|NDT.net - July 2002, Vol. 7 No.07|
This paper introduces and demonstrates the new Near Side Detection and Sizing (NSDS) ultrasonic transducers, from Eclipse Scientific Products Inc., that were designed to obtain accurate same side sizing measurements of isolated or colony surface breaking defects found in plate and pipe. The defects sized and discussed in this paper were isolated and colony Stress Corrosion Cracks (SCC) located in the body of gas transmission line pipe. This is paper is one in a series of articles that discusses same side sizing techniques to obtain point measurements or profiles of SCC to assist with In-Line Inspection (ILI) verification and Engineering Critical Assessment (ECA).
Ultrasonic, Crack, Stress Corrosion Cracking, Near Side Sizing, SCC
SCC and other forms of cracking pose a serious concern for operators of pipelines due to the safety, environmental, and economic impacts that result from the failure of a pipeline. Crack detection and sizing is of great importance to ensure pipeline failures are avoided. Once cracks are detected, accurate same side sizing depth measurements of the cracks is essential to evaluating the remaining strength of the pipe. Accurate same side sizing results also complements the current interest within the pipeline industry to obtain more information about defects from In-Line Inspection (ILI) tools.
Same side sizing of surface breaking defects requires special consideration and novel approaches. Crack sizing is a difficult task to perform and is usually only trusted to personnel with years of experience. There are several methods available that are commonly used for crack sizing; conventional and focused angle beam, creeping wave, tip diffraction, and phased array (ASNT 1989; Krautkramer et al., 1990).
Time of flight tip diffraction methods obtain same side sizing results by transmitting a longitudinal wave which activates the crack tip that generates ultrasonic signals traveling in all directions away from the crack tip (Harumi et al., 1989). A receiving transducer collects the signals that travel back to the surface. Time of flight tip diffraction methods have effectively been used in same side sizing applications (Ginzel et al., 1987; Ginzel 1997). In many situations this technology can be used with a standard hand held ultrasonic instrument helping to minimize costs.
The NSDS transducers use time of flight tip diffraction methods to obtain sizing information. NSDS transducers have successfully been used in a variety of applications to size defects, including lap and mechanical damage adjacent to ERW welds, isolated SCC, and colony SCC. The NSDS sizing examples provided in this paper include the sizing of SCC colony and isolated indications.
The NSDS H-series transducer design utilizes two piezoelectric elements in a tandem configuration that enable a time of flight tip diffraction measurement to be obtained from an indication. Figure 1 illustrates the path traveled by the sound beam when it is used to same side size crack like defects. To create this beam path the transmitting element (Tx) is positioned at the rear of the transducer and set at an angle such that a near 70o longitudinal wave is created. The receiving element (Rx), located at the front of the transducer, is positioned to receive signals at shallow angles <15o. The design allows the signal to be concentrated on the tip of indications resulting in a strong, clear response.
As illustrated in Figure 1, the transducer must be positioned above the defect to allow recovery of the signal. The exit point of the returning signal, labeled as "x" in Figure 1, is located approximately 9mm back from the front of the transducer. The exact position of the exit point will vary depending on the defect depth being sized.
|Fig 1: Conceptual drawing of the NSDS H-series transducer showing the path of the sound beam (longitudinal wave only) when crack defects are sized.|
The NSDS H-series transducer is designed for same side sizing of surface breaking defects in carbon steel utilizing piezoelectric elements that provide the optimum response. To provide the optimum inspection conditions for the material being inspected and the defects being sized, four NSDS H-series transducers are available. Wedge design, piezoelectric element selection, and element separation are all configured to provide clean, crisp signals with a good signal to noise ratio. The NSDS H-series transducers are capable of sizing defects as shallow as 0.5mm in material as thin as 4mm or defects up to 15mm deep in much thicker material. Defect sizing at even greater depths is also possible with alternative transducer designs.
The NSDS H-series transducers are designed to work with almost any UT instrument and can be used in a manual or automated inspection format. The NSDS H-series transducer is 38 by 26 by 40mm (1.5 by 1 by 1.6") and includes 4-40 mounting holes located in the stainless steel housing along with couplant lines for automated inspection formats. When sizing with the NSDS transducer in a manual format an RF waveform presentation or phase specific presentation is recommended. For automated inspection formats the collection of digitized waveform data is recommended with the option of monitoring the time of flight data and amplitude data.
|Fig 2: A B-scan of three SDHs that are used for calibration purposes. The A-scan at the top was extracted from the B-scan at approximately 160mm.|
Calibration of the NSDS transducer is required to ensure that the results from the sizing exercise reflect the inspection conditions. The calibration process involves obtaining time of flight information from a series of known targets at three or more depths. The information from the known targets is used to convert the time of flight measurements from the defects being sized into depth measurements.
