UCC completed their first Steam Dryer repair in 1983. Since then UCC has made various repairs or modifications to BWR Steam Dryer. To date, UCC has performed more than 30 different projects on Steam Dryers all over the world. These repairs range from repair of cracked welds on bolts and drain channels to a complete volume reduction and disposal project using underwater plasma arc cutting.
A related unique project involved separator support leg modification. The OEM removed the existing separator support leg with underwater EDM. UCC welded the new Separator legs in place. These legs were welded on the lower most part of the Separator where the on-contact dose was approximately 5 to 10 Rem depending on the survey point.
Over 300 dives were performed with no skin contaminations. Radiological conditions were extreme with contact dose rates in the sludge ranging from 0.050 Rem/hr to 30 Rem/hr and averaging about 2 Rem/hr. Divers encountered debris with dose rates of up to 600 Rem/hr on contact. To put this in perspective, high radiation is considered to be a dose rate exceeding 0.100 Rem/hr. An actual whole body exposure of 400 to 600 Rem can result in acute radiation poisoning causing death within 30 days.
Taking advantage of the shielding provided by the water and other dose mitigation techniques, we were able to keep the diver in a 0.001 to 0.005 Rem/hr field during the majority of the work. Even though brief excursions into high dose areas were necessary, an average exposure rate of 0.006 Rem/hr was maintained for the entire dive team for the course of the project. This includes Divers, Tenders, and support personnel. Divers typically averaged 0.003 Rem/hr during diving operations keeping total exposure well below expected levels.
Over 10,000 pounds of basin debris was managed and relocated. Divers operated a desludging system to collect approximately 1,500 ft3 of sludge in a fraction of the planned task duration. Overall, it was successfully demonstrated that, on a task by task basis, divers could work more efficiently while receiving significantly less radiation exposure than topside workers using remote tools or extension poles. Much of the work performed underwater would not have been possible without the shielding afforded by the water barrier and direct access by the divers.
Underwater coating repair in commercial nuclear plants is an accepted maintenance approach. Some of the early applications have now been in service for over 15 years. Surface area covered ranges from several square inches to several thousand square feet over both steel and concrete substrates. One of the first completely documented underwater coating projects took place in 1986. Thousands of repairs were performed in a BWR Mark I suppression pool. Additional repairs were also performed in the Condensate Storage Tank. The condition of these repairs has been periodically monitored and they continue to perform well while the principal coating continues to degrade.
Three case studies are presented that help to illustrate the effectiveness of the underwater repair technique.
Case Study #1: The Torus at a Mark I BWR was drained and completely recoated in 1986. During the following refueling outage, an underwater coating inspection was performed to assess the performance of the newly installed coating system. The inspection revealed extensive blistering and significant loss of adhesion over large areas. Fractured blisters and coating delamination exposed the steel substrate in numerous locations and pitting of the steel substrate had occurred.
Underwater coating repair was used to reestablish the protective corrosion barrier and prevent metal loss at the Torus pressure boundary. Coating repair application proved to be extremely difficult due to the lack of adhesion in the existing coating. Ultimately, the repair coating was used to tack down loose coating and protect exposed areas on the Torus shell.
Inspection during several subsequent outages has shown that the repairs are performing as expected. No significant defects were noted in repair areas and pull testing showed adhesion values were much higher in the repair coatings than in the principal coating.
Case Study #2: The Torus at a Mark I BWR was inspected underwater during the plant’s first refueling outage in 1990. The inspection revealed extensive blistering and moderate loss of adhesion. In addition, it was necessary to remove instrumentation that had been installed prior to start-up to monitor operating conditions. The removal of strain gauges, pressure transducers, and associated wiring and piping exposed large areas of steel substrate on the pressure boundary.
Numerous small coating defects were repaired as well as several large areas in excess of 100 square feet. Recent follow up inspections confirmed that repair-coating performance is excellent. No significant substrate corrosion was noted.
Case Study #3: Underwater coating repair was performed in the Torus at a Mark I BWR in 1996. At the next refueling outage, the Torus was drained for a recoat. The painting contractor used abrasive blasting to remove the old coating and provide a suitable surface profile. During the surface preparation, the contractor reported that blasting easily removed the original coating but left the repaired areas intact.
Coated carbon steel pipe in two concrete emergency cooling water basins demonstrated extensive coating failure and corrosion (see Figure 2) leading to thru-wall pitting of the pipe. Corrosion rates were extremely high due to water chemistry and microbiologically influenced corrosion (MIC).
Both small and large areas were coated with an underwater epoxy. Thousands of areas totaling several hundred square feet have been touched up. Underwater applied coatings have been in service for over 5 years and are performing well. Field adhesion testing was used to verify that adhesion values are adequate.
The concrete roof slabs of SSW Basins A and B demonstrated full thickness cracking of the concrete slab covering a large portion of each basin. The roof slab extends below the water line in each basin. A previous report recommended sealing the cracks on both the atmospheric and immersion faces of the slab to prevent exposure of the rebar to excessive moisture intrusion. The atmospheric side was sealed by the application of a coating system over the entire roof slab.
