By: Robert B. Candelaria, Salt River Project
The Navajo Generating Station has a history of completing major overhauls in eight weeks or less. This time frame has been maintained while completing all the major projects expected during a 30-year life, including the tie-in of scrubbers, and despite the lengthened time between outages that has become the industry norm in recent years. When estimates to completely rebuild NGS's hot side precipitators came in at 14 weeks, we went to work with our partners to fit the job within NGS's normal major overhaul window.
For large generating units like the Navajo Generation Station, forever gone are the days that overhaul maintenance activities are afforded the luxury of 3 to 4 month outage windows. Operating challenges, today, revolve around developing long-term maintenance strategies, 10-year budgets, aggressive work outage schedules and design improvements that serve to increase plant efficiencies and improve environmental stewardship.
Environmental stewardship is especially important at a high profile plant like the Navajo Generating Station, which is situated on the Navajo Nation in northern Arizona and near the Grand Canyon and an array of other National Parks, Monuments, Recreational and Wilderness Areas.
All these were major factors as we began the planning process for our 2003 through 2005 outage work projects.
Electrostatic precipitators at Salt River Project's Navajo Generating Station.
NGS is owned by the United States Bureau of Reclamation, Salt River Project, the Los Angeles Department of Water and Power, Nevada Power Corporation, Arizona Public Service Company and Tucson Electric Power. Salt River Project operates the plant. NGS operates three 750 MW net capacity coal-fired boilers. Our coal supply is low sulfur bituminous, mined at Black Mesa on the Navajo reservation. Particulate control is performed by three early 1970s vintage Joy-Western hot-side precipitators. Each boiler discharges particulate and gases through two outlet ducts to a precipitator, which is arranged in piggy-back fashion with eight chambers, six fields deep, on top, and eight chambers below. In the mid 1990s, we installed three wet limestone forced-oxidation scrubbers to reduce the plant's SO2 levels as part of a negotiated settlement.
We had been using Midwest Power, and later BHA Group, to provide support services for maintenance on the precipitators for years. The precipitators suffered from the usual problems you'll find with these hot-side units: warped collecting plates, unreliable discharge electrode wires and weights, deformed structural members, etc. But the ongoing maintenance work kept them running pretty well, and also provided a side benefit — the Navajo workforce was becoming familiar with precipitator projects.
When the scrubbers were installed, we did not need to make major improvements to the ESPs to achieve our emissions goals. By 2001, however, we determined that ESP modifications would be required to extend the useful life of the precipitators and to improve the operating efficiency of the scrubbers. The ESPs were approaching 30 years of service, and it became evident that maintenance was no longer the answer — the units needed to be rebuilt to prevent excessive particulate carryover to the wet scrubbers, which could potentially become "concrete factories" from too much ash. Our primary goal for investing in the ESP rebuilds was to increase equipment reliability. However, we also knew that improving ESP performance would have a positive impact on our overall emissions reduction efforts.
Time is NOT on Our Side
After Salt River Project conducted a technology review, we decided to upgrade the precipitators to a much better design to give us flexibility to meet future regulations, and keep them running for another 30 years. The upgrade involved demolishing the old system, Joy 6-foot by 30-foot 18-gauge steel curtains with a weighted wire system, and installing BHA's ESP-1 plates and RDE-1 Rigid Discharge Electrodes.
The target date for our first ESP rebuild (Unit 3) was February 2003. We began planning for this outage in April 2002, a full ten months before the project. After reviewing the unique aspects and challenges of this project, and benchmarking with other plants that had performed similar work, the original consensus was that a 14-week outage was necessary to complete our precipitator rebuild.
Installation of new rigid discharge electrodes through the hopper of the ESP
However, the14-week proposal wasn't acceptable. We needed to find a way to compress the project into an 8-week timeframe.
Most people have heard the old adage, "If you fail to plan, you plan to fail." One of the best decisions we made was to identify our project partners early and include them in our planning. After interviewing several vendors, we chose BHA Group for the engineering, parts, and voltage controls, and we chose Zachry for the construction management. These companies were critical in helping us find creative ways to accomplish this project in 40% less time than normal.
