D5-D5A Users: Productive year: two meetings conducted, website launched, brainstormed with OEM on fleet R&D needs

The previous 12 months have been the busiest period in the history of the 501D5-D5A Users. Last October, Chairman Gabe Fleck and Vice Chairman Barry Mayhew headed an ad hoc user committee that par­ticipated in a first-ever ses­sion with Siemens Energy R&D personnel to exchange product/service ideas, with the goal of improving fleet performance (Sidebar 1).

Early this year, the group’s first two-day Mid-Year Meeting, featuring a special seminar on gas-tur­bine (GT) repair technolo­gies, attracted the largest crowd ever to that event (Sidebar 2). Shortly thereaf­ter, the organization’s web­site, a couple of years in development, was launched and received very positive reviews from members of the user group (Sidebar 3).

Before the first week of June ended, the D5-D5A users had com­pleted their 12th Annual Confer­ence at the Colorado Springs Mar­riott with near record attendance. The meeting officially started Sun­day, June 1 with a reception and dinner hosted by Allied Power Group.

Traditionally, the first full day of a D5-D5A annu­al conference is set aside for team-building activities. Sounds like an opportunity to have a good time, and it is. But it has a much greater purpose: Make attendees forget the plant and focus on learning and building valuable relationships.

This year, Mitsubishi Power Systems and Emer­son Process Management celebrated their recent alli­ance (see sidebar in the 501F Users Group report elsewhere in this issue) by taking the group on a day-long white-water rafting adventure. Now this really was team building. People have been known to fall out of rafts in rough rapids. Could you count on your raftmates to pull you out of the drink were you to fall in? Luckily no one fell into the water before team-building was complete.

The serious side of the meeting began Tuesday morning. In his opening remarks, Fleck, an electri­cal engineer for Associated Electric Power Cooperative Inc (AECI) with mainte­nance responsibility for the company’s legacy frames, recognized eight members of the group with participa­tion awards (Sidebar 4). He also announced that there were 17 first-timers among the 45 attendees.

 1. Siemens, users benefit from collaboration

The first W501D5-D5A “Voice of the User” R&D brainstorming session exceeded expectations according to participants in the Orlando meet­ing, last October. A follow-up to what likely will become an annual event is planned for this fall.

This workshop featured an exchange of product/service ideas between Siemens R&D engineering and marketing personnel and user-group representatives from Caith­ness Energy LLC, Associated Electric Cooperative Inc, Empire District Elec­tric Co, and Cardinal Power of Canada LP. Chairman Gabe Fleck of AECI led the user delegation.

The formal portion of the meet­ing consisted of a planning and creative-process training session, plus Siemens presentations on long- and short-term business trends, the com­pany’s product-development process, and some FY2007 R&D programs. Next, user-group representatives shared the results of their prepara­tion assignment with the Siemens team. The users wrapped up their participation with each representative suggesting two or three “opportunity areas” for Siemens to pursue.

The bottom line. Kevin Carpenter, frame owner and manager of product development for mature Westing­house frames, and Jeff Kain, engi­neering frame owner and manager of service engineering for the same group of engines, proposed that several products and services ideas identified or reinforced during the meeting be included in the FY2008 R&D planning process.

Mayhew moderated the closed user sessions that dominated the day’s activities. Closed user sessions continued Wednesday afternoon after a morning of presentations by Sie­mens Energy personnel. The OEM completed its presentations Thurs­day morning and hosted a closing reception and dinner. More than 30 companies participated in the vendor fair on Wednesday evening.

What’s bothering the users

Mayhew, the maintenance manager at Cardinal Power in Ontario, is a superior session moderator. He knows the questions to ask, how to get his colleagues into the discussion, and is familiar enough with virtually every piece of the engine to keep the dialog going when others get stuck.

The Tuesday morning sessions were unstructured—that is, the group would entertain questions from colleagues about anything they needed help with. Mixed in with the questions were experiences that some users wanted to share with the group—in most cases to help others avoid “pain.”

Torque converters led off the discussion. They always are a hot topic at D5-D5A meetings, perhaps because Fleck and Stan Wilhelm, a colleague of Fleck’s at AECI, always seem to have one more operational or maintenance issue on torque convert­ers to bring to the group’s attention. With a claimed reliability exceeding three nines highlighted in one sup­plier’s literature, it’s hard to believe there would be need for any discus­sion at all.

One user said that the torque-con­verter manufacturer restricted spin cooling to 15 minutes at his plant—an inconvenience if emergency mainte­nance were necessary and the unit had to be returned to service quickly. The vendor told him that the oil would emulsify if he extended the run time and it wouldn’t lubricate properly.

Another user thought this ridicu­lous and said his GT is spin-cooled for about eight hours after shutdown, until turbine disc-cavity temperature drops to 110F. A hand shot up. The new participant in the dialog cau­tioned that compressor disk migra­tion might result from such long spin cools. Not necessarily so, said the first user. His plant has been doing 8-hr spin cools for 15 years without observing any disk migration.

The discussion was auto-igniting. Yet another user commented that a particular brand of booster pump was used for his torque converter and it was the source of operational problems. He called it a “repurposed sewage pump” and said that rust barnacles formed on internal flow-path surfaces and contaminated the hydraulic oil. With some digging, the user found out that the rust bar­nacles were the result of the pump being tested with water. A few others in the room said they had had the same problem.

Business plans for GT-based gen­eration facilities typically stress lean-and-mean operational staffing and the use of outside contractors for most anything other than ordi­nary day-to-day maintenance. This works well on paper. Fine for the real world, too—unless several plants need the same services at the same time and from the same firm, and the “new hire” is the one assigned to your plant.

Case in point: A user said his torque converter had to be rebuilt. Not taking any chances, he hired the OEM, only to learn that wasn’t a risk-free solution. On the first start following the overhaul, the over­running clutch didn’t kick free as it was supposed to and the machine destroyed itself.

