Back to Basics

Portable Phased Array Applications

by Jesse Granillo* and Michel Moles⁺

 

INTRODUCTION
Volumetric nondestructive testing (NDT) is typically performed in industry using either radiography or ultrasonics. Radiography has the disadvantages that it can be a safety hazard and is poor at detecting the more critical planar discontinuities (cracks, lack of fusion and lack of penetration). Manual ultrasonics is much better at detecting planar discontinuities, but it is slow and the results are highly dependent on the operator. Automated ultrasonic testing typically involves large, expensive and inflexible systems, though the results are reproducible. A new development - portable ultrasonic phased arrays - offers speed and flexibility.

Portable phased array ultrasonic equipment is highly computerized and can be operated in manual, semiautomated (encoded, with or without a scanning aid) or fully automated (operating a scanning rig) modes. This new generation of equipment offers many of the advantages of phased arrays: speed, flexibility, data storage, imaging, reproducibility and limited footprint, with many of the advantages of manual ultrasonics: portability, ease of setup and relatively low cost.

After briefly introducing the principles of phased arrays and the types of scans, this paper describes a series of portable phased array applications. As normal with new categories of equipment, many of the initial applications have been unusual in some way; more recently, general applications for weld testing have become viable. Perhaps more interesting is the observation that most of the applications are either fully manual or semiautomated. Very few portable phased array applications are fully automated.

 

ULTRASONIC PHASED ARRAYS
Ultrasonic phased arrays are a novel technique for generating and receiving ultrasound. Instead of a single transducer and beam, phased arrays use multiple ultrasonic elements and electronic time delays to create beams by constructive and destructive interference. As such, phased arrays offer significant technical advantages for weld testing over conventional ultrasonics. The phased array beams can be steered, scanned, swept and focused electronically. Beam steering permits the selected beam angles to be optimized ultrasonically by orienting them perpendicular to the predicted discontinuities, for example lack of fusion in automated welds.

---

Portable phased arrays are commercially and technically viable for a wide range of applications.

---

Electronic scanning permits very rapid coverage of the components, typically an order of magnitude faster than a single transducer mechanical system. Beam steering (usually called sectorial or azimuthal scanning) can be used for mapping components at appropriate angles to optimize the probability of detection of discontinuities. Sectorial scanning is also useful when only a minimal footprint is possible. Electronic focusing permits optimizing the beam shape and size at the expected discontinuity location, as well as optimizing the probability of detection. Overall, the use of phased arrays permits optimizing discontinuity detection while minimizing testing time.

 

How Phased Arrays Work
Ultrasonic phased arrays are similar in principle to phased array radar, sonar and other wave physics applications. However, ultrasonic development is behind the other applications due to a smaller market, shorter wavelengths, mode conversions and more complex components. Several authors have reviewed applications of ultrasonic phased arrays (Clay et al., 1999; Wustenberg et al., 1999; Lafontaine and Cancre, 2000), though industrial uses have been limited until the last few years.

From a practical viewpoint, ultrasonic phased arrays are merely a technique for generating and receiving ultrasound; once the ultrasound is in the material, it is independent of the generating technique (piezoelectric, electromagnetic, laser or phased arrays). Consequently, many of the details of ultrasonic testing remain unchanged; for example, if 5 MHz is the optimum testing frequency with conventional ultrasonics, then phased arrays would typically use the same frequency, aperture size, focal length and incident angle.

Phased arrays use an array of elements, all individually wired, pulsed and time shifted. These elements are usually pulsed in groups from 4 to 16 elements. A typical user friendly computerized setup calculates the time delays from operator input, or uses a predefined file: test angle, focal distance, scan pattern and so forth (see Figures in R/D Tech, 2004).

The time delay values are back calculated using time of flight from the focal spot and the scan assembled from individual focal laws. Time delay circuits must be accurate to around 2 ns to provide the accuracy required.

The setup information is electronically recorded and only takes seconds to reload. Modifying a prepared setup is quick in comparison with physically adjusting conventional transducers.

 

Types of Scans
Using electronic pulsing and receiving provides significant opportunities for a variety of scan patterns. The two basic patterns are electronic and sectorial scans.

Electronic scans are performed by multiplexing along an array. Typical arrays have up to 128 elements, pulsed in groups of 8 to 16. Electronic and linear (single axis mechanical scanning) testing permits rapid coverage with a tight focal spot. If the array is flat and linear, then the scan pattern is a simple B-scan. The data can be processed to provide a C-scan or combined scans (for example, top/side/end views or combined S- and A-scans).