A flexibilized underwater epoxy supplied by TFT of Houston was selected for repair of the identified cracks. Extensive testing was performed to verify that the material could be applied underwater on an overhead surface. The selected material demonstrated the ability to cling to an overhead surface even under the turbulent conditions produced by the divers exhausted air bubbles. It also demonstrated sufficient post cure adhesion to the concrete surface.
High pressure water at 4,500 to 4,800 psi was used to remove loose material and clean deposits from the surface of cracks and adjacent concrete.
Plural component caulk guns were used to mix and dispense coating material. This eliminated pot-life concerns and allowed efficient underwater transport and dispensing of coating. Storage and dispensing were carefully tracked to permit control of all material dispensed within the Basins. Material was applied at a wet film thickness of 10 to 20 mils. Wet film thickness was checked periodically during application with a wet film thickness gauge.
The application method forced the coating as deeply as possible into the crack. Coating extended out approximately 2” on either side of the crack but did not overlap any unprepared surface. After the initial pass to apply the coating, several subsequent passes were necessary to adjust wet film thickness and ensure the coating did not pull away from the substrate.
A total of 3,583 feet of crack equaling approximately 1,800 square feet was coated. Inspection of the applied coating was performed by ANSI N45.2.6 certified Level II and Level III Coating Inspectors. The inspection was conducted under the supervision of a Level III Coating Inspector.
Visual inspection of the applied coating shows that the coating was applied per the approved procedure. Post cure adhesion tests were performed and adhesion values were fund to be acceptable. A follow-up inspection was performed in August 2004. All coating appears to remain tightly bonded to the surface with no evidence of cracking or disbonding.
The plan called for divers to work at a depth of 90 feet. Diving work at this depth can be conducted safely using surface supplied air, but time on the bottom is limited. The no decompression limit at 90 feet on air is 30 minutes. After 30 minutes, the decompression obligation quickly escalates and work times beyond one hour are impractical and unsafe.
To allow increased bottom while minimizing decompression, it was decided that Enriched Air, or Nitrox, would be used. Nitrox is a mixture of nitrogen and oxygen where the percentage of oxygen is varied to reduce the amount of inert gas in the breathing mixture. Breathing gas with a higher partial pressure of oxygen allows a shallower equivalent air depth to be used when calculating decompression time.
In this case, the breathing gas was mixed using 65% oxygen and 35% nitrogen resulting in an equivalent air depth of 70 feet. This increases the bottom time to 50 minutes without any required decompression. However, to work productively, a diver needs to be able to spend two to three hours on the bottom. Two and one half hours at 70 feet would require over seventy minutes of in-water decompression time. At three hours bottom time, the decompression would approach two hours.
Using a procedure called surface decompression on oxygen, or SurD-O2 it is possible to dramatically reduce the decompression times and permit the diver to spend all or a portion of his decompression time in a decompression chamber making the process more comfortable and much safer in a controlled environment. In the case of a 70 foot dive for 2.5 hours, SurD-O2 reduces the decompression time by 40% all of which can be spent in the decompression chamber.
All other things being equal, most diving contractors would prefer to work a twelve-hour day. This usually allows them to establish a dive rotation that includes at least three dives. Table 1 illustrates the productivity gained by utilizing Nitrox and surface decompression on oxygen. In each case, the three divers would spend a collective nine hours involved in diving, but the divers using Nitrox and SurD-O2 would enjoy an additional two and one half hours of productive time over the air divers. They would also reduce their decompression time by 62% and spend no time decompressing in the water.
The clay liner of an 842 acre reservoir supporting a 1872 megawatt pump storage facility developed large trench-like features. Reservoir depth ranged from 40 feet to 110 feet. Before a repair could be designed and implemented it was necessary to determine the extent of the cracking. The size of the reservoir and the depth would make this a difficult task.
Engineers working with a diving contractor and marine navigation firm developed a mapping plan that included the use of side scan sonar, short baseline acoustic diver positioning, and differential global positioning (DGPS) of the support vessel. Sonar was used to scan large areas and roughly characterize the cracking. Divers equipped with helmet-mounted video cameras then inspected each feature.
The divers’ position relative to the support vessel was mapped by computer using the acoustic tracking system. The position of the support was plotted by DGPS. By correlating this data, a computer map could be generated to show the precise position of features in the liner.
Divers also entered the features, some of which were ten feet wide and twenty feet deep to provide additional data in support of the sonar findings. Video images and depth readings from the diver pneumo-fathometer helped to ground truth the sonar data.
Engineers were able to use the data to develop a grout mix design and accurately estimate quantities. Survey maps were then used to quickly relocate features during the repair phase.
A concrete gravity dam with an earth core and rock/soil foundation was completed in 1944. The dam is 8,422 feet long, 206 feet high with a 130 foot head. Maximum storage is 6,129,000 acre-feet and the purpose is navigation, flood control, power production, and recreation. The downstream side is a radial gate controlled spillway 960’ wide with a maximum discharge of 1,050,000 cfs.
