At annual meetings of the Western Turbine Users Inc (WTUI), the technical sessions begin on Monday afternoon. Over the next two days, users participate in the breakout session focusing on their engine of interest—LM2500, LM5000, LM6000, or LMS100. In sum, about nine hours of presentations and discussions in each track dig into the nitty gritty of gas turbine O&M. No better place for owner/operators of LM aeros to learn; if you don’t understand something, there’s always a colleague or supplier willing to help.

Hands down, the most popular of the four breakouts at WTUI is the LM6000, chaired by Andrew Gundershaug, plant manager, Calpine Solano Peakers. More than 200 owner/operators participated in these sessions during the organization’s 26^th annual meeting in Palm Springs, Calif, March 20-23, 2016.

One reason for this breakout’s popularity: There are more LM6000s in land-based electric generation service than the other three engine models combined. Units installed, under construction, and on order total a nominal 1200—about 30% equipped with DLE combustion systems. LM6000s serve in simple-cycle, cogeneration, and combined-cycle plants.

Roughly three hours is set aside during each breakout to review depot findings for the year prior to the meeting. Information gathering and presentation are collaborative efforts among key personnel at the participating Level 4 authorized depots—for the LM6000: IHI Corp (Ken Ueda), MTU Maintenance (Jens Arend), and TransCanada Turbines (Steve Willard).

The term “authorized” means qualified by the OEM and approved for access to GE technical documents, parts and service support, and the approved-vendor list for component repairs, etc. For more detail, consult service letter LM6000-15-002.

Significant to note, given the dog-eat-dog world of gas-turbine services, is that although the depots compete commercially they come together as an effective technology team to support owner/operators. One outcome of this effort is a nominal 100-page, full-color, printed summary of the depot findings containing scores of photos. In the editors’ opinion, there is no better training program available for O&M personnel. To obtain a copy of this publication you must attend the WTUI meeting and be an LM6000 user. Depot findings for the other engines also are published and made available to user participants in those breakout sessions.

Mark your calendar: The 2017 WTUI meeting will be held at the South Point Hotel in Las Vegas, Nev, March 19-22.

Depot findings in 2015 with the greatest cost impacts were selected by the editors for coverage here:

• Cracking of LPT first-stage blades.

• IGB gear-shaft pitting.

• HPT N1 leaf seals, SB-306.

• HPC S3 to S9 spool damage.

• HPC blades.

• DLE combustor change.

• Peak versus baseload operation.

• LP Sprint™ damage.

Cracking of LPT first-stage blades. Multiple cracks on the trailing edges (TE, see sidebar) of low-pressure-turbine first-stage blades have been found during HGP inspections (25,000 hours) and major overhauls, and occasionally during borescope inspections. The cracks, while relatively slow to propagate, should be reported to your CSM immediately. It was said that approximately three-quarters of the LPT S1 blades scrapped during shop visits are trashed because of such cracking. TE thickness and coating quality of the damaged blades were confirmed within the OEM’s limits.

According to one participating depot representative, there isn’t much users can do to mitigate cracking other than to operate the engine within OEM guidelines and simply deal with any wear and tear identified.

A user asked if cracking has been found on both peaking and baseload machines. The answer: Yes, but it’s more prevalent on peakers. However, no statistics were available. Cyclic operation also was said to impact crack propagation. When a crack gets to a certain point, the group was told, there is no choice but to stop operating and to repair/replace blades as required.

A new LPT S1 blade with a material change reportedly is under engineering review.

IGB gear-shaft pitting. Pitting/scoring outside of the OEM’s limits was observed on the gear teeth of the horizontal gear shaft for an inlet gearbox after about 25,000 hours of service. Experts were concerned that cracks could initiate at pits. Direct cause of the damage could not be determined, although the following possibilities were considered:

• Fatigue damage from repeated force/pressure.

• Contaminated oil.

• Dry condition (lack of oil).

• High moisture.

Primary recommendation: Keep turbine lube oil clean! Also, hand turn the engine before a start after long storage.

HPT N1 leaf seals, SB-306. A borescope inspection of the first-stage nozzle row for the HP turbine in a DLE unit, accessed via the combustion liner, identified burn-through conditions on the leading edges of some nozzles, making an outage necessary. Investigators found the burning associated with the leakage of cooling air, caused by the deflection of inner and outer leaf seals and the loss of some portions of the seals. The coil spring holding the leaf seals in place was ineffective.

Damage to the combustor also may occur with ineffective leaf seals. This was the focus of the HPT N1 discussion at the 2015 meeting in Long Beach.

Recommended action was inspection of the combustion chamber to verify proper seating of the leaf seals. If overheating or burning is apparent, inspect the leaf seals. Any rework required should be done during an engine overhaul, and the recommendations made in service bulletin LM6000-IND-306 implemented.