Sectorial scans use the same set of elements, but alter the time delays to sweep the beam through a series of angles. Again, this is a straightforward scan to program. Applications for sectorial scanning typically involve a stationary array, sweeping across a relatively inaccessible component like a turbine blade root (Ciorau et al., 2000), to map out the features (and discontinuities). Depending primarily on the array frequency and element spacing, the sweep angles can vary from ±20 to ±80 degrees.

Manual ultrasonic testing is performed using a single transducer, which the operator scans back and forth to cover the area to be tested. Many automated testing systems use a similar approach, with a single transducer scanned back and forth for corrosion or weld testing. This is very time consuming, since the system has dead zones at the start and finish of the raster.

In contrast, phased arrays use a linear scanning approach. Here, the probe is mechanically scanned in a line around or along the component (a weld in this example), while the array performs electronic or sectorial scanning. Linear scanning is frequently used in pipe mills and on pipelines.

 

PORTABLE PHASED ARRAY INSTRUMENT
A portable phased array unit with manual, semiautomated and automated capability has been developed. In practice, this is a multiple technology unit, with replaceable function modules (besides phased arrays, there are conventional ultrasonics, time of flight diffraction, eddy current and eddy current array modules available, with other technologies in development). The current phased array unit is a 16/128 unit (16 multiplexed pulsers with 128 channels), with up to 256 focal laws (individual beam pulses). The unit can perform electronic and sectorial scans. It has ultrasonic specifications similar to an upscale single channel discontinuity detector (frequency, filtering, time corrected gain, gates, alarms, range and so forth) and can operate as such. The instrument is fully digital and can perform encoded scans.

The phased array unit records full waveform data at multiple angles/positions and can display A-, B-, C-, D-, S- and combined scans. This gives much increased imaging capability. The unit also has built in reporting capability (using pasted in scans) and internal procedure capability. There is a special calibration process for phased arrays, to ensure uniform signal strength across the array (and wedge). The 4.6 kg (10 lb) unit also has a "probe recognition" function, where the array is automatically detected and characterized when connected; this eliminates programming the array parameters.

 

Arrays
As with all testing systems, the probe or transducer is critically important. This is perhaps even more the case with arrays, though typically a single array can perform multiple tests, often with appropriate wedges. There are technical limits to arrays; individual element sizes are limited in practice to around 0.15 mm (6 x 10^-3 in.) and are normally under 20 MHz. However, the real limitations of arrays are cost. The more advanced arrays, with hundreds of elements, can easily cost tens of thousands of dollars. These arrays can be matrix, circular, conical or complex. To reduce costs, automated manufacturing of a standard series of linear arrays has been developed.

 

APPLICATIONS
This section lists a dozen portable phased array unit applications. This list is far from exhaustive and new applications are arriving regularly. However, this provides a cross section of typical uses and covers a wide variety of industries: nuclear, petrochemical, defense, manufacturing and aerospace.

 

Detection and Sizing of Stress
Corrosion Cracking in Turbine Roots
This application involves a large number of components and high downtime costs, plus limited access in a nuclear reactor. False calls must be minimized, due to outage costs, and small discontinuities (1 mm [0.04 in.] high and as little as 3 mm [0.12 in.] long) must be detected. Discontinuity range and location varies.

The phased array solution was to model the application to optimize array design using ray tracing to optimize the testing. The solution was to use a relatively high frequency (6 to 12 MHz) and to plot the scans on a component overlay. (In practice, being a nuclear application, multiple units and multiplexed scans were used; however, this does not alter the application principles). S-scans were used, with minimal probe movement.

 

Small Diameter Austenitic Pipe Weld Testing
This application involved testing stainless steel pipe welds of variable diameters for a nuclear waste application. The welds were autogenous, made by orbital welders; as such, the weld profile was near vertical. Wall thicknesses were generally thin. Space between pipes was minimal, necessitating a manual scan or low profile scanner. Radiography was not permitted for safety reasons. Rapid and reliable testing was required, with full data recording.

The portable phased array solution used two arrays generating shear waves, one on either side of the weld with a splitter cable. Linear scanning around the weld and a low profile scanner with a small encoder was used for data collection. S-scans were used, with the data displayed as C-scans. Figure 1 shows the scanner and display.

(a)
(b)

Figure 1  - Small diameter austenitic pipe weld testing: (a) twin shear wave wedges with low profile scanner; (b) typical A-, S- and C- scan display showing 1.5 mm (0.06 in.) calibration hole.