Eddies created by a change in the gate operation sequence deposited gravel on the downstream concrete apron. A resulting eddy between the downstream flow blocks and the apron toe caused a ball mill type action that eroded the concrete apron.
The repair designed called for the placement of number six rebar in each erosion area as shown in Figure 9. Twelve-inch deep holes were to be drilled on three-foot centers and the reinforcing steel was epoxy grouted in place. Since underwater visibility was limited to a few inches, the Engineer directed divers to construct a wire template to aid in aligning each hole. Measurements were then taken so that each piece of reinforcing could be sized to leave three inches of cover at the repair surface. Concrete would be placed by bucket and designed so that it could be dropped through up to seven feet of water.
The success of the repair depended largely on the concrete mix design. Because the diving contractor was experienced in placing concrete underwater, he was able to make helpful recommendations to the Engineers designing the concrete. The mix would need excellent cohesive strength in the submerged application and had to be resistant to washout of cement. Its ability to self-compact and self-level and its workability at low water to cement ratios would also be important. Because of the poor visibility, the divers would need as much working time as possible to place the concrete.
Divers used oxy-arc cutting torches to remove damaged 1-1/4” steel reinforcing. Loose debris was removed from the repair areas using a combination of low pressure water jetting and hand dredging. Damaged concrete was then removed by 5,000 psi high-pressure water. Before placing concrete, divers conducted a video inspection so that the Engineer could verify the preparation of the repair areas.
A cable was anchored to the apron adjacent to the repair area. The concrete bucket was tethered to the cable at the surface and then lowered to the bottom as shown in Figure 10. Once the bucket reached the apron, the diver could safely move in and adjust its position. Buckets were positioned within twelve to eighteen inches of the steel reinforcing and then opened to release the concrete. Only minimal hand toweling was required to provide the desired surface. A final video inspection was performed to approve each repair.
On June 24, 1998, divers from Underwater Construction Contractors performed an underwater video and ultrasonic thickness inspection of a water intake barge located on Florida’s Hiwassee River. The ultrasonic inspection of the hull revealed plate thickness ranging between 0.360 and 0.380 of an inch. These measurements were consistent with as-built drawings indicating the hull was constructed from 3/8-inch steel plate. During the inspection, a pattern emerged showing reduced wall thickness (0.005 to 0.010) on the starboard side. Upon inspection of the inner hull, approximately two inches of standing water was found on the starboard side. The port side was found to be dry.
The hull coating was found to be about 90 to 95 percent intact, although rust nodules were found over the entire submerged surface of the hull near the bow. Upon removal of several of the rust nodules, pitting was found. Pit depths ranged between 1/16 to 1/8 of an inch. A majority of the rust nodules appeared to be concentrated along the weld seams. Several inches of water were found in the bilge. It was believed that at least some of the leakage was caused by pinhole perforations of the hull at pitting corrosion sites.
The barge is permanently moored and cannot be dry-docked. It was decided that underwater coating repair of the pitted area would be attempted in hopes that it would not only arrest corrosion but plug any pin hole leaks.
The divers cleaned the surface with a 3,500 psi pressure washer. This was followed by a mechanical surface prep to SP 11 using a 3M Clean N Strip to remove any remaining corrosion and roughen the coating around the pit. Bio-Dur 561 was applied at 40 to 50 mils. Approximately 500 repairs were made, ranging in size from 1/2 inch to three feet in diameter. TFT epoxies are distributed by Progressive Epoxy Polymers, Inc.
The work was performed in a one knot current. Water temperature was 50 degrees F, with visibility at approximately two feet. After the coating was fully cured, the main compartment was sealed and pressure tested to five psi. No pressure loss was detected over 30 minutes. Subsequent inspections indicate there has been no additional leakage.
UCC recently performed a complicated repair project at facility serving over three million people consuming one billion gallons of water per day. After years of immersion, the plant infrastructure was exhibiting signs of aging. Making the needed repairs required the careful selection of materials and the development of specialized underwater repair techniques. In addition, it was crucial that the repair not in any way adversely affect operation of the plant or water quality. Finally, the work would take place at water temperatures near freezing.
Early January 2006 — Working closely with the Owner and the Owner’s Engineers, UCC researched various cementitious and epoxy products that would yield good curing results in 34ºF water and still meet the water plant’s demanding performance and schedule requirements.
February 28, 2006 — Underwater mock-up testing was completed by UCC Divers in the St. Joseph River in Michigan. Four promising samples were sent for compression testing. One product stood out and was approved for cold potable water application.
April 21, 2006 — Underwater Construction Corporation began work to repair 12 leaking expansion joints that existed between the head house and filter building. Repair required that divers drill more than 1,000 holes and place material over approximately 500 linear feet of joint.
June 23, 2006 — Work on the initial phase was completed accident free and under budget.