrasonic testing that needs to penetrate the weld length of more than 15mm. After the longitudinal wave is adopted, the secondary wave detection is not applicable due to the waveform transformation of the longitudinal wave at the root of the weld, see Fig. 4.The primary wave detection cannot cover the near-surface area of the weld, and there is a blind zone. The solution is to excite the creeping wave to detect the surface area to make up for the detection blind zone caused by the primary wave, as shown in Fig. 4.

The creeping wave is also called "longitudinal wave under the surface". When the longitudinal wave is incident at the first critical angle of about 27.6°, the creeping wave can be obtained. Its velocity is about 0.96 times the velocity of the longitudinal wave. The angle of the main lobe is about 80°, which is almost perpendicular to the thickness direction of the part, detection sensitivity for vertical cracks is good. The effective detection depth is 1~9mm, which is rarely affected by the surface of parts such as nicks and unevenness, and is suitable for the detection of rough surfaces. However, after the creeping wave leaves the probe, the attenuation is very fast, and the general inspection length is only twenty some millimeters. Fig. 6 is the inspection diagram for artificial grooves 1~6mm away from the surface, which proves the effectiveness of creeping waves in the detection of near-surface defects.

2.2 Effects of anisotropy and solutions

Austenitic welds are coarse-grained, non-uniform in structure, and have obvious anisotropy, which can lead to twisting and scattering of ultrasonic beams, see Fig. 6. When ultrasonic waves propagate in anisotropic medium, the sound attenuation value and velocity are affected by the angle between the beam direction and the crystal axis. In the range of 0° and 90°, the sound attenuation is the largest and the velocity is the smallest. The propagation direction of ultrasonic energy is not perpendicular to the wavefront, which causes the twisting of ultrasonic beams. The maximum twist angle is 15~20°. It can be seen that the anisotropy of the austenitic weld brings the change of the velocity and the attenuation of the acoustic energy. In order to reduce its influence, the scanning angle is selected in the range of 45°. In addition, due to the change of the velocity of the weld, it is necessary to try to avoid using the echo passing through the center of the weld for positioning, so as to avoid larger errors.

2.3 Impacts and solutions of heterogeneous interfaces

The weld fusion surface of austenitic stainless steel is significantly different from the basic structure. When the ultrasonic wave is incident on the fusion surface, reflection, refraction and waveform conversion will occur, resulting in false signals, as shown in Fig. 7.

In addition, the superposition and accumulation of the interface reflection echoes between the grains in the weld will also produce false signals, and the wider the beams, the more obvious this phenomenon is; otherwise, the narrower the beams, the weaker the impact. Therefore, the use of dual linear array (DLA) or dual matrix array (DMA) probes and the use of acoustic beam focusing technology shall obtain focused narrow beams as a better solution.

The beam focusing of the phased array includes acoustic focusing and electronic focusing. Acoustic focusing is the focusing generated by the "pseudo-focusing effect" of the geometric beam diamond overlapping area of transmit-and-receive respectively, and the electronic focusing electronically controls the beam delay time, resulting in phase superposition for focusing.

Wherein the electronic focusing of the DLA probe can only be carried out in one direction, while for the other direction, acoustic focusing is produced by the wedge with roof angle, with the focal area of the acoustic focusing being fixed. For objects with large thickness that need to be detected in multiple focal areas, it is necessary to replace the wedge to achieve this purpose. The DMA probes can realize electronic focusing in two directions. By controlling the excitation delay of the array elements in two directions, the beams can be steered and focused in 3D space. The beams can be steered laterally to form a diamond-shaped focusing, which can be completed without the wedge with a roof angle, and the steering is adjustable and controllable. The focusing area can be adjusted to the required depth at any time without replacing the wedge, see Fig. 8.

Fig. 9 shows the use of DLA and DMA phased array probes with a frequency of 4MHz respectively to detect the φ2 side-drilled hole passing through the weld in the austenitic weld comparison test. When the DLA phased array probe penetrates the interface between the base metal and the weld, there are obvious interface clutter reflection signals, while the interface reflection signals of the DMA phased array probe is not very obvious, indicating that the focusing effect of the DMA probe is better, obtaining narrower beams and reducing the emission signals of the heterogeneous interface.

3.  Application Examples

3.1 Test object

The test object is a simulated specimen of austenitic stainless steel with a thickness of 40 mm. The locations and parameters of the simulated defects are shown in Table 1.

3.2 Testing equipment

The testing equipment is shown in Fig. 10. The main equipment and models are as follows:

Phased array detector: SyncScan 32PT       

Phased array linear probe: 2.0DL16-1.0-6

Wedge: D16N55L-I                         

Crawler: TSE crawler for Austenitic stainless steel

3.3 Test results

The test results are consistent with the location and parameters of the defects on the simulated test block, as shown in Fig. 11.