 

 

In Service Testing of Pipe for Stress Corrosion Cracking
This nuclear application is for detecting axial stress corrosion cracking in Canada Deuterium Uranium (CANDU) reactor feeder pipes. These pipes are ferritic steel, with very limited access between pipes. Radiation fields are high, so testing must be quick. Crack heights are less 1 mm (0.04 in.) and wall thicknesses are typically around 10 mm (0.4 in.).

The portable phased array solution is to use a small 10 MHz, 16 element array with a miniature wheel encoder attached (Figure 2). Once detected, discontinuities could be accurately sized using time of flight diffraction.

(a)
(b)

Figure 2  - Phased array detection of stress corrosion cracking in feeder pipes: (a) scanning setup; (b) crack detection.

 

 

Butt Weld Testing
In contrast to the nuclear applications above, butt weld testing represents a huge and varied application. Typically, this testing is performed according to an established code and approved procedure and technique. ASME code approval has been obtained using external consultants for pipes and butt welds up to 25 mm (1 in.). Typical testing criteria for practical applications include performing cost effective, rapid and reliable testing of butt welds in plate or tubes, storing the data for reference and imaging discontinuities for optimum sizing.

The portable phased array solution uses an array on a wedge (for wear and optimum angles) to generate shear waves as usual. S-scans or electronic scans are performed using a linear scan along the weld. The data are stored and displayed as S-scans or top/side/end views.

 

T-weld Testing of Bridge Structures
These weld tests are similar to butt weld testing, but can be more challenging due to the geometry. Typically, these applications involve thicknesses of 10 to 16 mm (0.4 to 0.6 in.) and reliable detection of planar discontinuities (cracks, lack of fusion and lack of penetration) is essential. Probe movement is limited, multiple test angles are necessary and a cost effective solution is required.

The portable phased array solution is to use an encoded hand scan with a small, linear, 5 MHz, 16 element array. S-scans are performed at between 40 and 70 degrees using shear waves and the results displayed as a combination of A- and S-scans. Other scanning and display options are possible. Figure 3 shows the T-joint geometry and testing in action.

Figure 3  - Testing T-welds using portable phased arrays with an encoded array: test geometry and procedure.

 

 

Hydrogen Induced Cracking
Hydrogen induced cracking involves the diffusion of hydrogen into steels, where it typically forms lamellar blisters at inclusions. Standard hydrogen induced cracking is benign and easily detected by ultrasonics, but stepwise cracking can occur between blisters, which is structurally undesirable. This stress oriented hydrogen induced cracking (or stepwise cracking) is more difficult to characterize using conventional ultrasonics. While hydrogen induced cracking forms lamellar reflectors parallel to the surface, stress oriented hydrogen induced cracking forms as cracking between hydrogen induced cracking blisters, at an angle to the surface. The objective is to reliably determine if stress oriented hydrogen induced cracking exists amongst regular hydrogen induced cracking. The testing must be rapid and comparatively inexpensive. Data storage is desirable.

The portable phased array solution is to use normal beam electronic manual scans to rapidly detect hydrogen induced cracking. To determine if stress oriented hydrogen induced cracking is present, a second setup file is loaded to perform S-scans using ±30 degree S-scans. A tracking function is used to display the A-scan angle with the highest amplitude waveform. The array is skewed back and forth to optimize the signals. Typically, the beam is focused at midwall, since most hydrogen induced cracking and stress oriented hydrogen induced cracking occurs at 1/3 to 2/3 depth. The operator looks for additional signals between hydrogen induced cracking reflections to identify stress oriented hydrogen induced cracking (Figure 4).

(a)
(b)

Figure 4  - Hydrogen induced cracking: (a) with no stepwise cracking visible (no stress oriented hydrogen induced cracking); (b) with stress oriented hydrogen induced cracking visible.

 

 

Flange Corrosion under Gasket
The requirement is to detect corrosion under a gasket seat, without removing the bolts. Testing is possible only from the pipe surfaces; scanning is needed, but the scanning area is limited. The angles are difficult for conventional ultrasonic testing (Figure 5a).

The portable phased array solution is to use a 16 element phased array probe with a 45 degree natural angle and to perform an S-scan from 30 to 85 degrees. To ensure maximum coverage with the bolts in place, a guide was used. Using a corrected B-scan ensured a good interpretation of the images (Figure 5b).

(a)
(b)

Figure 5  - Gasket corrosion mapping: (a) schematic showing flange gasket, area to be scanned, locations of bolts and the limited access (measurements are in degrees); (b) A-scan, B-scan and corrected B-scan displays of corrosion mapping.

