Volume 4, Number 1
January 2005
Articles one through five of this series discussed basic equipment setup and calibration techniques commonly used for contact angle beam testing. Remaining articles in the series are focused on what is required once the UT operator is given a specific part to inspect. In this article, full penetration plate butt welds and corner and T joint welds will be used as examples. Ultrasonic testing of nonwelded parts (castings, forgings and so on) is done in a similar manner, though the location reference points vary depending on the configuration of the part being inspected.Determining Accessibility
The first step in setting up onsite for an inspection is to determine accessibility to the area of interest for the weld to be inspected. In some cases, the governing code or specification might also require ultrasonic inspection of the base material for a short distance beyond the toe of the weld. This requirement should be taken into consideration when selecting the wedge angle and determining the length of the scanning area.
Common Weld Configurations
Figure 1 shows three common weld configurations; butt welds in a plate, a corner joint made by two plate ends set at 90 degree angles to each other and a T joint between two plates (typically seen in beam-column connections). The surfaces adjacent to the weld (or in the case of corner and T welds, the face directly behind) are the surfaces from which the weld can be ultrasonically scanned. As shown in all three figures, these surfaces are commonly labeled Faces A, B and C as applicable.
The two weld sketches in Fig. 1a show plates with two weld configurations, a single-V butt weld at the top and a double-V butt weld at the bottom. While the scanning surfaces for both are labeled the same, the location of the root of the weld is different and will affect the location of certain types of discontinuities. Face A is on the plate side with the weld crown for a single-V weld with Face B on the root side. In the case of a double-V weld, Face A is usually the side of the weld most easily accessible.
Figure 1b shows a typical corner weld with the Face A scanning surface on the weld crown side of the beveled plate, Face B on the opposite side of the same plate and Face C directly opposite the beveled plate on the reverse side of the other plate. These face designations are the same for a typical T weld as shown in Fig. 1c. Scanning from
Face C is done with a straight beam transducer and is used primarily to detect laminar tears in the vertical member behind the weld or sidewall lack of fusion at the vertical face of the weld groove as shown. In all inspections, the face from which the weld is being inspected should be recorded on the inspection report form.
Determining the Scan Surface
Once the scanning face has been selected, it is necessary to determine how much of that surface is required to perform a full coverage inspection of the area of interest. This is determined by material thickness and choice of the wedge angle if that choice is left up to the operator and is not specified by code or specification. In order to attain full volumetric coverage, it is necessary to ensure that the weld area is scanned by both the first and second legs of the sound beam. If an additional portion of the base metal requires inspection, that distance must be added to the scanning area also.
Minimum Scanning Surface. Figure 2 shows the minimum scanning surface required on each side of a plate butt weld to provide full coverage of the weld volume. In the figure, Face A is the scanning surface and the weld is being scanned from both sides. Notice that at position A1, where the nose of the transducer hits the edge of the weld crown, the sound beam only covers a small portion of the bottom of the weld. At position B1, from the other side of the weld, only an additional small volume that was not covered by A1 is scanned. However, as the transducer is moved away from the weld, toward positions A2 and B2 respectively, the sound beam will interrogate the entire cross-section of the weld, resulting in full coverage of the weld volume. If the governing documents require that a portion of the base metal also be interrogated, then that distance must also be included when the scanning surface is prepared. Operators should keep in mind that a 70-degree probe would require a much longer scan path than a 45-degree probe. However, a 45-degree sound beam may be too short to get the second leg completely through the weld cross-section before the nose of the probe hits the edge of the weld crown.
Surface Preparation. The portion of the plate to be used as the scanning surface must be free of any loose scale, rust or other foreign material that might cause bad coupling and thereby prevent sound from entering the part. If weld spatter is present, it may be necessary to use a cold chisel to remove it. A 3 in. wide chisel is wide enough to clean a large area quickly and is a good tool for this purpose. If used properly, it will not gouge the plate surface. Once the scanning surfaces have been prepared, couplant can be applied and the weld scanned.