Turning gear. Another “sure” topic at the annual meeting is time on turning gear and associated wear and tear on engine rotating parts. One user mentioned that his unit had been accumulating more than 150 hours of turning-gear time at 2-3 rpm between starts and he knew this was conducive to parts wear.

Decreasing turning-gear speed may be one way to reduce wear. But at the 1 rpm investigated by one attendee, long-term operation can mean loss of the oil wedge. Bearing damage could result. Might there be a benefit to reducing speed further? A user said he was experimenting with 0.1 rpm. There’s nothing to report yet, however.

 2. Mid-Year Meeting highlights

The 501D5-D5A Users con­duct a mid-year meeting, typically in the mid-January to mid- February timeframe, to ensure continuity of informa­tion exchange between Sie­mens and owner/operators on critical issues. Chairman Gabe Fleck and Vice Chair Barry Mayhew find that two meet­ings annually are much better than one for keeping users involved and sharing experiences.

The 2008 Mid-Year Meeting, Orlando, January 31-February 1, was the first in the group’s history to run two days; it featured a special repair technology seminar on Day Two. The conference began with a two-hour user-only open forum—that is, any subject on the mind of the 33 attend­ees (a mid-year record) could be brought to the floor for discussion.

Note that half of these users were participating in a D5-D5A meeting for the first time. Fleck attributed this, in part, to the high rate of turnover in an industry sector that continues to grow and where experienced person­nel have access to a steady stream of promising opportunities.

Another reason, he said, is that those who regularly attend the Annual Meeting in June know how valuable the group’s conferences are and a second meeting allows them to get others in their respective organiza­tions involved. Given current staffing levels, it’s virtually impossible to have two top people out of a plant at the same time.

After the coffee break that fol­lowed the closed session, Siemens personnel joined the users for presentations and col­laborative discussion on topics of special inter­est. This activity filled the remainder of the day.

Craig Weeks, CEO of the service fossil business unit, opened the Siemens portion of the meeting and explained what the com­pany’s reorganization into three sectors—energy, industry, and healthcare—only two months earlier meant to customers.

The Siemens Energy portfolio, he said, extends from primary energy through power distribution and is comprised of six divisions. One of those, Energy Services, is a global organization of 6600 employees that conducts more than 3000 outages annually and supports about 400 sites with gas-turbine-based generation. It is headed by CEO Randy Zwirn.

Weeks stressed the division’s abil­ity to respond quickly to unplanned outages. Its resources are regionally based, he continued, and the 24/7 organization’s 400 field service engi­neers and 500 technicians are sup­ported by rapid response trailers and the latest field-based tooling.

The day-to-day impact on custom­ers of Siemens’ sweeping reorganiza­tion, however, appears minimal. Key contacts in the OEM’s organization for D5-D5A Users, Kevin Carpenter, service frame owner, and Jeff Kain, engineering frame owner, continue in their positions.

Next, Weeks asked customers to be more proactive in providing information requested by the frame owners. The company’s surveys and needs assessments are critical to the delivery of improved solutions that benefit users. The customer’s voice, he added, is critical to the prioritiza­tion of funding and resources.

Weeks then discussed Siemens’ role in user-group meetings and in company-sponsored dialog work­shops. He urged owner/operators to actively participate in the Customer Extranet Portal (CEP) as a way to facilitate access to technical infor­mation required in support of plant O&M activities—including product information and updates, availability improvements, parts catalog, and outage reports and other unit-specific information. New users unfamiliar with the CEP were urged to apply for site access by contacting Dawn McCarter at dawn.mccarter@sie­mens.com.

Weeks finished his presentation with thumbnail sketches on the company’s progress in further developing its qual­ity culture (six sigma) and driving con­tinuous improvement throughout the OEM’s organization; lifetime extension programs to increase asset value; the high degree of success in hiring techni­cians and engineers; and fortifying the field-service infrastructure.

Three presentations at the Mid-Year Meeting of particular interest to users that were not updated at the Annual Meeting in June are recapped below: Row 1/R19 compressor-blade 100,000-EBH (equivalent base-load hours) analysis, lube-oil filtration, and the repair technology seminar on the second day.

Lifetime-extension assessment. Kevin Sheehan, manager of materi­als systems development, walked the group through a lifetime-extension assessment for compressor blades approach­ing 100,000 EBH. Sheehan proceeded with the presenta­tion’s objective: Identifying the information needed to assess what the 100,000-EBH condi­tion of W501D5 compressor blades in the operating unit investigated would be and to identify some of the potential risks associated with operation beyond 100,000 EBH. His assessment focused on R1 and R19 blades

Engineers found that the micro­structure and hardness of the R1 blade material was comparable with the microstructure of new blade material. However, materials data only tell you that the inherent strength of the material in the specific location where the sample was extracted had not deteriorated. Note that the photo­micrographs and test data compiled for this investigation and the one for the R19 blade are available on the CEP where you can access Shee­han’s presentation.

The photo of the R19 blade that Sheehan put up on the screen revealed oxidation and oxide buildup. He pointed out that even relatively minor buildup can cause very small surface pits where stresses concen­trate and cracking can begin. The photomicrographs and test data showed the material’s microstruc­ture and creep rupture strength were comparable with those found on new blading.

But the service-exposed material was of significantly lower hardness than the new material. Data available on the CEP show a drop in both the percentage shear and absorbed ener­gy. This indicates that the long-term exposure of the blade to high temperatures has raised the ductile-brittle transformation temperature. Based on these data, Shee­han said the risk of a brittle fracture during a cold start is higher on units with more than 100,000 EBH.

Sheehan summarized this way: Operating, main­tenance, and repair histories, as well as any existing surface damage or discontinuities, can mean that the component may not achieve expect­ed life. He believed that the replace­ment of both R1 and R19 compressor blades after 100,000 EBH was a sound recommendation.