 

 

Nozzle Testing
The requirement is to detect and measure erosion/corrosion on a 175 mm (6.9 in.) nozzle inside surface. The testing must be performed rapidly and in service and must be cost effective.

The portable phased array solution is to use a 32 element, 10 MHz linear array and perform S-scans using L-waves from 0 to 70 degrees (Figures 6a and 6b). The nozzle is imaged as a volume corrected (true depth) S-scan. Erosion/corrosion is measured from the image (Figure 6c). The image can be zoomed, if required.

(a)
(b)
(c)

Figure 6  - Nozzle testing: (a) 175 mm (6.9 in.) calibration block and bevel end; (b) S-scan of nozzle, showing bottom surface, corner and smooth end surface; (c) S-scans showing eroded corner (the right side is a zoomed image).

 

 

Thread Testing
The requirement is to rapidly and reliably test threads on many munitions shafts to determine if they are correctly threaded or double threaded (Figures 7a and 7b). The output display should be easy to interpret. All data must be stored.

The portable phased array solution uses a linear array with a custom wedge to fit the shaft. Focused ultrasonic beams are used for resolution and a B-scan display to show correct or bad threading (Figure 7c). The operator can readily distinguish between good and double threading by interpreting the B-scan patterns (Labbé, 2004).

(a)
(b)
(c)

Figure 7  - Munitions thread testing: (a) munitions tail and mockup of probe on custom wedge; (b) cross section through shaft showing double threading; (c) B-scan of threads showing correct threading.

 

 

Spindle/Shaft Testing
The NDT required in this case involved testing a long spindle for cracking (Figure 8). A rapid and reliable test was required, which should both detect and size any discontinuities. The main concern was that data interpretation could be difficult due to multiple reflections. This type of testing is required for bridge pins, vehicle shafts and similar applications.

The portable phased array solution used a single array rotating on the top of the spindle (Figure 8a), performing a narrow angle S-scan to sweep from the centerline to the edge of spindle. The results were displayed as a corrected S-scan and known features (for example, lands) were used to determine the locations of reflectors. Calibration used machined notches.

(a)
(b)

Figure 8  - Spindle/shaft testing: (a) spindle and true depth (or volume corrected) S-scan display with known reflectors; (b) typical location of cracking in spindle.

 

 

Testing of Bridge Bolts
Bolts hold bridges together and undergo significant fatigue cycles. The bolts are large (around 220 mm [9 in.] long) and fatigue susceptible areas are typically hidden (Figure 9a). Normal ultrasonic testing does not offer the multitude of angles required, nor appropriate data storage and imaging. Testing must be rapid, reproducible and convenient.

The portable phased array solution is to perform a 0 to 15 degree L-wave S scan, focused at 100 mm (4 in.). This is a manual scan (no encoder) with the operator manipulating the array to get full volumetric coverage. The imaging makes interpretation much easier and more reproducible (Figure 9b) and tests were much faster than with conventional ultrasound. It would be possible to include distance amplitude correction or time corrected gain.

(a)

(b)

Figure 9  - Bolt testing: (a) typical bolt with two reference notches and array on accessible area; (b) A-scan and S-scan image from typical bolt, showing threads, reference notch and backwall.

 

 

Landing Gear Testing
Aircraft landing gears undergo considerable stress on landing and takeoff, and are potentially susceptible to fatigue cracking. The area to be tested has three different diameters, which makes conventional ultrasonic testing difficult.

The portable phased array solution is to use an S-scan to generate 40 to 65 degree shear waves inside the component, with a wedge specifically contoured to the cylinder's outer diameter. This permits a single pass test of the cylinder, with full data collection. Though there are several different cylinder outer diameters and multiple diameters within each, electronic setups make this testing straightforward. The imaging permits discontinuity identification.

 

Laser Weld Testing
This is an aerospace test for laser weld construction. The component has a complex geometry, rapid testing is required and full data storage is needed.

The portable phased array solution is to use a linear array with a water box for coupling. A 10 m (32.8 ft) long linear scan manual test is performed, using an encoder at 25 mm/s (1 in./s). The array performs a normal beam raster test (electronic scan), giving a real time C-scan display. All the data are stored.

 

Composites
There are many composite testing applications in the aerospace industry. This particular application is for a 6 mm (0.24 in.) thick carbon composite. A sample simulating layup tape commonly found during the manufacturing process was made with known discontinuities (Figure 10). The objective was to reliably detect and size discontinuities and to store all data.