Discontinuity Orientation
In order to detect both transverse and axially oriented discontinuities, the weld must be scanned in two directions from both sides of the weld.
Axial Discontinuities. For axial discontinuities (those that run parallel to the weld length) the transducer is aimed at the weld and moved back and forth as shown on side A of the plate surface in Fig. 3.
Transverse Discontinuities. For transversely oriented discontinuities (those that run across the width of the weld) the transducer should slide along the side of the weld in both directions as shown on side B of the surface in Fig. 3. Both types of scan are performed on both sides A and B of the surface to ensure that all orientations of discontinuities are found.
Manipulating the Probe
Overlap. As the probe is moved back and forth, the sound beam is required to overlap the previous scan by some percentage. The scan pattern shown in Fig. 4 illustrates a scan pattern with a 50 percent overlap. If the amount of overlap is not detailed in governing documents, a 20 to 50 percent overlap is usually used. The easiest way to estimate the percentage of overlap is to look at the tracks the transducer makes in the couplant on the surface of the plate.
Oscillating the Probe. In addition to being moved back and forth, the probe should also be oscillated from side to side to detect discontinuities that may not be oriented at exactly 90 degrees to the sound beam. If no oscillation requirement is specified in the governing code or specification, a range of approximately 15 to 20 degrees can be used (see Fig. 4).
Difficulties of Probe Manipulation. At first glance, these manipulations may appear relatively straightforward and not a complex task. One might assume that all that is required is to spread the couplant and then slide the transducer back and forth far enough to ensure full coverage by the sound beam while wiggling the probe back and forth sideways. However, the combination of movements is more difficult than first perceived and, in addition, probe manipulation must be done while the operator is watching the CRT screen and not the probe. The process is further complicated by the fact that operators routinely use probes that vary from 0.25 in. to 1.0 in. or greater in diameter. All of these factors make it difficult to maintain a good scanning pattern with consistent overlap. It’s helpful to remember that proficiency improves with practice.
Scanning Rate. Most codes or standards require that the scanning rate not exceed a maximum travel speed, typically not more than 6.0 in. per second. The actual scan speed will either be detailed in the governing specification or in one of the specification reference documents. This restricted scan rate ensures that any reflected sound has sufficient time to return to the transducer before the transducer moves on.
Reporting Discontinuity Locations
Using an XY Coordinate System. The final step before beginning an inspection is to select a set of reference points so that any discontinuities can be accurately located. This is most commonly done with an XY coordinate system, where the X-axis represents the distance across the weld and the Y-axis represents the distance along the length of the weld. Figure 5 shows typical locations for these points on different types of welds.
Locating the Zero Point. As shown in Fig. 5a, the zero point for X on a plate weld is the centerline of the weld and the zero point for Y is the left end of the weld. For corner and T welds, the zero point for X is at the back of the weld (Fig. 5b). For horizontal runs in pipe welds, it is common to place the Y reference point (Y = 0) at the top of the pipe (Fig. 5c) and in line with an elbow or fitting on vertical runs. The point at which X is equal to zero is again the centerline of the weld. Note that some codes and specifications do have specific conventions for these locations so they should be checked prior to setting your own locating marks. If no specific location instructions are given, the operator should select one set of conventions and record them on the report form. It is also smart to mark the point at which Y equals zero on the weld. Then, if repairs are needed, the welder will know the point to measure from.
The distance of a discontinuity from Y = 0 is fairly obvious; the distance is measured from the left end of the weld to the nearest end of a discontinuity. Therefore, a discontinuity that starts at 6.0 in. from the left end of the weld and ends 8.0 in. from the left end would be listed as a 2.0 in. indication at Y = 6.0 in.
