Remote Testing of Underground High Level
Radioactive Waste Storage Tanks

by J.B. Elder*

INTRODUCTION
This paper provides a description of the ultrasonic nondestructive tests that were performed on a high level radioactive waste storage tank which has been in service for 46 years at the Department of Energy's Savannah River site, near Aiken, South Carolina. These tests were performed in accordance with WSRC-TR-2002-00061 (Wiersma et al., 2002). The in service testing program for high level waste tanks was developed using BNL-52527 (Bandyopadhyay et al., 1997) for waste tank in service testing programs as a guide.

The tests were performed from the contaminated, 762 mm (30 in.) wide annular space of the inactive, 3.9 ML (1.03 x 10⁶ gal), underground waste storage tank. A steerable, magnetic wheel wall crawler was used to simultaneously collect data with up to four ultrasonic transducers and two cameras.

---

NDT of the tank included remote automated ultrasonic testing supplemented by remote visual testing.

---

The purpose of this test was to verify known corrosion models and to investigate the possibility of previously unidentified corrosion mechanisms. The tests included the testing of previously identified leak sites as well as thickness mapping and crack detection scans on specified areas of the tank covering welds and all past and present interface levels.

Figure 1 - Type II high level waste tank.

Tank Design
Figure 1 shows a schematic of the type of waste tank that was tested. The following is a summary of the tank features:

• construction - 1955 through 1956; entered high level waste storage service in 1960
• capacity - 3.9 ML (1.03 x 10⁶ gal)
• material - ASTM A285, grade B carbon steel (not stress relieved)
• construction code - ASME-52
• 1.5 m (5 ft) steel secondary containment pan - material is A285; grade B carbon steel
• annulus ventilation - normally positive pressure (changed to negative during testing)
• annulus access - constructed with 127 mm (5 in.) carbon steel risers at the south, west, north and east annulus risers. Additional access provided through 152 mm (6 in.) diameter drilled testing ports. There are 12 testing ports plus the four 127 mm (5 in.) risers spaced around the 81.4 m (267 ft) circumference of the tank. The testing ports are identified by the distance in feet from the south riser (Figure 2).

 

Figure 2 - Tank riser layout.

NDT REQUIREMENTS 
The in service testing program for high level waste tanks dictates the frequency and extent of the areas to be tested, as well as the damage mechanisms to be detected. The program states that the specific tank "shall be inspected two times within a five-year time span to validate current degradation models. Known leak sites will be characterized in addition to the normal extent of examination. If leakage occurs in unexpected regions and unknown degradation mechanisms are suspected, additional inspections will be performed" (Wiersma et al., 2002).

The in service testing program for high level waste tanks calls for the following regions of a high level waste tank to be tested:

• liquid vapor interface
• liquid sludge interface
• upper weld of lower knuckle of primary tank (5% of accessible circumference)
• lower knuckle base material
• external surface of primary tank (includes vapor space)
• vertical and horizontal welds other than the lower knuckle weld (one vertical course section and 5% of middle horizontal weld).

These general requirements are further delineated in a tank specific test plan. The tank specific plan stipulated the following tests specific to the primary wall:

• four vertical strips for the entire accessible height of the tank, one each under risers IP55, IP107 and IP182 as well as the east riser (Figure 2)
• 9.1 m (30 ft) of middle horizontal weld between riser IP171 and IP207 (10% of circumference - additional 5% in lieu of 5% of upper weld of lower knuckle which was inaccessible due to tank geometry)
• lower primary shell plate vertical weld below riser IP182
• five previously identified leak sites.

Figure 2 illustrates the approximate radial location of annulus access risers. Locations are in feet from the south riser. The north, south, east and west risers are 127 mm (5 in.) carbon steel pipe. The other testing ports were added using a 152 mm (6 in.) diameter core drill.

 

NDT TECHNIQUES 
NDT of the tank included remote automated ultrasonic testing supplemented by remote visual testing. The following techniques were used to test the tank:

• thickness mapping
• weld testing/crack detection
• ultrasonic discontinuity sizing
• through wall bleed out.

 

Testing Equipment
All ultrasonic tests were performed using an automated ultrasonic system and a remotely operated magnetic wheel scanner. The prescribed regions were tested utilizing two basic data collection techniques:

• vertical strips - base material thickness mapping and crack detection scans
• weld testing - scans of weld and heat affected zones to detect and characterize cracking oriented parallel or perpendicular to the weld seam.