Scott Freeman, service engineer­ing manager, tackled the subject of lube-oil cleanliness and filtration. The W501D fleet dates back into the 1980s and lube-oil cleanliness stan­dards and filtration nomen­clature have changed in the last 10-15 years. This opens the possibility that O&M staffs tending some engines in the fleet may be using old terminology, others the new terminology. Tough to com­municate when you’re speak­ing different languages.

Freeman began by saying that lube-oil cleanliness is related to the condition of the fluid. Bearing requirements for lube oil evolved from the previous industry standard of nominal/absolute rat­ing in the mid 1990s to the current international standard described in ISO-4406 which specifies the number of particles of a given size that are allowed in a 1 milliliter sample. Users should consider spending a few min­utes on the web to familiarize them­selves with the ISO terminology.

Filtration is a rating dependent on the filter media. This is quantified against a cleanliness standard. The previous nominal/absolute rating, Free­man said, while widely used, typically was filter-manufacturer dependent. Siemens started using the ISO nomen­clature to communicate both particle size and efficiency in the filter rating in 1999. For example, a 25-micron-rated filter with an efficiency of 99.5% would be written as B25=200.

Lube-oil filters for D5-D5A units were specified with a nominal/abso­lute rating of 15/25 prior to 1999. The supplier later changed nomen­clature to a B25=75 rating. Units in the fleet sold after February 1999 typically have filter elements rated at B25=200.

The need to upgrade to more effi­cient filter elements depends on sev­eral factors, including bearing and oil system condition and oil cleanliness. Siemens’ rec­ommendations include limit­ing the pressure differential across the filter assembly to 10 psi.

George Van Deventer, repair marketing manager, led the valuable and well-received full-day repair technology seminar. Some of what he had to say can be found at www.combinedcyclejournal.com/archives.html, click 1Q/2007, click 501D5/D5A Users on the issue cover. Van Deventer’s course notes are available on the CEP.

Another alternative is to increase shaft speed dramatically—to, say, 160 rpm—where centrifugal force should “lock” blades in place, thereby reducing wear. One user was pursu­ing this alternative thinking that an added benefit could be elimination of spin cooling.

On Siemens Day, mention was made of a new procedure being inves­tigated by the OEM. The plan: Cool the engine appropriately after shut down but do not run the turning gear continuously. Prior to a restart, engage turning gear for an hour, do a 5-min spin cool, coast down to 50 rpm, and hit the “start” button.

Vibration/balancing was anoth­er topic brought to the floor. Here’s where a good moderator shines. Users needing help can’t be offered concrete advice most times because there are no data to analyze. And even if data were available, time is needed to review it. So the moderator must move the discussion to another topic—such as alarms.

A word of advice from the back of the room: Sometimes alarms don’t “ring.” One of his machines was set to alarm at a vibration of 6 mils and trip at 8. There was no alarm on a V-band failure (bolts not tack-welded and backed out) and the turbine was destroyed. A second “no alarm” dam­aged the unit.

A few more discussion topics from the opening session:

  • Lube-oil filters should be specified to capture particles larger than what size? Everyone probably has an answer here, but users were quoting numbers from a low of 3 microns to 30.
  • Air-to-air rotor air coolers gener­ally were criticized as being inef­fective after a few years of service. A couple of users reported “fins coming loose” as the cause of inef­fective heat transfer. Suggestion: Spray water over the bundle.
  • Questionable thermocouple (TC) readings. Check thermocouple wells to assure no foreign material is present and that good contact is being made by the TC. Check to see if leakage from a nearby seal might be skewing readings.

Support systems

Hydraulic and lube oil systems were the focus of discussion as the group returned from the morning coffee break. The type of hydraulic oil used and varnish problems dominated the exchange. Interestingly, some sites have varnish issues, some don’t—even when those that don’t are using the same oil and have the same engine as those that do.

Servo maintenance had a high profile in all this, of course. Servos require periodic overhaul at a mini­mum; one user suggested having a spare set and swapping them out every 18 months. Set removed is sent out for maintenance on a relaxed schedule and returned to the plant storeroom.

Group consensus: Varnish typi­cally is found in cool sections of the hydraulic system, especially where flow rates are low. If this happens to you, try heat tracing to hold the var­nish in solution.

 3. A website for the ages

Gabe Fleck, chairman of the 501D5-D5A Users since the group was formalized in 2001, has been working on the organization’s web­site for a long time. At every annual meeting for the last three years or so, he would update the users on his progress and close with the old Brooklyn Dodger promise, “Wait till next year!”

Well, 2008 was the breakout year, and it would be difficult to find some­one in the group who didn’t believe it was worth the wait. Fleck is a perfec­tionist and the website, www.501D5-D5Ausers.org, certainly reflects his persona. That’s not to say the entire site is built-out at this time—it isn’t. But the work remaining pales in com­parison to that completed.

Standard fare for user-group websites in the gas-turbine-based generation sector of the electric power industry is a chat room with questions/answers archived, general announcements, history, bylaws, names and contact info for steering-committee members, presentations from past meetings, and information/contacts for companies participating in the group’s vendor fair.

The D5-D5A website has all that. In addition, it offers in a secure area, valuable plant and user databases. These help registered group mem­bers access in a matter of minutes both engine data fleet-wide and contact information for colleagues to approach with questions. With­out such electronic files, it probably would take the average user two or three meetings—a year or more—to identify participants with the same equipment they have and to gather the e-mail addresses and phone numbers specific to their needs.

The secure unit and user data­bases section of the website allows members to install and update their personal professional profiles (includ­ing name, title, company, address, phone, fax, e-mail address) and com­pany information (descriptions of all gas-turbine-based generation assets, plant addresses, phone numbers, etc).

Data on the 90 operational D5s and 57 D5As in the global fleet include plant and unit names, engine class, commercial operating date, turbine and unit serial numbers, pri­mary fuel, generator style, starting package supplier, type of combus­tion system, and type of power aug­mentation system.

