The 6B gas turbine (GT) is a durable workhorse that’s highly valued by many American refiners and chemical producers. Typically, half of the attendees at a Frame 6 Users Group meeting manage cogeneration plants “inside the fence” at large industrial facilities; most of the remainder serve at electricity-only generating facilities often owned by independent producers.
Versatile and reliable, the engine works well in peaking, intermediate-load, and base-load service as a standalone generating unit or one integrated into a cogeneration or combined-cycle system. GE Energy, Atlanta, says more than 1100 of these nominal 40-MW machines are in service worldwide and collectively have accumulated over 50-million operating hours.
The Frame 6 Users Group, guided by a steering committee of six (Sidebar), meets annually—usually in the first half of June. The organization’s 22nd conference and vendor fair was held at the South Shore Harbour Hotel & Conference Center in League City, Tex, just three months before Hurricane Ike devastated the area.
Format of the organization’s meetings varies little from year to year, which is a characteristic of virtually all user groups serving the GT-based generation sector.
A welcome dinner and reception are held on Monday night (June 9 in 2008); the formal meeting is conducted all day Tuesday and Wednesday, and Thursday morning. The vendor fair is on Tuesday evening. This year, Wednesday morning was reserved for presentations by the OEM’s frame experts.
The technical program begins with the compressor section and moves aft through the machine. Miscellaneous topics—maintenance execution, component repair, I&C, fuel valves, and such—generally are on the Thursday morning agenda. To keep the proceedings moving at a crisp pace, formal presentations by outside experts alternate with user-only discussion sessions. Latter often include brief, informal show ’n tell presentations by plant personnel.
Perhaps there’s no better way to start the meeting than to invite Advanced Turbine Support Inc’s Rod Shidler and Rick Ginder to tell you what they find, and where, when borescoping compressors for this frame. ATS conducts hundreds of GT inspections annually so Shidler and Ginder most probably have seen what you should know about.
About half of what they presented this year was repurposed from their 2007 Frame 6 effort (access www.combinedcyclejournal.com/archives.html, click 3Q/2007, click “Borescope inspection” on the issue cover). Most of the rest was included in 2008 presentations before the CTOTF meeting (p 92) and the 7F Users Group annual conference (report follows this article). If you need more information, return to the Web page noted, click 2Q/2006, click CTOTF on issue cover and scroll to the subhead “GE Roundtable. . . .”
The two sleuths mentioned that they are finding some shims migrating from between stator vanes in 6Bs as they have in Frame 7 engines. One of the solutions discussed in the CTOTF report referenced above is “pinning” of stator vanes and shims using a process developed by Rodger Anderson, manager of gas-turbine technology for DRS-Power Technology Inc, Schenectady, NY.
Sam Moots, a member of the Frame 6 steering committee, and production manager for Colorado Energy Management (CEM), was the first to use Anderson’s solution in a 6B and made a short presentation on the experience.
Moots said that ATS found two protruding shims last year during a routine borescope inspection of CEM’s only Frame 6, located at the company’s Brush generating facility. He addressed the issue during a partial hot-gas-path inspection which began May 27. The engine was returned to service June 7, only a couple of days before the Frame 6 meeting started.
When the outage began, the engine was approaching 73,000 hours of service and 4300 starts. Turbine work conducted included change-out of row 1 (R1) nozzles, replacement of a cracked R2 bucket, and replacement of R2 shroud blocks. Compressor work focused on pinning of shims and selected vanes.
Most of the work was done by plant personnel; new parts required were purchased from third parties. The R1 nozzles that replaced the set in service were spares that had been repaired by Liburdi Turbine Services Inc, Dundas, Ont, Canada. Liburdi also repaired spare combustor liners and fuel nozzles.
Anderson was onsite to provide technical support for the shim/blade pinning activity. This portion of the work began by removing the upper half of the compressor case and setting it on the plant’s shop floor as shown in Fig 1. Hook-fit wear was in evidence on the protruding shims (Fig 2).
