By Jeff Benoit, VP Combustion Engineering, Power Systems Mfg., LLC (PSM)
Chris Johnston, Airfoil Engineering Program Lead, PSM
Martin Zingg, Manager, New Technology Introduction, PSM
In a highly competitive market environment, power generation companies are constantly seeking ways to drive operation and maintenance (O&M) costs down on their industrial gas turbine units to maintain profitability. With soaring fuel prices there is an increasing focus on reducing maintenance cost to remain ahead of competition. Specifically for 501F gas turbine operators, lower prices in conjunction with extended durability of turbine blades and vanes, transition pieces and other high-cost capital spares can significantly improve bottom-line profitability. In addition, increased parts longevity will not only yield reduced refurbishment cost, but will also allow extended operating intervals between scheduled maintenance, thus providing better availability in the marketplace.
Over the last seven years, Power Systems Mfg LLC (PSM), a provider of re-engineered hot gas path turbine and combustion system components for the power industry’s F-class gas turbine fleet, has mantained close partnerships with many 501F operators. This collaboration provided insight into some of the problematic O&M issues facing 501F operators, along with first-hand exposure to fleet 501F engine operation. This made it possible to incorporate and validate hardware redesigns that improve part durability while ensuring optimum engine operability. As an example, redesigned combustion transition pieces (TP’s) for 501F engine types have accumulated 24,000 fired hours without replacement or refurbishment, well beyond the current fleet standard 8,000 to 12,000 fired hour inspection/refurbishment intervals.
Identifying the Issues
Unscheduled engine outages can occur as a result of premature part failure in the gas turbine flow path. As a consequence, initiatives were launched to address these fleet-wide combustion and turbine hardware field issues. The effort was initiated at user request after a large number of 501F machines began exhibiting chronic TP failures across the fleet, causing forced outages, which significantly affected revenue generation. Investigations revealed that the 1st and 2nd vanes experienced premature platform oxidation/erosion and airfoil cracking. Creep on the outer diameter hooks of the 3rd vane allowed the parts to move aft and contact the rotating hardware. The 1st blade frequently showed brazed tip plate failures and developed a large platform crack at the leading edge suction side requiring the part to be removed prematurely. The 2nd blade showed cracks in the airfoil trailing edge root area. In cooperation with a operators at a large number of units it was possible to clearly identify the issues and to initiate the necessary steps to address them.
The largest 501F operator requested to deliver redesigns to solve field issues as quickly as possible, after multiple part failures throughout their fleet had affected fleet availability, causing significant revenue losses. A detailed root cause assessment (RCA) of the failed parts was initiated in 2001. For the TP, the initial focus was on analyzing dozens of failed 501F transition pieces from multiple gas turbines. Subsequently, combustion systems were instrumented on several machines to collect engineering data during unit operation and also to determine the thermal, pressure and dynamic boundary conditions on the combustion system parts. Armed with this information, it was possible to develop finite element analysis (FEA) models for structural assessments and computational fluid dynamics (CFD) models for air flow variation.
At the same time, mobile laser scanning technology was used to characterize combustion components and create statistical geometric and cooling flow models. The RCA revealed a number of key factors that contributed to the TP failure, including the existing body shape, construction and fabrication methods, material, cooling scheme and mount design to the first stage nozzle retaining ring. Similar RCA procedures were applied to understand the root causes of the various failures observed on the turbine blades and vanes. This RCA was indispensable in determining redesign solutions.
Redesigning the Part and Collecting Runtime Hours
Equipped with the RCA results and the life cycle cost requirements from the largest 501F operator, the engineering team leveraged an ISO-certified engineering design process to eradicate existing design flaws.
Transition Piece: The new transition piece features a rounded body shape that balances the heat transfer loading both internally and externally and eliminates resonant frequency concerns (Figure 1b, page 140). The body wall itself is made from Haynes 230, an oxidation-resistant nickel-based superalloy. The TP seals, which engage the first vane row and control leakage flows, along with the mounting design, are changed to prevent the seals from binding with the vane. This has been an issue with the current original equipment manufacturing design and allows better leakage flow control and assembly. An improved TP exit structure (called the “picture frame”) reduces the thermal fight between the frame, which tends to run cooler, and the hotter TP body. Coupled with a thermal barrier coating, wear-resistant couplings and an innovative, patented effusion cooling system, the part was redesigned for 24,000 equivalent fired hours of operation between replacements. The first set of TP’s, some of which were instrumented and covered with thermal paint as shown in Fig 1a (page 140), was installed in a 501F in June 2003. The parts met all design objectives. In particular, measured combustion dynamics pressure levels (one of the key drivers in component failures) were cut in half.