Side Drilled Holes (SDHs) are typically used as the known targets from which the time of flight measurements are obtained during calibration. Figure 2 contains a B-scan and an extracted A-scan of a typical set of SDHs used for calibration. Although a B-scan is shown here, the calibration can be completed using point measurements established from the A-scans. An important consideration when calibrating the NSDS transducer is that the calibration block should have the same velocity properties as the material being inspected. In addition to material properties, the temperature of the calibration block should be comparable to ensure an accurate calibration.
The SDHs used in the calibration should have similar depth properties as the defects being sized since there is a non-linear relationship between the depth and the time of flight measurements. Using calibration targets of comparable depth will provide the basis for accurate sizing results.
Once the calibration is complete, defect depths can be obtained by utilizing a time to distance conversion chart, a direct depth reading from the UT instrument, or screen graduations. The method selected will depend on the operator’s preference.
The NSDS transducer can be used to size a variety of defects but was designed for same side sizing of surface breaking defects in carbon steel that are located in areas free of surface obstructions. This limitation, that defects must be in areas free of surface obstructions, is due to the transducer having to be positioned above the crack in order to complete the sizing exercise. Typically the defect should be at least 12.5mm (½") from any surface obstructions.
Prior to sizing the defects it is recommended that fluorescent MPI or wet black and white contrast MPI be used to initially identify and view the surface breaking indication pattern. For our presentation the use of white contrast paint and magnetic powder provided good contrast in photographs to illustrate the selected crack patterns.
The following sizing examples illustrate sizing
|Fig 3: Photograph of the magnetic particle inspection of an isolated SCC crack.|
The crack being sized was 16mm in length and was located in a pipe coupon obtained from a 36" diameter carbon steel pipe. Figure 3 shows the photograph of the magnetic particle inspection results.
Sizing of the isolated crack was completed by mounting the NSDS transducer in a laboratory based automated inspection system and using data acquisition software to acquire full waveform ultrasonic data. The data was then reviewed to obtain crack depth information. Manual methods could have been used for the inspection of the isolated indication but for this sizing exercise automated inspection methods were employed.
Figure 4 contains the C-scan image obtained from inspecting the crack shown in Figure3. The C-scan that is presented is showing response signal amplitude data only. The data in Figure 4 illustrates how the signal is strongest (represented by red) when the tip of the crack is excited by the center of the beam and the signal weakens (represented by blue) when only off-axis portions of the ultrasonic beam are exciting the tip of the crack.
The data in Figure 4 also reveals that the crack shown in Figure 3 is in fact two cracks that have linked to form one crack. This observation was confirmed by verification tests that used high frequency immersion transducers in fine resolution scans from the opposite wall. The detailed opposite wall scans were carried out at 50MHz and revealed significant detail pertaining to the crack facets. Normal beam scans and slightly off normal angle scans from the opposite wall were used to image the crack and the results revealed information about the cracks as well as details regarding the plate material, such as very small inclusions located at varied mid-wall positions and changes in material properties in proximity to the crack region.
|Fig 4: C-scan showing amplitude data for an isolated SCC crack.|
Figure 5 contains two B-scans (located at position 190mm and 195mm) that were extracted from the dataset shown in Figure 4. At the top of each B-scan there is an extracted A-scan showing the peak and collected signal that would be used for sizing the defect based on the time of flight information. Both A-scans have very clear, strong signals from the tip of the crack enabling consistent interpretation.
|Fig 5: B-scans extracted from the C-scan shown in Figure 4. The B-scan on the left is located at 190mm and the B-scan on the right is located at 195mm. At the top of each B-scan are the peaked and collected signals that were used for sizing.|
The depths obtained for the crack at the two locations (at 190mm and 195mm) compared very well with the results obtained through the validation work. Table 1 contains the results of sizing the crack at the two locations indicated using the NSDS transducer as well as the depth measurements obtained using the validation technique.
|Position||Measured Depth||Validated Depth|
Time of Flight |
|Table 1: Results of sizing the isolated crack shown in Figure 4. The measured values were obtained using the NSDS transducer and the validation values were obtained using high frequency normal beam scans from the opposite wall.|
As seen in Table 1 the measured values compare well with the depths obtained through validation methods. The results demonstrate that the difference between the measured and validated depth is 0.12mm and 0.05mm (0.004" and 0.001").