Next, shims were removed from between the ring segments that comprise vane rows 1-4. Moots reported that this was accomplished relatively easily with vice-grip type pliers and muscle. These shims were not replaced. All vanes and shims in R5 (an air extraction stage) were pinned in segments of six vanes each to eliminate rocking. Note the pinning procedure in progress in Fig 3.
Holes were drilled in the in the base of each vane to be pinned at a local machine shop, using a fixture provided by DRS (Fig 4). DRS also supplied all shims and pins required. A six-vane segment for R5 is shown in Fig 5.
For rows 6-17, the first four vanes on both sides of the upper and lower casing halves (moving inward from the horizontal joint) were pinned together. Shims installed by the OEM had been positioned between vanes closest to the horizontal joint; new shims replaced them in the same locations and were pinned as part of the four-vane groups (Fig 6).
In the discussion following Moots’ presentation, a user said the OEM told him that shim migration is common and “getting more frequent.” That user peened the shims to eliminate rocking. Another attendee reported shim migration in S5-S9.
Of dings and blending. A user who had just completed an annual inspection before the meeting brought along a half-dozen photos to show the group some of the 19 R1 compressor blades hit by a foreign object that caused a total of 24 dings and tears on the leading edges of those airfoils (Fig 7). All users are encouraged to bring a few photos—nothing “formal”—for show ’n tell. Least costly way to learn is from the experiences of others.
The OEM recommended blending the damaged blades and provided instructions. The field crew initially resisted the owner’s request to conduct the blending operation through the inlet guide vanes (IGVs)—without lifting the compressor case (Fig 8). The owner prevailed based on the success of others with the method requested. Experience indicates that it’s possible to also blend R2 blades using this technique, but it is difficult.
Disabling of the IGVs was a concern. They were “decommissioned” by pulling the actuator pin, locking the vanes in place, and locking out the hydraulic oil system to disable the ratchet.
Important to do a fluorescent penetrant inspection before work is started—this to be sure everything that has to be repaired has been identified and there are no “surprises.”
Questions/comments from the floor included the following:
- When does balancing become an issue? Response: You can blend fairly aggressively without affecting performance.
- Where did the FOD come from? Response: Never determined; possible link to filter-house repairs.
- A user suggested icing as a possible cause. A leaking valve at his plant caused icing that damaged compressor blades.
The open discussion on the air-inlet house and compressor was typical in that it touched on many subjects and most concerns could be addressed adequately by someone in the room. Examples:
- Wastage of galvanized silencer sections in a seaside location; replace with stainless steel components.
- Rusting of the air-inlet house in a desert location attributed to evaporative cooling; just clean up and apply rust-inhibiting paint.
- Increasing pressure drop across inlet chiller coils cleaned periodically with a household pressure washer. One user’s comment: Washer may not be adequate for task and dirt released from the front end of the coil bundle may be accumulating on and plugging heat-transfer surface downstream.
Use of dry ice cleaning suggested with recommendation of not being too aggressive (can damage coils). Another person suggested formulations used by contractors to clean coils in large commercial systems. Warning there was to check label to see if the product contained sodium hydroxide, which can be harmful to aluminum fins.
- The subject of fogging was introduced and proper nozzle location/orientation discussed. Not much interest to this group because many plants represented are on the Gulf Coast where benefits of fogging generally are not compelling given the high humidity.
- Woven debris screens fabricated of stainless-steel wire were discussed next. Fretting at contact points can release small pieces of wire into the air stream. Frequent inspections were recommended; replace when necessary.
- IGV upgrades and maintenance generated plenty of chatter, including bushing wear. One user was concerned that in going from base-load service to cycling he might experience a significant increase in wear and tear.
- A question was raised regarding destacking of the compressor to correct a balance issue. It came from a user who destacked when his machine required reblading; found rust, which could cause dynamic imbalance if it accumulated. Another user said he was chasing unbalance gremlins and destacked; little to no benefit.