 

Ultrasonic System
The ultrasonic system used for these tests is capable of performing tests with multiple transducers and techniques simultaneously. It is also capable of performing thickness mapping, weld testing and A-scan recording all at the same time. During tank tests, it was used to operate two angle beam and one thickness mapping transducer or four angle beam probes simultaneously. It is operated through a laptop computer as the user interface. The system also controls the wall crawler.

 

Wall Crawler
The wall crawler (Figure 3) is a commercially available crawler, which attaches to the steel tank wall through strong, permanent magnetic wheels. It is capable of being installed through a 127 mm (5 in.) carbon steel riser. It can scan with up to four transducers. The wall crawler is typically outfitted with a remote control pan and tilt camera system with auxiliary lighting.

The wall crawler included a pneumatically activated camera boom arm to lift the pan and tilt camera about 254 mm (10 in.) off the surface. It also has pneumatic lifting feet to decouple it from the tank wall to allow removal from the annulus.

Figure 3 - A wall crawler

 

Procedure and Equipment Qualification
The in service testing program for high level waste tanks states that the ultrasonic system (instrument, transducer, scanning device and cables) shall have the following detection capabilities (tested at 12.7 mm [0.5 in.] nominal thickness of the tank sidewall plate):

• general corrosion/thinning detection greater than 0.5 mm (0.02 in)
• pitting detection (elliptical or hemispherical) greater than 1.3 mm (0.05 in.) depth
• crack detection greater than 2.5 mm (0.1 in.) deep and greater than or equal to 12.7 mm (0.5 in.) long. In the absence of an acceptable cracked sample, a machined notch 1.3 mm (0.05 in.) deep by 25.4 mm (1 in.) long can be used instead of a crack for qualification.

The procedures and equipment easily met the above requirements (Elder, 2002).

 

Thickness Mapping
Thickness mapping includes wall thickness measurement as well as the detection and sizing of corrosion, pitting and liquid/air interface attacks. Thickness mapping was performed in four vertical strips. Individual vertical strips were 216 mm (8.5 in.) wide so the combined width of all four strips provided coverage of 1% of the circumference of the tank. Thickness mapping data were collected over the entire accessible height of the tank to ensure coverage of all areas and environments in the tank. By collecting data in a continuous strip from top to bottom, all present and historic interface levels are examined as well as the vapor space of the tank.

Thickness mapping data were collected using the automated ultrasonic system. A thickness mapping program was utilized to provide color coded thickness plots from the top, side and end views. These data were collected using a dual element, 0 degree, longitudinal wave transducer operating at 5 MHz.

 

Weld Testing and Crack Detection
Weld testing and crack detection were performed with the same ultrasonic system using amplitude based weld testing software. Crack detection was performed using single element, 45 degree shear wave transducers operating at 4 MHz. This technique was incorporated into the thickness mapping vertical strips and was used to examine welds for cracking oriented parallel and perpendicular to the weld seam.

 

Ultrasonic Discontinuity Sizing
When indications were detected with ultrasonic techniques, the extent of the indications were measured. The location and length/width in the X and Y directions were determined based on where the indication was discernable from the background noise or thickness.

Pitting indications were reported based on the remaining sound metal (ligament) above the pit. The depth of any pit indications was determined by subtracting the minimum thickness reading obtained from the pit from the thickness of the area adjacent to the pit.

Cracking lengths were reported to the points where the indication was no longer discernable from the noise. Crack depths were determined utilizing planar discontinuity sizing techniques. Using the same transducers that were used for detection, the amplitude was adjusted to locate the deepest point on the crack. When indications were less than 100% through wall, a measurement of the remaining metal (ligament) was made using the absolute arrival time technique. The absolute arrival time technique is a planar discontinuity sizing technique used throughout industry that provides a direct reading of depth to the crack tip.

 

Through Wall Bleed Out
Through wall bleed out is the term being used to describe the field implemented variation of a liquid penetrant surface test technique. It was noted that the water being used for an ultrasonic couplant would penetrate (through capillary action) surface cracks. Due to the elevated temperature of the tank wall (around 322 K [120 °F]), the wetted surface would dry after a few minutes. Where there was a crack open to the exterior surface, the water drawn into the crack would then bleed out, providing a high contrast image of the open crack. Video cameras were used to view these indications and make crude measurements of length as the crawler was driven along the indications. Figure 4 shows an example of the video image of the bleed out region.

