Summer 1996 © 1996 All rights reserved. Photos by
Related articles: Repairs carried out under intense media scrutiny Cause linked to guide structure damage
|
When the bulk of the core had been removed, the refueling crew members maneuvered an underwater camera to the bottom of the Alpha 7 assembly. They saw that the flow plate at the bottom of the fuel assembly was ruptured, and seven fuel rods protruded through the opening. The legs on the bottom of the assembly were pushed outward, wedging the assembly in place. A check of industry operating experience and calls to other plants and the fuel vendor revealed that Palo Verde personnel were facing a unique situation. Before the remaining fuel bundles were offloaded, Palo Verde’s refueling crew had a preliminary plan and had contracted with Framatome Technologies, Inc., to supply hydraulic jacks to go under the assembly to aid in lifting it. Once all the other bundles were removed, the refuel crew was able to remove the undamaged bundles that hadn’t moved earlier, and was left with the problem of A7. The hydraulic jacks lifted the assembly a half inch before reaching the assigned pressure limits. The refuel team then tried to remove the bundle by normal manipulations of the refuel machine – and heard a loud, metallic bang. The upper end of the fuel assembly had moved upward slightly, with no corresponding movement of the bottom portion. This indicated that the guide tubes, which give rigidity to the assembly, might have broken. Developing a recovery plan
Attacking the problem John Steward, Radiation Protection Department leader, helped make sure that the proper radiological controls were in place. "As far as radiation protection was concerned, the most important thing was maintaining fuel clad integrity." The radiation protection staff set up special monitoring equipment to give prompt indication of fuel clad damage. Long stainless steel straps fabricated on site were hooked to the assembly to add stability. At the same time, other response teams brainstormed on how to move the assembly. The first step for many team members from other engineering disciplines was to learn the exact dimensions and structure of the fuel assembly and fuel-moving equipment. The task of educating fellow team members mainly fell to a relatively unique group. Palo Verde has 13 employees who hold limited senior reactor operator licenses just for refueling activities. Through the year, the LSROs work in maintenance and engineering positions, but six of the 13 are generally activated each outage. Says LSRO Steve Jones, a system engineer, "We do requal training just before the outage, going over procedures, machine operations and industry events." Catching the fuel rods Design Engineer Frank Nagode and System Engineer Kevin Graham were assigned to the team charged with designing apparatus that would ensure the fuel bundle’s stability – both at rest and while being moved to its final storage location. Nagode was on the day shift design team led by Mike Hodge and Graham on the night shift team led by Mohammad Karbassian. Essentially, the engineers had to design a cage to hold the fuel assembly together. ABB Combustion Engineering, the fuel vendor, recommended assuming the guide rods were broken and the assembly had no structural stability. Says Nagode, "We designed assuming the supports were severed and any shifting might cause damage to the fuel pins." For this reason, ABB-CE also suggested inserting steel rods into the guide tubes. Because the machine that takes the fuel to the spent fuel pool – called the "upender" at Palo Verde – has only about a quarter of an inch of clearance around a fuel bundle, the "cage" had to add almost no width to the bundle. The quarter-inch steel straps initially used to stabilize the assembly were too thick, so the engineers thought of aircraft cable. They determined that two doubled loops of aircraft cable, threaded through openings at the bottom of the assembly, crossed and tightened with a jacking plate at the top, should do the trick. The support installation had to be done with long-handled foreign object search and retrieval tools. Says Graham, "When you’re designing this stuff, sitting in a conference room with coffee and doughnuts, almost anything’s possible, but when you think about maneuvering a piece of wire through water 50 feet below you through a hole about an inch and a quarter wide, with only six inches on the other side to retrieve it, it puts a whole new light on things." The tools needed to do this tricky work were joint creations of the people doing the underwater work, the engineers planning the support mechanism and the machine and welding shops. Engineer Boris Bolf served as a liaison between the various groups and was heavily involved in the bear claw design. Graham says designers "pretty much lived in the shop to expedite the whole process. I built relationships with a lot of people I hadn’t had a chance to work with before." Working the poles It’s like threading a needle from the top of a turbine building. It’s like trying to get something out from under a table you’re standing on. While the people who worked the poles used all these analogies to try to explain the difficulties involved, welder and LSRO J.D. Kingston says, "You can’t really appreciate what we did until you’re on the end of that pole. You’re working through 50 to 60 feet of water, and the pole is very flexible." Palo Verde’s refuel group did most of the work and assisted in the design and fabrication of the support apparatus. The group, formed in late 1990s, includes team leader Bill Ryder, plus Bob Anderson, Dave Bell, Michael O’Connor, Bob Crook, Mike Hansford, Thad McCloud and Gary Pace. They are engineers, electrical specialists, mechanics and I&C technicians who have all been cross-trained. They work on refueling at all three units, and between outages do maintenance on all the equipment associated with refueling, from the refueling machines to the cameras. The LSRO program is maintained by the group and four team members are licensed. Says Ryder, "Having this group in place was a definite help. Because of our expertise, we knew what would work with the refueling equipment, and we have considerable practice in working with underwater equipment. "It’s hard to get your orientation looking at the monitor of an underwater video camera. Depending on the camera’s position, sometimes what appeared to be up in the monitor was actually to your left – and after you move the top of the pole, it may take a few seconds before something happens at the bottom." Radiation Protection’s Steward, who had experience using the long-handled tools, also spent many hours operating the equipment. "On a number of days, when I left for home it was very hard to open my hands." The poles are in 10-foot sections, and as each section was pulled