- Compressor bleed-valve discussion focused on hang-up of internal piston, use of dual solenoid for redundancy, maintenance, etc.
Think for a moment about the value proposition of attendance at the user-group meeting supporting your GTs. You can come armed with questions and get guidance from colleagues at no cost; you have the opportunity to develop relationships with people whom you can call when a problem pops up unannounced; you learn about issues you never knew existed and how to avoid them; you get to “kick the tires” on new products and services at the vendor fair, etc. You can’t afford not to attend.
Co-chair Jeff Gillis opened this segment of the meeting by asking a couple of questions:
- How many attendees had DLN (dry, low NOx) combustion systems, how many did not? By show of hands, it looked 50/50.
- How many attendees have units with over 150,000 fired hours? More hands were raised than you might have imagined.
Reason for the second question was obvious: This ageing fleet with so many base-load units has several engines at or near the 200,000-hr mark established by the OEM for a rotor lifetime-extension evaluation. If your rotor passes examination (most probably after spending bucketfuls of cash for the inspection work and necessary refurbished/replacement components that likely will be required), the engine can be cleared to operate for up to another 50,000 EOH (equivalent operating hours).
When you bring up the “lifetime” subject, the hair typically bristles on the necks of the users. Most are convinced the 200,000-hr inspection trigger is arbitrary at best and they know it means a two- to three-month outage and sticker shock. It may be of little comfort to Frame 6 users but they’re not alone in the rotor-life debate. Lifetime limits have been established for all frames. More detail is available in the report on the spring meeting of the CTOTF’s Legacy Roundtable which begins on p 102.
There was considerable discussion about a controls upgrade from Mark IV to Mark VI to support a combustion system upgrade to DLN1+. Tuning was a major challenge; retuning for wintertime and summertime operation typically is required. Fuel-gas preheat to 110F-120F contributed to more stable operation.
Gas quality was another issue discussed. Lube oil from fuel-gas compressors in the natural gas continues to be a problem. Treatment between the plant fence and GT is necessary to assure reliable operation, particularly for units with lean, premix combustion systems. Treatment skids typically have a knockout tank, heater, and coalescing filter/separator. You’ll need expert help on this assignment.
The discussion just seemed to roll along, briefly touching on many important topics: combustor dynamics monitoring, importance of proper fit-up between transition piece (TP) and first-stage nozzle, floating seals, TP cracking and picture-frame distortion, CO excursions, etc.
Finally, just as many were beginning to feel a bit overwhelmed with all the information coming at them, Conference Coordinator, Wickey Elmo of Goose Creek Systems Inc, rang the lunch bell. Those who hadn’t had to deal with many of the issues discussed breathed a sigh of relief. They had action items that could be incorporated into a proactive plan to avoid some problems tormenting colleagues.
Some great raffle gifts got everyone back in the room, on time, after the lunch break. A good lead-in to the open discussion on the turbine section was the presentation on “Metallurgical Evaluation of Frame 6 Components” by Hans van Esch of Houston-based TEServices Inc. What attendees should have come away with was a healthy respect for metallurgists and why you want advice from the best one available before sending out your hot-section parts for repair and recoating.
van Esch’s practical lecture was a one-hour summary of his popular three-day course that someone from your plant should attend. It provides the background necessary to communicate with repair shops and pointers on what to ask for in a repair spec, plus what to look for when doing shop/job inspections.
To dig deeper, access van Esch’s four-part series outlining the six critical steps to successful refurbishment of GT parts at www.combinedcyclejournal.com/archives.html.
Click 2Q/2005 for the first two steps, onsite assessment of component condition and the development of repair specifications; 3Q/2005 for Step 3, guideline for selecting the appropriate repair vendors to meet your plant’s specific needs; 4Q/2005 for Step 4, the vendor verification process for incoming inspection; and 1Q/2006 for the last two steps, vendor verification during repairs and recoating, and final inspection of refurbished components.