 Figure 1a 501F TP installed |
null
 Figure 1b 501F TP model |
Field results of the redesigned 501F parts have been excellent. Experience demonstrates that the parts exceed original part performance. The transition piece fleet leader has accumulated 24,000 operating hours without refurbishment and is in excellent condition. Figures 2a, b, c and d (page 142) show the new transition piece after 23,659 actual fired hours and 368 starts. As the pictures show, panel cracking (a typical failure with the current fleet) is completely eliminated with the new design, avoiding forced outages and improving the 501F RAM performance (reliability, availability and maintainability). As for refurbishment cost, the redesigned seal system eliminates the sometimes expensive rework or replacement of the TP’s picture frame.
 Figure 2a-d Transition piece after 23,659 operating hours and 368 starts |
null
 |
The redesigned TP also features a rebalanced and homogeneous seal leakage flow at the interface to the 1st vane. This measure results in more uniform metal temperature profiles and lower average temperature at the 1st vane ID platform, thus reducing thermal loads (Figure 2e).
 Figure 2e Controlled uniform seal leakage at interface TP to Vane 1 (view is downstream through the TP into 1st vane row). |
Turbine Blades and Vanes: The redesigned 1st stage vane includes several significant design improvements relative to the original part. Of primary benefit to the plant operator, redirected cooling air allows better cooling of the ID platform to directly attack the persistent ID platform erosion issue widely seen in the fleet. This allows the part to run up to three times longer. In addition, cooling air is better distributed around the airfoil and in the leading edge region of the airfoil to help avoid occasional local TBC spallation and base alloy erosion issues. In a baseload unit application, these changes are proven to result in a part life that exceeds one interval of run time (24,000 hours). The fleet leader is currently at well over 30,000 hours without repair and is still running. Representative photos of this hardware are shown in Figures 3a and b (page 148).
 Figure 3a-b 1st vane design after 28,402 hours of operation |
null
The redesigned 1st stage blade, like the 1st stage vane, incorporates several improvements. Modified platform cooling, a more robust electron beam welded tip plate and a patented trailing edge undercut for stress reduction were added to eliminate existing fleet issues. In a baseload application, the fleet leader is approaching 24,000 hours of runtime without repair. The parts will meet the 24,000-hour requirement despite thermo-mechanical fatigue (TMF) cracking in the platform that was observed in cyclic application. In an effort to further enhance 1st blade longevity, a fix for this issue will be released as well as a repair in the spring of 2008. Figure 4 (page 148) shows the redesigned1st blade after approximately 19,000 hours of operation.
 Figure 4a-b 1st blade design after 18,883 hours of operation |
Significant 2nd vane fleet issues include severe airfoil cracking and an ID platform oxidation/erosion issue. To combat the former, the vane segment was redesigned to a bolted pair configuration. Further, the base alloy was changed from MarM509 (cobalt based) to IN939 Weldable (nickel based). Ample field experience shows the bolted pair successfully eliminated the airfoil cracking issue. Also, design changes were recently implemented to target what is known as the “ID platform erosion” issue. These design changes are currently undergoing validation. The redesigned 2nd vane fleet leader has accumulated over 16,000 hours of operation.
Other major hot gas path component modifications include changes to the 2nd blade and 3rd vane designs. Changes were implemented to further combat the trailing edge stress issue in the 2nd stage blade and the hook creeping issue in the 3rd stage vane. The redesigned 2nd blade design goes beyond the original part’s trailing edge stress-reducing undercut, in an effort to further reduce the stress. The 3rd vane incorporates a base alloy change from X-45 (cobalt based) to Weldable IN939 (nickel based). Inspections show both design changes are successful. Fleet leaders for both components are in excess of 20,000 hours of runtime.
 Figure 5a-b 2nd vane bolted design |
Over the past seven years, the engineering team has worked hard, in a close partnership with numerous 501F operators, to help turbine owners remain competitive. In a commitment to solve the issues on the 501F turbine, extensive gas turbine engineering expertise was leveraged, in conjunction with an integrated product development approach and latest engineering tools and technologies, to yield turbine and combustor hardware that meet industry needs. Ample fleet experience has demonstrated that the improved parts meet the industry’s component durability requirements, significantly improve RAM performance and reduce refurbishment cost. Parts performance in the field is constantly monitored and further improvements implemented as needed, thus constantly pushing the limits of airfoil and combustor technology to create more robust, higher efficiency 501F components, helping the industry move forward. All redesigned parts not only promise, but also have demonstrated, superior performance.
Authors:
Jeff Benoit is Vice President of Combustion Engineering for PSM. He has over 21 years of technology development experience in the power generation industry, ranging from aircraft engines to the nuclear, wind and industrial gas turbine fields. He holds engineering degrees from Clarkson University and the Massachusetts Institute of Technology.
Chris Johnston is an Airfoil Engineering and Program Lead for PSM Airfoil Products, focusing primarily on F-class hardware design and manufacturing. He holds engineering degrees from Clemson University and Univerity of Texas at Austin and an MBA from University of Florida
Martin Zingg manages new technology introduction at PSM. He gained global experience in the power generation industry in various functions, including R&D, field service and product management. He holds a master degree in mechanical engineering from the Swiss Federal Institute of Technology.