Figure 4 - Example of through wall bleed out.

 

NDT DATA COLLECTION

Field Conditions
Tests were performed from the annular space of the high level waste tank. The wall crawler and cameras were installed in the annulus and operated from the NDT control trailer which was up to 61 m (200 ft) from the riser. Access to the annulus was through testing ports or risers inside contamination control huts. These risers are approximately 1.2 m (4 ft) long and are either 127 mm (5 in.) carbon steel pipe or 152 mm (6 in.) diameter concrete holes. All ultrasonic tests were performed by inserting the wall crawler through the 152 mm (6 in.) concrete risers. Remote pan and tilt cameras were also inserted into the annulus to monitor crawler movement. The tank has a history of through wall leaks, therefore the annulus is contaminated. The tank ventilation was shut down and auxiliary ventilation was installed to provide negative pressure ventilation during the tests. Huts were set up around each riser that was used for crawler access to provide contamination control. In addition to the huts and ventilation, respiratory protection was typically required during crawler installation, removal and maintenance activities.

 

Test Areas
All of the required tests were performed by deploying the crawler through three risers. Ultrasonic tests were performed with the wall crawler in risers IP55, IP107 and IP182.

 

NDT RESULTS
All test data were analyzed by certified Level III personnel.

Summary of Test Results
The tested tank was not stress relieved and had a history of stress corrosion cracking with 18 previously identified leak sites (Waltz and West, 2001). The tank is presently an inactive tank.

Several leak sites were selected to be ultrasonically tested to determine the length, depth and contributing factors (for example, weld attachments, weld beads and so on) of cracks. These indications are scheduled for retesting in five years to look for any changes and to test crack growth. Maximum crack lengths were determined to be longer than previously expected but still well within the established critical crack lengths at the crack locations (Wiersma and Elder, 2003).

Thickness mapping was performed on 1% of the tank circumference for the entire accessible height of the tank (Figure 5). This thickness mapping was performed to detect and measure any general wall loss, pitting or interface attack in all regions of the tank including the vapor space. No reportable wall loss or pitting was detected.

Figure 5 - Average thickness summary plot.

 

Vertical Strip Results
Tests of the tank through riser IP55 included one vertical strip for the entire accessible height of the tank. No reportable areas were detected in the vertical strip. The minimum thicknesses detected in the upper and lower plates are above nominal thickness. The minimum thickness detected in the upper plate was 16.2 mm (0.639 in.) and is near the edge of the plate toward the middle weld. The minimum thickness detected in the lower plate was 16.05 mm (0.632 in.). This minimum thickness area is a 6.4 by 10.2 by 0.8 mm (0.25 by 0.4 by 0.03 in.) deep indication approximately 63.5 mm (2.5 in.) from the middle weld. The minimum thickness at the bottom of the same plate is 16.1 mm (0.634 in.). The indication inside the black circles on Figure 6 was approximately 0.8 mm (0.03 in.) deep. The results also show the plates to be thinner at the edges. There are several noise spikes shown in the side and end views. These noise indications were tested and determined to not be relevant. No cracklike indications were detected in this strip.

Figure 6 - Thickness mapping image of lower plate, riser 55.

 

Previous Indication Investigation
One previously identified indication in the upper plate vertical weld at approximately 16.2 m (53 ft) was also examined under riser IP55. The indication was observed at 5.1 m (200 in.) above the tank bottom. Due to the high weld profile and limited time and the fact that the examination of this leak site was not a requirement of the test scope, the indication was only tested from one side of the vertical weld. The tests were performed on the side of the weld opposite the riser. The crack was confirmed to be through wall, but also had a partial through wall segment. Measuring the indication on the right side only, the through wall portion was 35.6 mm (1.4 in.). The total length was 94 mm (3.7 in.) on that side of the weld.

Figure 7 shows some of the ultrasonic data from this indication. The through wall portion of the crack is represented by the blue portion of the ultrasonic data in Figure 7. The yellow and green portions of the ultrasonic image indicate the part of the crack that is only partially through wall.