The Discussion Board for users only is divided into a dozen sections as follows:

  • General: Administrative, long-term contracts, operations, out­ages, parts needed and available.
  • Inlet: Air inlet houses, filters, evap coolers, pulse cleaners, wet com­pression, fogging, chillers.
  • Compressor: Blades, dia­phragms, bearings, wash skids, cleaning solvents/solutions, bleed valves.
  • Combustion section: Fuel noz­zles/support housings, baskets, transitions, V-bands, clamshells, scoop tubes, bypass ring/valves, CDMS.
  • Turbine: Blades, vanes, disccavity cooling valves, orifices, and thermocouples, blade-path TCs, exhaust bearing, exhaust-bearing tunnel.
  • Exhaust: Struts and fairings, dif­fusers, transition, duct, silencers, stack.
  • Balance of plant.
  • Controls: WDPF, TXP, Ovation.
  • Starting package: Motor, torque converter, worm gear, turning gear.
  • Mechanical package: Lube-oil system (pumps, motors, filters, valves, vapor extractor), instru­ment air system, PS&G cabinet, vent fan.
  • CEMS/PEMS: Data acquisition system, data logger, analyzers, compliance testing, RATA testing.
  • Generator: Bearings, jacking oil system, exciter, voltage regulator, cooling-air filtration.
  • End note: 501D5-D5A users who have not registered for group membership should contact Chairman Fleck at gfleck@aeci.org to arrange for website access. There is no charge to join the group nor are there any dues.

A couple of attendees viewed peri­odic replacement of oil as a solu­tion. Intervals suggested: two years, three years, eight years. Others warned against this practice unless the hydraulic system was flushed thoroughly. Reason: Constituents in the replacement oil might react adversely with the residual original oil remaining after the system was drained.

Conditioning systems for hydrau­lic oils, such as filters and the par­ticle agglomeration system made by New England-based ISOPur Fluid Technologies Inc, got their share of the available air time.

Coking of oil in cold climes received a mention, too. It can occur when heaters come on. Evidence: the color of the oil darkens over time.

Lube oil. A user affiliated with a base-load plant said they gravity-drain their lube-oil system every six years (each major) and refill with new oil—because it is thought to be a good idea. Two such cycles have been completed already. Plan is to flush-clean the system before oil is replaced for a third time. A con­taminant spike has been observed on restart with new oil, presumably because contaminants hide out some­where in the system.

This same user rebuilt the lube-oil pump after the second major even though it was working well. OEM suggested it should be done, and was. Another suggestion: Inspect shoes on oil-filter transfer valves regularly; test leaktightness during every out­age. Be prepared to replace shoes triennially.

Endnotes, morning session.

  • There are several engines in the fleet operating beyond 10 years with their original control sys­tems—a WDPF (Westinghouse Digital Processing Family) in most cases. Serious consideration should be given to replacing these systems—even if they are work­ing properly—because the turbine OEM considers them obsolete and parts and service are in phase-out.

Conversation focused on the battle for the hearts and minds of users by Siemens, which wants asset owners to upgrade with SPPA-T3000, and Emerson Pro­cess Management, which wants them to migrate to Ovation®. Recall that Emerson’s Power and Water Solutions began life as Westinghouse Electric Corp’s Process Control Division, which designed the WDPF. Siemens chose not to acquire that division when buying the company’s rotat­ing equipment assets.

  • Fire protection cross-talk was on replacement of Haylon systems because of environmental con­cerns. Users thought CO2 and Dupont’s FM200 (a waterless fire-suppression formulation) were the leading candidates. The water mist system that was presented to the group a couple of years ago has not gained much traction.

A couple of users discussed their efforts in upgrading controls and voting logic to prevent false trips—which occur too frequently according to a show of hands. Causes identified for malfunction include corrosion of components, vibration, sensors in high-heat areas, etc. The need for annual testing was stressed with this warning: Carefully select the ven­dor for this service. Be sure they have powerplant or industrial experience and provide references for you to check.

  • Even bearing vents got air time. One experience was that the No. 2 bearing is prone to coking up. To avoid oil leaks and the possibility of a small fire, clean the vent pipe each outage.
  • Reinsulating the turbine no later than the second major was a recommendation from the floor. Logic: It just falls apart and insur­ance companies are sensitive to the risk bare sections of casing pose for package fire protection. Crossby Dewar Inc was recom­mended by one user.

The other major player in seg­mented, form-fit insulation sys­tems for this sector of the industry is Arnold Insulation USA LLC. Wolfgang Arnold presented on his firm’s product line after lunch. It was the only vendor presenta­tion on the program—other than Siemens, of course. Both Crossby Dewar and Arnold had booths at the vendor fair.

  • Combustion inspection. Last item before lunch was a short user presentation on a recent CI for a base-load unit. After 8000 hours of operation, visual and NDE (nondestructive examination) inspections of standard transition pieces (not extended-life parts) presented good results and the plant was aiming for 16,000 hours. The thermal barrier coating (TBC) was credited for the pristine con­dition. Be sure to inspect coatings for holidays which create places for hot gas to damage the base metal.

4. Eight users recognized for continuity of participation

Chairman Gabe Fleck and the Board of Directors of the 501D5-D5A Users recognized eight members of the group for their valuable participation in five or more annual meetings and for the personal sacrifices made to assure that participation (time away from family, etc). This is the first year participation awards were conferred; the program will be ongoing. Recognition consists of a cer­tificate and a $250 Marriott gift card, good at any Marriott facility at any time.

The 2008 recipients were the following:

  • Craig Davis, Black Hawk Cogen Station, Borger, Tex.
  • Tim German, Elgin Energy Center, Elgin, Ill; Gibson City Power Station, Gibson City, Ill.
  • John Guillory, Taft Cogeneration Plant, Taft, La.
  • Luis Martinez, Enertek Cogen Plant, Altamira, Tamauli­pas, Mexico.
  • Barry Mayhew, Cardinal Power, Cardinal, Ontario, Canada.
  • Jaime Pardo, Termoflores SA Colombia, Barranquilla, Atlantico, Colombia.
  • Doug Prindle, Elgin Energy Center, Elgin, Ill; Gibson City Power Station, Gibson City, Ill.
  • Stan Wilhelm, Nodaway County Power Plant, Maryville, Mo.