Locating a defect on the X-axis requires a little more thought. Centerline indications are easy because they are at X = 0, as shown by indication 1 on the shaded weld in Fig. 6. Indication 2 is 0.5 in. from the centerline away from the inspector and would be recorded as being at X = +0.5 in. Because indications 3 and 4 are both on the operator side of X = zero, they would be recorded at X = –0.5 in. and –0.25 in., respectively. There is one note of caution when scanning the weld from the other side; the operator should make sure that the ±X convention is not accidentally reversed.
Sizing the Discontinuity
Six Decibel Drop. When sizing a discontinuity for length, the operator should refer to the governing documents for direction. However, if a specific method of determining the end point is not given, it is common practice to use the six decibel drop method for making that determination. If used, note of its use should be recorded. The six decibel drop method refers to the fact that a 6 decibel (dB) decrease in gain results in a 50 percent decrease in screen amplitude. The operator manipulates the transducer to maximize the signal from an indication. When this is achieved, the maximized signal is then set to 80 percent full screen height (FSH) and the transducer is moved slowly along parallel to the axis of the defect until the screen signal drops to 40 percent FSH, or 50 percent of the maximized signal (a 6 dB drop). A mark is then made at the centerline of the transducer to denote that end of the discontinuity. The transducer is then moved back toward the other end of the discontinuity until the signal peaks at 80 percent FSH and again drops to 40 percent at the other end of the indication. This end is marked as before using the centerline of the transducer. The distance between the two marks is considered the length of the discontinuity.
Irregularly Shaped Discontinuities. Note that since many discontinuities are irregular in shape, the operator should continue the sizing scan past the point at which a 50 percent amplitude drop is first seen. It is possible that a discontinuity will have a varying orientation that will drop off in amplitude at one point then increase in amplitude again farther away from the center of the discontinuity. If this condition occurs and the scan is stopped when the signal first drops to 50 percent, the recorded length will not reflect the actual discontinuity length. If the acceptance-rejection criteria being used is based on a combination of length and decibel rating, miscalculation of the length could have serious results.
Discontinuity Depth
The depth of a discontinuity can be calculated using a trigonometric function based on the sound path or surface distance and the inspection angle. This can be read directly from a graphic UT calculator. On newer machines, depth calculation is an integral part of the programming and can be done by simply pushing a button. Regardless of how depth is calculated, the operator should bear in mind that depth calculation is based on the nominal wedge angle (45, 60 or 70 degrees). In addition, most codes and specifications allow the transducer to vary within a range of ±2 degrees. As a result, a three decimal place depth number may not be as accurate as the operator might think. It should also be noted that the welder that must dig out the defect identified by the UT operator will not be able to remove the weld in 0.001 in. increments when using a 0.375 in. diameter carbon air arc rod or 36 grit grinding wheel.
It has been the author's habit to report the defect depth to the next lower full 0.0625 (1/16) of an inch. Reporting a defect in this manner is done for two reasons. Repeated handling of the transducer results in more wear at the nose of the wedge face, thus causing the angle to decrease over time. In addition, digging a little deeper the first time ensures removal of the defect and is quicker and less expensive than if a second attempt is required. TNT
*Jim Houf is Senior Manager of ASNT’s Technical Services Department and administers all ASNT certification programs. (800) 222-2768 X212, (614) 274-6899 fax, <jhouf@asnt.org>.
[ The NDT Technician ]
Copyright © 2008 by the American Society for Nondestructive Testing, Inc. ASNT is not responsible for the authenticity or accuracy of information herein. Published opinions and statements do not necessarily reflect the opinion of ASNT. Products or services that are advertised or mentioned do not carry the endorsement or recommendation of ASNT.
IRRSP, NDT Handbook, The NDT Technician and www.asnt.org are trademarks of the American Society for Nondestructive Testing, Inc. ACCP, ASNT, Level III Study Guide, Materials Evaluation, Nondestructive Testing Handbook, Research in Nondestructive Evaluation and RNDE are registered trademarks of the American Society for Nondestructive Testing, Inc.
ASNT exists to create a safer world by promoting the profession and technologies of nondestructive testing.