Figure 7 - Ultrasonic data for crack in upper plate vertical weld, riser 55.

 

Weld Testing Results
The following is a summary of the indications detected in the lower plate vertical weld. The through wall indication (Figure 8) measured 114 mm (4.5 in.) total length. The photographic overlay is from a liquid penetrant test performed on a similar tank in 1962. The recently detected crack is nearly identical to the crack from 1962.

Figure 8 - Image of perpendicular crack with liquid penetrant results overlay.

 

Approximately 10.1 m (33 ft) of the middle horizontal weld was tested for horizontal and vertical cracking. The middle horizontal weld was examined in three sections, as noted:

• from 52.1 to 54.9 m (171 to 180 ft)
• from 55.8 to 59.4 m (183 to 195 ft)
• from 59.7 to 63.4 m (196 to 208 ft).

Previously identified leak sites on the middle weld were also tested.

 

Leak Site
The leak site at 58.5 m (192 ft) is a horizontal/arched crack in a lower plate at a weld repair location (Figure 9). The through wall portion (verified visually with the bleed out technique) of this crack was measured ultrasonically to be 259 mm (10.2 in.). The indication is arch shaped around a weld repair region in the horizontal weld. The weld repair area appears to be approximately 203 mm (8 in.) long and centered on the through wall portion of the crack. The total length of the indication was measured at 460 mm (18.1 in.). As shown in Figure 10, this indication is longer at the inside surface than on the outside.

Figure 9 - Crack at 58.5 m (192 ft.), middle horizontal weld: (a) as found; (b) after bleed out.

Figure 10 - Ultrasonic image of crack 58.5 m (192 ft), middle horizontal weld.  The through wall part of the crack is shown in the boxed area. The ultrasonic image has been rotated to the same orientation as in the visual image (Figure 9).

 

 

CONCLUSION
A new ultrasonic in service test program for high level waste tanks at the Savannah River site has been implemented. The test details and results from the testing of a specific tank that has been in service for over 40 years have been summarized.

No indications of reportable wall loss or pitting were detected. All thickness readings were above minimum design thickness. Several small indications of thinning were detected. The crack detection and sizing tests detected five previously undetected indications, four of which were only partially through wall. The lengths of cracks that were examined were slightly longer than expected, but well below instability lengths.

 

ACKNOWLEDGMENTS
The authors gratefully acknowledge the outstanding work of the Savannah River site waste tank ultrasonic testing team of E.R. Holland and R.W. VandeKamp. The work was supported by the United States Department of Energy under contract number DE-AC09-96SR18500.

 

REFERENCES
Bandyopadhyay, K., S. Bush, M. Kassir, B. Mather, P. Shewmon, M. Streicher, B. Thompson, D. van Rooyen and J. Weeks, BNL-52527, Guidelines for Development of Structural Integrity Programs for DOE High Level Waste Storage Tanks, Engineering Research and Applications Division, Upton, New York, Brookhaven National Laboratory, January 1997.

Elder, J.B., "Procedure and Equipment Qualification," TSD-NDE-20020726, Aiken, South Carolina, Savannah River National Laboratory, December 2002.

Waltz, R.S. and W.R. West, "Annual Radioactive Waste Tank Inspection Program - 2000," WSRC-TR-2001-00149, Oak Ridge, Tennessee, US Department of Energy, Office of Scientific and Technical Information, May 2001.

Wiersma, B.J. and J.B. Elder, "Structural Evaluation of Flaws Detected during Ultrasonic Examination of a High Level Radioactive Waste Tank," Proceedings of the ASME Pressure Vessels and Piping Conference, New York, ASME International, July 2003.

Wiersma, B.J., K.H. Subramanian, R.L. Sindelar, M.E. Dupont, J.B. Elder, W.R. West, R.S. Waltz and V. Cech, WSRC-TR-2002-00061, In-service Inspection Program for High Level Waste Tanks, US Department of Energy, Office of Scientific and Technical Information, February 2002.

 

* Savannah River National Laboratory, Materials Technology Section, Aiken, SC 29808; (803) 725-9844; fax (803) 725-1744; e-mail <james.elder@srs.gov>.

Copyright © 2004 by the American Society for Nondestructive Testing, Inc. All rights reserved.

 

---

 
Copyright © 2012 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.