Exhaust cylinder cracking was repaired during the outage. A sheet of Hastelloy X (4 ft × 8 ft × 3/16 in.) was purchased for this purpose. Expensive: $8000. For more on exhaust-cylinder repairs see the 501F User Group report elsewhere in this issue.

User experiences

The survivors of the white-water raft­ing adventure made it safely through lunch, listened to Arnold talk about turbine insulation, and settled in for a few war stories from colleagues. Adult beverages and chips would have been a positive addition, but they were lures to the exhibition hall along with lots of good food at carving stations.

The case of the cracked TPs. Here are the clues: D5, steam injec­tion, diffusion flame combustion sys­tem, all TPs show considerable dis­tress (cracks near the exit mouth and stiffeners), only one other repair cycle on this starts-based machine. What’s the cause?

This is why you go to user-group meetings: Sympathetic listeners with experience to tap; plus, there’s no consulting fee. A colleague reaches for the floor microphone. “We had the same problem,” he said. “Found we were firing at peak rather than full load. Once we dropped the tempera­ture by about 60 deg F everything was OK. Siemens came in, checked the firing curves, and adjusted the control system. Siemens had missed this operating anomaly for 10 years.”

Then came a suggestion from Moderator Mayhew: Before you over­fire on a short-term basis to make more power be sure you know its impact on parts life. One user said they disabled peak-firing capability at his plant because the economics just didn’t work out.

The life-extension conundrum. The facts: D5 went commercial in 1982 as part of a combined-cycle sys­tem (one of the first D5s installed). Gas only. Water injection for NOx control. More than 165,000 operating hours. During CI last February, Sie­mens was engaged to evaluate casing metallurgy and to install casing ther­mocouples to collect temperature dis­tribution data. Casing samples were not removed, but some etchings were done with nitric acid and ethanol.

Warning: Don’t attempt the last step without an expert present. Mix­ing both constituents creates a very loud “bang.” Proper handling, com­munication, and disposal are critical to personnel safety.

NDE results were favorable. No creep or decarburization was in evi­dence.

Next step: Rotor condition assess­ment to determine if life extension would be cost-effective. Be aware that a proper condition assessment requires a major plant effort to col­lect appropriate data for meaningful decision-making (Sidebar 5). Infor­mation provided to Siemens by this user included—among many other things—temperature excursions, trip history, and disk-cavity temperature history. If your rotor is approach­ing a milestone anniversary, visit Siemens’ Customer Extranet Portal (CEP) to get the full list.

Next year a “go/no go” rotor life-extension decision will be made by this plant. One of the first steps in the process is a comprehensive rotor inspection. For more infor­mation on rotor inspections, access www.combinedcyclejournal.com/archives.html, click 1Q/2007, click “501D5-D5A Users: Rotor exchange, inspection highlight ‘best ever’ Mid-Year Meeting.”

Is life extension in the cards for this rotor? Will another rotor be withdrawn from the exchange pro­gram? Will a new rotor (or a new gas turbine) be ordered? Come to the 13th Annual Conference next June for an update on a project of great impor­tance to the entire fleet.

Anatomy of a second major inspection. The background: D5A went commercial 12 years ago, about 100,000 fired hours, DLN combustion system, fuel is natural gas and some hydrogen, partial discharge activity was indicative of generator issues. Third-party service providers were engaged for disassembly/reassembly, Class IIB rotor inspection (five weeks vs the OEM’s three months, or more), and generator overhaul.

The total outage ran two months and one week. Here’s a summary of findings:

  • Oil seals very dirty.
  • Grease “cooked” in trunions at the exhaust end of the turbine. A grease groove was added (none existed) to allow the grease to flow.
  • Rotor cooling pipes were cracked; inner pipes had collapsed. Note that the configuration of these air­ways is pipe-within-a-pipe.
  • Torque-tube clearances were with­in original tolerances.
  • Disks and curvic couplings were in good shape.
  • Through-bolts all were straight and most could be reused; nuts were changed, however.
  • Compressor section received a new baked-on coating to increase efficiency by 1% to 1.5% and to provide an extra measure of ero­sion and corrosion protection.
  • Recoated compressor bellmouth with a two-part epoxy system.
  • Turbine rotor was in excellent shape; maximum runout of 1 mil.
  • Westac generator required extensive work, including stator rewind and rotor recondition­ing. Note that this is the third generator coupled to the original turbine.

After the case histories, Modera­tor Mayhew began dissecting the unit to trigger questions or comments that may have been tucked away in the subconscious. Rotor was first: disk failures, through bolts, mar­riage coupling (suggestion here was to replace bolts on older units with IN738 material), belly bands (a user suggested replacing them every time the rotor is removed from the casing), balancing, etc.

Regarding balancing, a user point­ed out that when you have a severe balance problem you can’t get beyond the critical speed at 2600 rpm. Exam­ple: Their unit required replacement of three turbine blades in one row. When the row was reinstalled, the blades were put back out of sequence and the unit had 20 mils of vibration. Couldn’t start it.

The exhaust section was next: casing, expansion joints, tangential strut shields (cracks typical here). Turbine section: R1 blades (mention was made that D5 and D5A blade geometries are identical; can run a D5A in a D5, but not vice versa because D5 blades are uncooled), R2, R3, R4, blade locking hardware (a few users have installed oversize seal pins; one reported checking blade rock on all four rows when they have the rotor out of the unit), bearings (group consensus is that replacement generally is condition-based).

Gremlins. Posco Power Corp, Incheon, South Korea, is an impor­tant participant in meetings of the 501D5-D5A Users: It owns and oper­ates more than 10% of the global D5 fleet. Bo-Hyun Kim updated the group on the company’s four 450-MW 3 × 1 combined-cycle power blocks which start/stop daily. Primary fuel for the 12 D5s is LNG, backup is dis­tillate oil.