The NSDS transducers can be used to size SCC cracks that are located within colonies. Although this type of inspection can be completed by manual inspection methods it is recommended that a semi-automated or automated method be used. Figure 6 contains the magnetic particle inspection results for the SCC colony used in this example.
|Fig 6: Photograph of magnetic particle inspection results for an SCC colony. The area scanned was 315mm by 60mm.|
Similar to the inspection of the isolated SCC crack, full waveform data was acquired by mounting the NSDS transducer in a laboratory based automated inspection system. Figure 7 contains the C-scan image obtained from the crack colony shown in Figure 6.
|Fig 7: C-scan image of the SCC colony shown in Figure 6. The displayed data is gated amplitude data of the full waveform dataset. This view shows the overall pattern of the SCC colony and allows the operator to quickly evaluate the quality of the scan data.|
|Fig 8: Composite of the MPI pattern and Ultrasonic image.|
The C-scan shown in Figure 7 is amplitude data with maximum interpolated values shown. This presentation shows only amplitudes of signals within a gated region of the waveform dataset. The gate size is adjusted to envelop the region of interest at various depths. For example, the gate setting can be set to eliminate very shallow indications or less significant cracks as it was in Figure 7.
Overlaying the C-scan data on the MPI image results in a good comparison between the UT image and the MPI pattern. Figure 8 presents the MPI image overlaid on the UT data.
With the full waveform data collected many presentation and processing options can be performed. For example, the pattern shown in Figure 9 is similar to the pattern shown in Figure 7 with the exception that the data displayed is time of flight data instead of amplitude data. As in Figure 7, the gate in Figure 9 was set to highlight the more significant cracks (depths greater than 1mm).
|Fig 9: Time of flight C-scan of the SCC colony shown in Figure 6.|
Table 2 shows the results of sizing two cracks from the image shown in Figure 6 at axial locations 133 and 293mm. Sizing was completed in a similar manor as presented for the isolated crack (using B-scans and A-scans) and the crack depths were validated using the opposite wall, high frequency delay line normal beam.
|Position||Measured Depth||Validated Depth|
Time of Flight |
|Table 2: Results of sizing two cracks from within the SCC colony shown in Figure 6.|
The results in Table 2 show that the measured values compared well with the depths obtained through validation methods. The difference between the measured and validated depth is 0.02mm and 0.15mm (0.001" and 0.006").
In this example two co-parallel SCC cracks are sized using the NSDS transducer. Figure 10 shows a photograph of the MPI pattern of the SCC colony containing the co-parallel cracks. The two co-parallel cracks were located within 2.3mm of each other.
|Fig 10: Photograph of MPI results for an SCC colony. The boxed region contains two co-parallel cracks that were sized. The red arrow indicates the location of the B-scan shown in Figure 11.|
Figure 11 contains the B-scan from the scan path indicated in Figure 10. As seen the two co-parallel cracks show up as distinct signal envelopes. The A-scans extracted from this B-scan contain components from both cracks; however, due to the quality of the signal, each crack can be sized accurately. The most efficient method to complete the sizing is to scroll through the various A-scans within the B-scan to locate the peak and collected signal for each crack.
|Fig 11: B-scan obtained from the two co-parallel cracks indicated in Figure 10. The peaked signal used to size each crack is clearly shown along with the weaker signal component from the adjacent crack|
It has been demonstrated that the NSDS transducers can be used to accurately size isolated and colony SCC, including co-parallel cracks located within a few millimeters of one another. Using accurate time measurement techniques produces reliable and repeatable results. The signals obtained from the NSDS transducer are strong with a good signal to noise ratio that enables operators to obtain repeatable results and report sizing measurements with confidence.
In this paper the results of sizing the isolated and colony SCC have been presented through manual interpretation. Although this method is effective the time required for interpretation may be significant. Application of automated processing methods will allow efficient manipulation of the full waveform data. Utilizing the NSDS transducers to acquire full waveform data, and post processing software to obtain crack profiles as well as interlinked geometries, full volumetric imaging of isolated and colony SCC for ECA can be completed.
The results obtained from the NSDS transducers provide the ability to complete quantitative comparisons between the same side sizing data collected during verification programs and the results reported by ILI programs. The same side sizing information obtained from the NSDS transducers may be used to validate ILI data interpretation to provide operators with confidence in ILI sizing results. This has the potential to reduce the number of verification excavations required and thus a cost savings to pipeline operations.
Eclipse Scientific Products Inc. specializes in the development of leading edge tools for the Non-Destructive Testing (NDT) industry, primarily in the ultrasonic (UT) discipline. The focus of the company is to enhance the accuracy of inspection results while making NDT inspection methods more efficient. Eclipse provides ultrasonic transducers (standard and special application), scanning systems, calibration blocks, and NDT/NDE software for the power generation (nuclear, hydro, fossil), the petrochemical (refinery, transmission, storage), rail, medical, and aerospace industries.
|© NDT.net - firstname.lastname@example.org|||Top||