The turbine-section open discussion included one user’s experience with component lifetimes. Some snippets:
- New first-stage nozzles will last 24,000 hr, sometimes longer. Lost the thermal barrier coating (TBC) but didn’t understand why at the time of the meeting; base metal looked good, however.
- First-stage buckets: 16-hole buckets run 48,000 hr and are scrapped, but only 24,000 hr when steam injection is used. Ran one set of 12-hole buckets at 2050F for 52,000 hr.
- Second-stage nozzles: Refurbish every 24,000 hr, expect to get 100,000 hr. Have spare set in during refurbishment of other set.
- Third-stage buckets: Approach in 48,000-hr increments; some are over 100,000 hr, and counting.
- Abradable coating is used on first-stage shroud blocks to reduce clearance. Coating does not normally last a full cycle. When recoated, there’s an immediate 2-MW bump in output.
- Users report difficulty replacing wheel-space thermocouples and in getting accurate readings long term.
Bob Steel, director of engineering for Strategic Power Systems Inc, Charlotte, closed out the first day with his presentation “Plant Operating Data: Automating Data Capture and Understanding its Business Impacts.” A good summary can be accessed at www.combinedcyclejournal.com/archives.html, click 3Q/2007, click “Optimizing the flow, analysis of plant operating data. . . .” on issue cover.
The OEM brought in a team of experts, led by Frederick Delaval, the 6B product manager, to conduct formal presentations and answer user questions. Delaval is based at the company’s factory in Belfort, France.
Rob Berry kicked off the four-hour session Wednesday morning with a commercial on the company’s capabilities, facilities, personnel, etc. Coverage included nozzle repairs, an advanced coating removal process, laser welding techniques, adaptive machine, advanced coatings, etc.
Tim Lloyd provided a valuable review of Technical Information Letters (TILs) pertinent to the 6B. These sometimes get overlooked in staff-challenged plants. Lloyd brought the group up to date on whom to contact if TILs are not being received. It’s easy to sign up online. In case you weren’t at the meeting, here are the documents Lloyd thought particularly important to 6B owner/operators:
- 1067-R2, Bucket tip shroud deflection. Check during every hot-gas-path (HGP) inspection.
- 1157-R1, GT fuel specifications. Second- and third-stage buckets and nozzles are most susceptible to corrosion initiated by fuel contamination.
- 1382-R1, Compressor R1 blade inspection.
- 1420-R1, Lube-oil control logic enhancement.
- 1454-2R3, Chisel-staked stator.
- 1571-R1, Inlet filter purchasing requirements.
- 1577, Precautions for filter-house ladder hatches. This is a safety issue, personnel must close trap door.
- 1579-R1, Turbine-compartment water systems maintenance.
- 1585, Proper use and care of flexible metal hoses. Avoids failures caused by improper handling and installation.
- 1595, DLN1 purge-valve shutdown controls. Owner/operators should verify proper software in Mark VI control systems.
- 1607, Bently Nevada equipment inspection/maintenance.
Kevin Spengler, the technology leader for the group, did most of the heavy lifting at the podium. One area of coverage was wheel-space temperatures, which are key indicators of turbine health. Elevated temperatures accelerate rotor life expenditure. Erroneous measurements most often result from improper installation of thermocouples or defective T/Cs. Spengler conducted a short clinic on proper installation, how to verify the T/C is correct for the given location, how to verify the T/C is positioned properly in the thermowell, etc.
When erroneous measurements can’t be traced to installation issues, sleuthing begins. Check tuning pins to ensure they match unit configuration, verify cooling circuits are not blocked by debris, etc. You may have to reduce load to stay below alarm levels until experts can visit the site and do a more exhaustive investigation.