5. Early planning recommended for lifetime-extension initiative

Equivalent base-load hours and equivalent starts for a gas turbine are much like the physical ages of power­plant personnel—they keep accumulating, and before you know it, retirement is on the horizon. Big difference, of course, is that you generally can extend the careers of GTs by replacing or refurbishing critical parts.

It’s important to begin planning the lifetime-extension initiative for your gas turbines about three years prior to the OEM’s suggested retirement dates for the units. The first step in the process generally is a thorough engi­neering evaluation of the ageing units. A proper analysis requires, at a minimum, knowledge of the following:

  • Engine component condition (comparison of actual engine degradation to that expected). This step typi­cally requires a battery of materials tests—such as tensile/yield strength, impact toughness, stress rup­ture, microstructure evaluation, and hardness.
  • Individual life-limiting mechanisms—that is, low- and high-cycle fatigue, creep, embrittlement, and surface degradation (wear, erosion, corrosion, oxidation).
  • Interaction between the mechanisms noted in the pre­vious bulleted item.

Here’s a quick refresher on the life-limiting mecha­nisms of interest:

  • Low-cycle fatigue. Cracking by cyclic loading with less than 10,000 to 100,000 cycles. It is caused by relatively large stress/strain amplitudes characteristic of start/stop cycles; surface roughness, such as that caused by oxidation, can reduce cyclic life. Failure proceeds this way: crack initiation at the surface, cyclic crack growth, final fracture. It is not possible to determine the probability of cracking by examining blades.
  • High-cycle fatigue. Rupture by cyclic loading with more than 100,000 cycles. It is caused by relatively small stress/strain amplitudes attributed to in-service stress fluctuations such as those produced by vibration. Failure proceeds as it does for LCF, except that rupture normally occurs quickly when the endurance limit is exceeded because of the high frequency of the loading. Remaining HCF life cannot be predicted from micro­scopic investigations of the material.
  • Creep. Rupture by time-dependent deformation and damage at elevated temperatures. There are three phases of creep: Primary, the shortest phase of the creep process and irrelevant in the prediction of remaining life; secondary, the longest segment of the creep process, is characterized by a constant, or near­ly constant, rate of creep; tertiary, increasing creep rate and damage throughout the volume of material. End of creep life is only predictable once the part gets into tertiary creep.
  • Embrittlement can be induced by contaminants (such as hydrogen, water, tramp elements), stress (work hardening), and/or temperature (phase transforma­tion).
  • Surface degradation includes wear, erosion, impact damage, oxidation, and corrosion any one or more of which may lead to fatigue crack initiation. Further, reduced cross sectional area of the blade is conducive to higher stresses. Quantifying the remaining fatigue life is not possible and lifetime assessment is difficult.

Abnormal vibration was observed on one of the GTs about six months before the unit was scheduled for its first major inspection. Observations included vibration hunting during both cold starts and load changes; phase angle of vibration measured at the turbine exhaust bearing had changed; thermal sensitivity was increasing and stiffness decreasing; vibration amplitude at the exhaust bearing was at nearly 6 mils continu­ously. An unbalance condition was suspected.

Most experienced plant superviso­ry personnel have chased the vibra­tion gremlin at least once in their careers and know how maddeningly difficult it occasionally can be to find its hiding place. Kim certainly would agree.

He approached the problem in a very organized manner, as would any good engineer. First operating data were collected and analyzed, but no cause was apparent. Next, the main­tenance history of the machine was reviewed thoroughly, but, again, the gremlin remained hidden. Kim ran some tests, but still nothing.

A runout check revealed a prob­lem at the marriage coupling. Spe­cifically, the runout value at the cou­pling was about four times greater than the value for the rotor itself. At that point in time, engineers weren’t sure what had happened at the cou­pling. They considered that perhaps the bolts had relaxed.

Although only one month remained until the start of the major, the wors­ening vibration condition dictated that the rotor be swapped out with an onsite spare. During that activity, engineers found that the marriage coupling was damaged. Bearings were repaired, the rotor replaced, and the unit was field-balanced and returned to service.

The rotor removed was sent for a Class IIB inspection (refer to article referenced previously for details), marriage coupling bolts and rotor through bolts were exchanged, the rotor was rebalanced, and then returned to the plant.

Lesson learned: When you’re having trouble finding the vibra­tion gremlin, don’t forget to check runout.

Minimizing mistakes is the goal of every plant manager, one user said when addressing the group, but “talk” does not beget results, hard work does. This participant recently assumed the position of plant manager at a nine-year-old facility that has required seven outages to correct problems cre­ated while making repairs.

Case in point: Two fuel hoses were installed incorrectly following a recent hot-gas-path inspection. Result was $2 million in damage and the need for what amounted to another HGP to make repairs. Both the OEM and plant lose when some­thing like this happens. Who was to blame? The speaker said the person who reattached the hoses incorrectly, obviously. However, he added that everyone connected with the project was responsible for what happened because they did not anticipate that a mistake might happen and work proactively to avoid it.

We’re all human, the plant man­ager continued, we all make mis­takes. It’s up to us to prevent a mis­take from becoming a loss, he added. A plant manager is responsible for everything that happens in his facil­ity. This involves reviewing contrac­tor procedures for equipment disas­sembly, for deciding on what work is required and how it will be done, for inspecting the job while it’s being done and after it is completed, for identifying hold points and who signs off after each critical step, etc.

Managing a power plant is not for everyone.

Siemens presentations

Siemens Energy’s formal participa­tion in user-group meetings typi­cally begins with a presentation that answers the question: “Where are we going?” It reviews the OEM’s investment in fleet solutions, its stra­tegic roadmap for the development of enhanced products and services, the company’s vision, organizational and personnel changes, etc. Har­ald Griem, service product line director, handled this at the 501D5-D5A Annual Meeting.