The portable phased array solution was to use a linear scan with electronic (normal beam) scanning. A 5 MHz, 32 element probe with a 1 mm (0.04 in.) pitch was used. (In practice, a 64 element probe with 0.6 mm [0.02 in.] pitch would give greater resolution.) Contrary to many applications, the element grouping was set at 5. Loss of backwall was used for discontinuity detection. The scans were displayed as C- and A-scans and the data stored as usual.

Figure 10  - Scan results from composite specimen: loss of backwall is visible (arrowed).

 

 

DISCUSSION
The applications listed above show that portable phased arrays can perform many different types of NDT, from generic weld testing to more specialized applications. All these applications have one or more of the following advantages:

• speed: scanning with phased arrays is an order of magnitude faster than single transducer conventional mechanical systems, with better coverage and focusing
• flexibility: setups can be changed in a few minutes and typically a lot more component dimensional flexibility is available
• testing angles: a wide variety of angles and wave modes can be used, depending on the requirements and the array
• imaging: S-, B- and C-scans offer much better data interpretation than simple A-scans
• small footprint: small matrix arrays can give significantly more flexibility for testing restricted areas than conventional transducers.

As mentioned earlier, most of the listed applications are unusually specialized, largely because this is how most new NDT products make it into the marketplace. These special applications will continue, diversifying into applications not currently thought of. Some may even use the fully automated scanning capability.

Most important, portable phased arrays now appear cost competitive for a number of applications. While it is too early to determine the cost of weld testing using portable phased arrays, early evidence shows that such testing is approximately five times faster than with conventional manual testing.

Besides the major labor savings, evidence also suggests that portable phased array weld testing is significantly more reliable than manual testing; the operator's interpretation of a waveform is no longer such a key factor. Once the setup is prepared, the same results are repeatedly obtained. We look forward to the first weld testing trials using portable phased arrays.

The arrival of portable phased arrays may have one other major effect on the NDT industry: significantly increased productivity could offset the upcoming shortage of qualified inspectors.

 

CONCLUSIONS
Portable phased arrays are commercially and technically viable for a wide range of applications. They have major advantages for high speed testing, setup flexibility, multiple test angles and wave modes and limited access testing. They should be cost effective for a number of standard applications (for example, welds) and standard code compliant procedures should significantly increase their application. One should expect more portable phased array applications in the near future.

 

ACKNOWLEDGMENTS
Many people at R/D Tech have assisted in the development of this instrument. In particular, Pierre Langlois spearheaded the development and Chris Magruder, Philippe Cyr, Simon Labbé and others worked on various applications. Also, several external companies have assisted with one or more of the examples here, including Eclipse Scientific Products, OPG, Materials Research Institute, Washington Group International and Northwest Airlines.

 

REFERENCES
Ciorau, P., D. MacGillivray, T. Hazelton, L. Gilham, D. Craig and J. Poguet, "In-situ Examination of ABB 1-0 Blade Roots and Rotor Steeple of Low-pressure Steam Turbine, Using Phased Array Technology," 15th World Conference on NDT, Rome, Italy, October 2000.

Clay, A.C., S.-C. Wooh, L. Azar and J.-Y. Wang, "Experimental Study of Phased Array Beam Characteristics," Journal of NDE, Vol. 18, No. 2, June 1999, p. 59.

Labbé, S., "Signal Analysis for Automated 'Go/Nogo' Inspection of Complex Geometries Using Ultrasonic Phased Arrays," 16th World Conference on NDT, Montréal, Canada, August-September 2004.

Lafontaine, G. and F. Cancre, "Potential of Ultrasonic Phased Arrays for Faster, Better and Cheaper Inspections," NDT.net, Vol. 5, No. 10, October 2000, <www.ndt.net/article/v05n10/lafont2/lafont2.htm>.

R/D Tech, Introduction to Phased Array Ultrasonic Technology Applications: R/D Tech Guideline, Quebec City, Canada, R/D Tech, 2004.

Wüstenberg, H., A. Erhard and G. Shenk, "Some Characteristic Parameters of Ultrasonic Phased Array Probes and Equipments," NDT.net, Vol. 4, No. 4, 1999, www.ndt.net/article/v04n04/wuesten/wuesten.htm.

 

* R/D Tech, 4615 E. Broadway, Suite 2, Long Beach, CA 90803; (562) 439-3102; fax (562) 439-2102; e-mail <jesse.granillo@rd-tech.com>.

⁺  R/D Tech, 73 Superior Avenue, Toronto, ON M8V 2M7, Canada; (416) 831-4428; fax (416) 255-5882; e-mail <michael.moles@rd-tech.com>.

 

 

Copyright © 2005 by the American Society for Nondestructive Testing, Inc. All rights reserved.