Spengler then moved to the thorny issue of TIL 1576 and the requirement for rotor life assessment at 200,000 EOH noted earlier. GE service center inspection procedures were included in this part of the presentation. At the present time, he said, there haven’t been a sufficient number of inspections completed to gauge component fallout. A few important points made by Spengler:
- Current guesstimate of inspection time is about 10 weeks when the shop is at full capacity; could be less if the shop is not at capacity. However, if any repairs are necessary, in-shop time would likely increase.
- Allow about 14 months of lead time for a rotor inspection. One reason: Operating data must be obtained from the plant and analyzed.
- Current thinking is that starts-based units are more susceptible to crack initiation in critical components.
- Develop an asset plan to mitigate downtime risks should inspection reveal non-repairable indications. Work thus far indicates turbine-wheel rabbets and dovetail wear are repairable.
Spengler closed this portion of the program by saying that rotor life assessment would be covered at each meeting to keep owner/operators current on the state of inspection technology, trends in inspection results, total outage time, etc.
Thomas Bouvay and Eric Smith followed with the details on a range of mods and uprates that included a pitch for a flange-to-flange replacement as an alternative to rotor life extension. Some of the plusses: Increase in output of 14.5%, reduction in heat rate of 4.5%, NOx emissions with the DLN1+ combustion system of 5 ppm or less, compliant with the latest codes and standards, minimal plant disruption, save $3.5 million in avoided maintenance costs that would be required to operate an aged machine. The last number includes $2.6 million for a new rotor assuming you fail inspection or the unit has accumulated more than 5000 equivalent starts.
David Clarida, the CHROEM™ product line leader, returned to the podium this year with much the same material on inlet and exhaust system replacement that he covered at the 2007 meeting (access a summary at www.combinedcyclejournal.com/archives.html, click 3Q/2007, click “How to replace an exhaust plenum” on the issue cover.
The Wednesday afternoon session began with a user presenting on a torque converter failure. Problems with torque converters are relatively common at GT-based powerplants and sometimes difficult to correct. Reasons include: Torque converters are an oft-forgotten item in the plant maintenance plan, they work basically like an automatic transmission and have lots of parts, plant personnel generally are unfamiliar with them, OEM service personnel sometimes lack the expertise to do the required job correctly the first time, etc.
For the case history presented, the torque converter failed to turn the turbine rotor for crank cooling after a planned maintenance shutdown. The unit had been in continuous operation for 18 months. Plant personnel found a block valve in the priming line to the torque converter oil pump closed.
After re-establishing prime to the oil-pump suction line, the torque converter was able to turn the turbine. However, it was making a grinding sound and was leaking a significant amount of oil at the diesel mounting flange. The unit was shut down to investigate.
Good news: The problem was traced to a ball-bearing failure (Fig 9); there was essentially no collateral damage. The torque converter was cleaned up, reassembled, hoses replaced. The speaker showed about two-dozen slides documenting key steps in disassembly, inspection, and reassembly. Several in the group were seeing the inside of a torque converter for the first time; for most of the others, the presentation was a good refresher. When the first presentation after lunch gets high marks from attendees, you know it was top-notch.
Paul Heikkinen of Wood Group Generator Service Inc, Farmington, NM, closed out the second day of the meeting with “Frame 6 Generator Rotor Rewind Practices.” It closely paralleled an article in 3Q/2006 (“Key steps in inspecting, reconditioning generator rotors”) based on an interview with Heikkinen at the Wood Group shop. Access via Web page noted earlier.
The final formal presentation on the program was made by Lawrence Mitter, VP of engineering support services for Young & Franklin Inc, on Thursday morning. “Combined Stop/Ratio and Control Gas Valve for Frame 6 Applications” began with a view of Y&F’s shop in the 1930s and walked attendees through GE frame gas-fuel control methodology from the 1950s to today. Why start in the 1950s? Interestingly, some of that equipment still is operating (see “These baby boomers also deferring retirement” elsewhere in this issue).
The value of this presentation was that Mitter dissected fuel control valves, explaining what all the parts were for, trim details, packing details, etc. It was a productive review. Maintenance tips and recommended spare parts closed out the presentation. ccj