Customer survey results and the company’s improve­ment efforts traditionally are next on the agenda. This presentation, made by Marketing Manager Dawn McCarter showed customer ratings of nearly four dozen evaluation areas—such as quality of repairs, workmanship, heat rate, engine emissions perfor­mance, parts availability, etc. To its credit, the OEM openly highlights areas that customers perceive need improvement and then outlines how it expects to address the issues, and presents a target timeline for that effort. Progress is updated as part of the following meeting.

This presentation also encouraged customers to access the Customer Extranet Portal (CEP) for the lat­est technical bulletins, shop reports, parts information, etc; and stressed the importance of user participation in the customer surveys.

Jeff Kain, engineering frame owner and manager of service engi­neering for mature frames, presented the fleet statistics—including ser­vice-factor and duty-cycle trends, availability, reliability, starting reli­ability, etc. This information helps guide the development of new prod­ucts and services for the fleet, to keep the performance numbers moving upward.

Kain then reviewed the technical advisories (TAs) issued by the com­pany since January 2007 with a one-slide summary of each. Details are available on the CEP.

R1 ring-segment update. Next, some recommended measures on R1 ring segments were summarized by Kain. His entire presentation can be accessed on the CEP.

Rotor cooling-air piping was another topic covered. The OEM’s planned actions were outlined.

Compressor hook-fit wear. Ser­vice Bulletin 53011 includes recom­mendations on inspecting for wear at each major. It also provides guide­lines on wear limits and repairs.

Kain’s slides have excellent sketches to explain where and how wear has been observed to occur on some units and what it can look like. Portions of the recommended repair process and durability enhancements available from Siemens also were highlighted.

R1 diaphragm update. The development of this enhancement is virtually complete. The new-design R1 diaphragm has been released for fleet implemen­tation. Details of the design upgrade were first published four years ago (access www.combinedcyclejournal.com/archives.html, click Summer 2004, click 501D5-D5A Users on issue cover, scroll to p 83). At the Annual Meeting, Kain told the users that since the prototype was installed in July 2004 all test results are consid­ered by Siemens to be positive.

After testing the prototype, design validation proceeded to the next step: New diaphragms were installed on 10 engines. A comprehensive moni­toring and inspection program—visual, borescope, and tap test—has not identified cracking or other sig­nificant issues in these machines over hundreds of starts and tens of thousands of hours of operation. Monitoring of operating experience is expected to continue for a period of time.

Turning-gear time-reduction update. Kain began this segment of his presentation by reviewing the industry issues associat­ed with turning-gear opera­tion, specifically:

  • Rotors will sag and devel­op runout if maintained in a stationary position. This would result in vibration—possibly vibration severe enough to trip the unit—on startup.
  • High time on turning gear is con­ducive to disk-serration/blade-root wear. Excessive blade rock may result and seal pins may be liber­ated. In the extreme, it can con­tribute to disk cracking.

At the present time, turning-gear operation is suggested 12 hours prior to startup. For units in peaking ser­vice, continuous turning-gear opera­tion is considered necessary by many users to maintain immediate start-readiness.

Siemens is pursuing alternatives to continuous turning-gear opera­tion. Objectives are to (1) maintain start-readiness without continuous operation, (2) require no pre-start measures, and (3) enable automatic turning-gear operation. An interim objective is to reduce pre-start mea­sures for units that have been off turning gear. Success is defined by Siemens to be that no vibration trips occur on startup or spin-hold.

A procedure to reduce pre-start measures is being evaluated and has been demonstrated successfully on eight units. Evaluation will con­tinue on still more units with active involvement by the OEM. Plan is to eventually replace the current guide­lines with the new procedure.

Siemens has developed a method­ology for reduced turning-gear time while maintaining start-readiness and has successfully demonstrated a prototype of this system on four engines. The system currently is being developed into a product with a 2009 target availability date.

The procedure for aligning the GT with its generator was requested by users. Kain gave a short pre­sentation that included a review of why proper alignment is important (maintain the design loads on bear­ings, avoid excessive bending stress­es in the turbine/generator rotor assembly); how alignment gets out of adjustment (foundation settling is one possibility); what issues can be caused by misalignment (generator rotor vibration, for example).

He said that some potential issues can be mitigated by check­ing alignment during major inspections, after a cylin­der change, and whenever the coupling between the GT and generator is broken. This is a practical presenta­tion worth accessing on the CEP. Diagrams and data are valuable at the deck-plates level.

Steam injection. Selvam Veerappan, an applications engineer in gas turbine service engineering, followed Kain with presentations on power augmentation using steam-injection and wet-compression tech­nologies. Veerappan told the editors that his presentations were prompt­ed by user inquiries for information on how to boost power output during periods of peak demand. Properly designed, operated, and maintained power-augmentation systems, he said, may help many users increase power output without some adverse impacts.

Steam injection technology has been used to improve GT performance for 50 years or more, Veerappan told the D5-D5A users. In the mid 1950s, steam was first injected into combus­tors; in the 1960s, into the compres­sor discharge air via special ports; in the 1980s and 1990s, into combustor baskets to control NOx production in oil-fired engines. End notes to this portion of the presentation highlight­ed Siemens’ extensive experience with both steam and water injection for NOx control and power augmen­tation.

Steam for boosting power output is injected into the combustor shell, thereby increasing the mass flow through the turbine. Veerappan said the good operating experience with this system has confirmed its flex­ibility and adaptability for W501D5A users interested in increasing power production.

Then Veerappan dug deeper into the engineering aspects of power augmentation. He reviewed the ther­modynamics of steam injection, com­plete with a temperature/entropy diagram and then offered schematics to illustrate how both simple- and combined-cycle systems would be arranged for steam injection. This material is available to subscribers on the CEP.

Wet compression. The most compelling arguments for wet com­pression: It can restore the drop in GT output attributed to high inlet-air temperature and it has the potential to make a huge step change in power output when needed. In Siemens’ view, wet compression can achieve these objectives in a more cost-effec­tive manner than can competing technologies.