out of the highly contaminated water, radiation protection technicians wiped them to remove any contamination. Workers were limited to four-hour sessions because of the exhausting nature of the work. The workers used L-shaped pole attachments made on site to thread the aircraft cable. At the top of the assembly, the end loops of cable were supposed to loop over hooks on the jacking plate – but the springy cable wouldn’t cooperate. The cable resisted being placed over the hooks and, once on the hooks, the loops tended to spring off. As they struggled, Kingston said, "I’ve got an idea," and in a few hours returned with a new pole attachment that worked. Then Nagode designed weights to hold the loops on. Once all four loops were in place, turning the screws on the jacking plate tightened the cables. Meanwhile, work was under way to free the lower end fitting from the guide pins by cutting a slot in each foot with an electro-discharge machining tool that vaporizes metal with an electrical arc. Design Engineer Winston Borrero served as Palo Verde’s liaison to ABB-CE’s Connecticut office, where ABB-CE engineers applied pressure to a fuel assembly model with a hydraulic jack to recreate the damage. They used the model to determine the best place to cut. Because of the buckling of the lower end, the assembly no longer fit into the refueling machine hoist box – so the refuel crew used a manual hoist with a mechanical stop that ensured at least eight feet of water above the fuel assembly. Every pull on the hoist chain brought the assembly up only a quarter of an inch – and it had to be hoisted more than 50 feet. Murphy’s Law strikes
again Even though the numbers said the aircraft cables were strong enough to hold the bundle together, everyone was more comfortable with the suspenders-and-belt method – lifting the bundle with the heavy steel straps still on, then removing them prior to placing the bundle in the upender. To do that, the assembly had to be placed in a temporary storage stand in the deep end of the pool. The stand – variously nicknamed the "turkey shooter" and "rocket launcher" – was made by the welding shop out of large pieces of pipe. It took about a day to untangle the cables, move the assembly over the bear claw and clamp the claw on. Then the machine operator moved the assembly to the temporary stand, the steel suspenders were removed, and the assembly was ready to go to its final resting place. Says Jones, "It spun a little again, but that was all right, because I wanted to check it. It spun all the way around in front of me so I could see everything without moving the camera. It decided to be a little nice to us there at the end. It gave up, I think, like a fish. We’d practiced with a dummy bundle, and there was no problem putting it in the upender at all." Then, says refuel team leader Bill Ryder, "Everyone said, ‘I’m glad that’s over.’ But it wasn’t over for us. We modified all the equipment to get the bundle out, and had to put it back in condition for reload. We then assisted in the upper guide structure repair and went immediately into core reload – and did it in 59½ hours." Behind the scenes Paperwork was a crucial part of the job. The removal was controlled with appendices to the core reload procedure. In addition, the design had to be reviewed by the Nuclear Regulatory Commission under the provisions of federal regulation 10 CFR 50.59. A Plant Review Board of senior plant managers had to review the revised procedures and 50.59 explanations. Nance Henry, technical management assistant, operations, helped coordinate day and night shift plant review boards. LSROs Brian Blackmore and Mark Brutcher wrote much of the 44-page 50.59 package. Mike Dilorenzo, maintenance engineering, worked on the appendices to the refueling procedure. "It would change direction several times per shift. Basically, it was the normal routine for getting procedures reviewed, but at 100 miles an hour instead of five miles an hour. We wanted to make sure we got comments from every group that had a legitimate reason to comment on it." Ralph Rothenberg led the team of engineers, including Chengyu Hsia, Kevin Jones, Warren Jones and Chris Wandell, who did structural analyses of the components used in the removal. Designers Don Thompson and Steve Kretzinger created the computer drawings needed for the procedures and regulatory approval and to help explain the situation to the news media. The four-foot-tall team
member When the final piece needed to support the jacking plate was designed, the stress analysis said that the plate should be three-quarter-inch steel. There was none to be found, until one engineer said, "I’ve seen some three-quarter-inch steel recently." A plate that was part of the model is now on the Alpha 7 assembly in the spent fuel pool. Outage ends on schedule Palo Verde managers reexamined the outage schedule, finding ways engineering could work in parallel with operations and maintenance – similar to the way the fuel assembly project had worked. Volunteers ranging from front-line workers to vice presidents pitched in on cleanup – and the outage was completed within the original 49-day schedule. Lessons for the future "We told the media that about 50 people were working on this problem, but when we totaled it up to invite people to the picnic we held to recognize their contributions, we found more than 200. It’s a real challenge to communicate to that many people something that’s changing by the hour." Joe Flynn, welding shop team leader, says, "During normal maintenance, sometimes there can seem to be lines drawn between departments. All of that disappeared. I think we learned a lot about what can be accomplished when you get these barriers out of the way. Design engineering spent a lot of time in the shops and gave our people the opportunity to know them a little better. It wasn’t just getting the fuel bundle out. It was showing each other what we’re capable of." Bailey agrees, "That’s our reengineering vision when you think about it – assembling teams that have the talent to solve the problem, working side by side with the same goal. This type of event helps ignite the culture change it takes to move in that direction faster. It shows what can be accomplished when those barriers are broken down. I think that transferred to our outage success." Says Executive Vice President Bill Stewart, "I’m very proud of
the Palo Verde people. I’m proud of their professionalism, the way they performed
under intense scrutiny and in ultimately making the schedule in a good, safe and effective
way. I think it’s very important to celebrate success." So they had a parade,
complete with the high school marching band. "I think something like the parade is an
important way to reflect on success, feel good about the future and remember that we have
shown we can make it happen." |