A key challenge of this rebuild was the lack of space around our precipitators. Each boiler has two precipitators stacked vertically, with tight clearance on all sides. This prohibited us from removing or installing internal components through the top of the units. Instead, all components would have to be moved through holes cut into the hoppers underneath. Another challenging factor in these projects is our use of local Navajo labor crews. These crews already had some experience with precipitator maintenance work, but the rebuilds presented new challenges. Intensive training was identified as a critical aspect of the project's success. Further complicating the job were two other 'major' jobs during the outage — an economizer change-out and replacement of the hot re-heat piping.
We knew that the complexity of the project plan would require everyone from NGS, BHA, and Zachry to be on the same page every day of the outage. That's why we started the planning process ten months in advance. Around May 2002, our meetings with BHA focused on the engineering plan and the pre-fab staging process. During the summer 2002, BHA began manufacturing the new internal parts and developed a staged delivery schedule. We identified a large lay-down area where the components would be staged as they were delivered over the summer and fall. In concert with BHA, Zachry joined in the planning to detail the construction execution plan. It was critical for our partners to communicate with each other on each aspect of the plan.
Work on the Unit 3 precipitator began on schedule in February 2003. The project included (1) replacing precipitator voltage controls; (2) washing down the unit; (3) removing 864 collecting plates and high-voltage rappers, 3,456 collecting plates, 26,880 wires and associated 30-pound weights, and frames; and (4) installing 2,592 new collecting plates, rappers, 9,984 rigid electrodes, and frames. Safety and communication were extremely important since most of the work was from the "bottom up" on the units. Most of the new internal components were about 30-40 feet long, and relatively fragile.
Staging area for new ESP components
Critical to the smooth completion of the job was developing a traffic pattern for all materials for each job, including the economizer and the hot re-heat job. Avoiding logjams was of the highest priority. Once materials were staged on the hopper decks, they were attached to monorail trolleys for transport to the specific chamber. Loads had to be angled up into the hoppers using air tuggers mounted on the roof and lined through to the deck below the hoppers. A third party contractor was hired to handle all the material for salvage. Material was removed and loaded directly onto waiting trucks for salvage.
The Unit 3 project was completed in 63 days and the unit was back on line four days later than the projected schedule. Even with the extended planning period, there were many unforeseen issues we had to address. These issues were resolved, but they forced us to take an even more critical look at our planning process. Problems happen on almost every major project, and you can choose to either learn from them or leave them behind. We chose to learn.
Looking down into the ESP, showing the new collecting plates and V-pin rigid discharge electrodes
In April 2003, we began the process all over again to prepare for our rebuild of the Unit 2 precipitator in January 2004. Again, we selected BHA and Zachry as our partners. Learning from the challenges in the first project, our planning was even more detailed and more exact. We expected to complete the project in 54 days and we expected an even higher level of coordination to eliminate unnecessary problems. But we were wrong. The Unit 2 precipitator project was completed 10 days faster, in only 52 days! We are pleased to report that the Unit 2 ESP rebuild project was very successful and that preliminary testing indicates the unit is performing well.
What did we learn? First, we knew that planning would be important. We attacked the Unit 2 precipitator, which was much more logistically challenging than Unit 3 because of the proximity to Unit 1, from two sides instead of one, adding a crane and a crew. We added several engineering pieces that allowed for more expedient handling of the materials, we added a construction elevator, and we added more training. We assisted the training effort by constructing a model of one of the fields with plates, rigid discharge electrodes, frames, conical insulators and all structural support pieces complete with ceramic bar stabilizers.
We could not have expedited the project without having a complete plan and the input from our partners BHA and Zachry. We also learned there is no such thing as "overtraining." We spent many hours working with all our crews to explain the process and the timelines and it paid off. It was impressive to see how well the Navajo crews performed and developed during the outages. Zachry had utilized Navajo supervision and added a lead man program to further develop more Navajo worker supervision.
We learned a lot on our Unit 3 outage and applied it to our Unit 2 outage. Applying the lessons learned paid off. We brought Unit 2 back on-line two days early. Before the completion of the Unit 2 job, just prior to start-up and after air loads, we called our key people together to once again brainstorm for the next outage. We put together our next list of "lessons learned" covering the good, the bad and the ugly. Those lessons are now being applied for our Unit 1 major outage coming up in January 2005. Stay tuned U there's more to come.
Robert B. Candelaria is the O&M Manager at Salt River Project's Navajo Generating Station.