A great deal has been written about the potential issues associ­ated with wet compression. Many problems with non-Siemens systems have been traced to poor atomization/spray pattern and an incompatible inlet-air system. Excessive use—that is, running the system during peri­ods when there’s little or no economic justification for doing so—is another possible issue.

The benefits of this technology are hard to overlook. Veerappan men­tioned a power boost of 10% to 20%, a 1% to 2% heat-rate improvement, and more exhaust energy available for combined-cycle steam produc­tion—among other plusses observed on some units where a Siemens wet compression system is installed. Any user wanting to quantify the poten­tial value of wet compression at his or her plant should access the charts Veerappan presented on the CEP. Also read Service Bulletin SB55007, which presents compressor inspec­tion recommendations.

Scott Freeman, service engi­neering manager, responded to user questions on the topics of flexible tubing at fuel oil nozzles, torque-con­verter clutches, and control method­ologies for inlet guide vanes, exhaust temperature, and fuel-flow control. Readers who might benefit from this information should visit the CEP and access Freeman’s presentation.

Powerplant automation. Con­trol-systems presentations can be snoozers when the speaker digs into the details of logic, cabinet layout, wiring, etc. But not when Ron Hitzel, international I&C sales manager, is in the front of the room. His high energy level and polished delivery keep the audience involved from start to finish.

Hitzel had the task at the Annual Meeting to explain how easy it is for users still running WDPF (for Westinghouse Distribut­ed Processing Family) con­trols to migrate to Siemens’ SPPA (for Siemens Power Plant Automation)-T3000 system.

Hitzel began with boiler plate on the Web-based con­trol platform, how its three-tier architecture mirrors the structure of the Inter­net, and how easy it is to replace WDPF with T3000—so easy, that many plants choose to do their own installation work with technical field assistance representatives, con­tracted from the OEM, onsite.

Features and benefits of the T3000 and early conversion experience can be accessed at www.combinedcy­clejournal.com/archives.html, click 2Q/2006, click “Upgrading controls. . .” on the cover of the issue. Much more detail is available on the CEP, of course.

Hitzel spent a few minutes on two projects: Progress Energy Florida’s Hines (power block 1) and Korea East-West Power Co’s Ilsan com­bined-cycle powerplant. The Hines project was a complete plant migra­tion from WDPF to T3000. The 2 × 1 combined-cycle facility’s dual-fuel 501F GTs, steam turbine, heat-recov­ery steam generators, and balance-of-plant controls all were upgraded.

The 840-MW Ilsan, located near Seoul is powered by six D5s (four dual-fuel machines, two gas-only) arranged in two power blocks—one 4 × 1, one 2 × 1. There only the GTs were upgraded to T3000, a requirement to support the installation of advanced dry low-NOx combustion sys­tems on all engines.

Hitzel closed by saying that, in most cases, Siemens needs six months of lead time and a 10-day outage to go from WDPF to T3000.

Mike Koenig, manager of combustion aero design in the GT engineering group, opened his presentation on combustion sys­tems for the D5 and D5A engines with a review of combustion fun­damentals: theory (including key definitions), emissions (NOx versus CO tradeoffs), and diffusion versus premixed flames. Terse, easy to

understand, and ideal for incorpora­tion into any staff training program; you can pick up the slides from the CEP.

Next, he reviewed the charac­teristics of the various combustion systems that have been used in the D5-D5A fleet over the years. First slide in this series showed the “con­ventional” combustor, which relied on air jets for flame stabi­lization and downstream dilution air.

Several slides were devot­ed to the popular DF-42, which is an upgrade of the conventional combustor. Its attributes include more air in the primary zone, no dilu­tion air, and lower emissions (the “42” means 42 ppm of NOx when burning oil). CFD (computational fluid dynamics) illus­trations helped users understand how air jets are used to stabilize the diffusion flame and how this burner design enables deep penetration of combustion air to cool the core flame.

The DLN combustor for the D5-D5A fleet is of the same basic design as that used for the F and G Frames. Relatively minor changes are required to optimize the system for each frame.

Koenig saved the best for last: the new ULN (ultra low NOx) combus­tion system designed to be capable of maintaining NOx emissions in the single digits. Design work and testing are complete and the ULN is ready for operational validation at a customer site. The system engi­neered for the D5A is based on tech­nology leveraged from F- and H-class engines. Validation already is underway within the F fleet, which should speed up the validation process for the D5A.

Koenig said he expected that the ULN would be com­mercially available to the D5A fleet by year-end. Sales to the F fleet are brisk, he added. Koenig also reviewed the company’s experience firing oil in DLN combustion sys­tems, noting that distillate oil gener­ally is used as a backup fuel for dual-fuel engines.

As for bio-diesel experience, which is of interest to many owner/opera­tors today, Siemens has sold one engine with the ability to burn that fuel; it will begin rig testing later this year.

Product development. Perhaps the ideal way to wrap up a user-group meeting is for the sponsoring OEM to tell attendees what it’s developing to help improve performance, extend inspection intervals, enhance operat­ing flexibility, reduce emissions, and extend the lifetimes of engines in the fleet.

That’s what Paul Wesson, lead engineer, mature frame GT modern­ization product development, did at the Annual Meeting. Many attend­ees had at least heard about at least some of the things Wesson discussed because of the group’s participa­tion in the collaborative roundtable on product development hosted by Siemens known as the “Voice of the User” (Sidebar 1).

He began his presentation with a review of the OEM’s survey of what users perceived as their immedi­ate needs. Sixteen made the “very important” list. Another 13 were rated “important.” Wesson then ran through the Siemens process for product development, annual expen­ditures for R&D, progress in develop­ing modernization products for the D5-D5A fleet, etc. He closed with details on a couple of projects of importance to the group. Users were urged to visit the CEP periodically to chart the company’s progress on development projects of interest. ccj