SB-306 suggests the addition of brazed doublers to the inner and outer seals as well as replacement of the Rene 41 coil spring with one made of Inconel X-750. Manufacturing changes include a different nozzle casting and addition of a lug on the inner band to limit leaf-seal deflection.

HPC S3 to S9 spool damage. Inspection of the Stage 3-to-Stage 9 spool on an engine with more than 50,000 hours of service and 400 starts revealed damage on the upper surface of third-stage dovetail slot rails. The IRM does not address repair limits for damage found in this area, so the spool was removed from service.

During disassembly at the depot, damaged balance weights were found. Investigators believe fragments of broken weights caught between the lower blade platform and upper dovetail slot rail surfaces caused the damage. The depot set aside balance weights with the same part number until a proper engineering review could be done.

HPC blades. Discussion revolved around SB LM6000-IND-310, “HPC Rotor Stage 3 – 5 Blades Dovetail Coating Refurbishment,” released Feb 26, 2016—just prior to the WTUI meeting. This portion of the presentation began with a case history carried over from the 2015 meeting offering the following “observed condition”: HPC Stage 3 blade lift on a unit with about 12,000 hours of service and 6000 fired starts.

A borescope inspection identified both lifting of several S3 blades bearing the “K” designation on their respective platforms as well as blade movement when the rotor was turned over by hand. Note that the latest production blades (2015), distinguished by an “H” on their platforms, are recommended for peaking and load-following service. At last year’s meeting the message was that “K” blades were fine in baseload units and can still be used in peakers if coating repairs are made within a normal interval.

Back to the case history: The top case was lifted for closer inspection and removal of the S3 blades. Heavy wear was noted along the dovetail coating faces, with mating wear in the dovetail slots of the S3-S9 spool. Most of the airfoils removed were scrapped because wear extended into the parent metal; the spool also was removed and declared “unrepairable” based on today’s experience. This finding was of considerable interest because of the possibility that one or more blades could fail in service if the wear issue was not addressed.

A user reminded the group in Palm Springs that at the previous meeting in Long Beach, GE said it was planning to publish a SB based on starts. The discussion leader acknowledged that and alerted to all in attendance that the SB-310 released three weeks before the 2016 conference was that document (thorough reading suggested) and it recommended coating refurbishment after every 1500 starts.

Also reported: The OEM is working on an improved blade, one made of Inconel 718 instead of the current titanium and having a larger dovetail to improve stress distribution. Airfoil geometry of the improved blade—planned commercial release is 2017 at the earliest—will remain unchanged. Benefits of Inconel 718 include increased fracture toughness, a reduction in fretting wear, and no need for a wear coating.

It didn’t take long for an attendee to recognize that a new spool would be required to accommodate the new dovetail configuration and ask: Do we keep reblading our existing spool until that time? “Yes” was the reply from the podium. Another user said he has experienced four of the failures described in the last 10 years. The floor leader understood his colleague’s frustration and told the group he was not aware of any limits on allowable coating loss. “It’s a judgement call,” he said.

Discussion continued, prompted by another question: “For fatigue failure, is it better to use a new blade rather than a refurbished one?” A depot representative said generally there is no reason not to use a refurbished blade provided it has gone through the refurbishment process, which includes a health assessment and recoating.

DLE combustor change. DLE combustors come in two styles, the group was told: Those with brazed heat shields, those with threaded. The discussion leader reminded attendees that brazed heat shields are non-repairable because they must be machined out of their locations. The benefit of threaded heat shields is that they would be repairable, to reduce costs.

There was talk that at some point brazed heat shields may no longer be available. However, it was reported that the OEM has not yet confirmed the future will be threaded heat shields only. Nor has a repair process for threaded heat shields been finalized; field trials continue with positive results. Plus, users should not expect there would be a 100% yield on threaded heat shields sent for repair.

Attendees were made aware that a switch from brazed to threaded heat shields would require an upgrade to the combustor dome. Also, that the OEM would not recommend field replacement of threaded heat shields because of possible challenges associated with properly removing the locking pin.

A question from users during this portion of the program concerned HPC S11 vanes, which are addressed by service bulletin LM6000-IND-315, “S11 Compressor Stator Vanes Part Number Identification and Replacement.” The attendee asked, “Can these be changed out during a site outage.” The reply: “Yes, if you’re in there for something.”

Peak versus baseload operation. An open discussion on this subject exposes attendees to a wide range of valuable O&M experience. Definition of terms was the starting point both at the 2015 and 2016 meetings:

• Baseload means continuous operation 24/7 for power production, except for inspections at intervals of 4000 hours/450 starts, or annually, whichever comes first. It’s important, the presenter said, not to expect that the engine always will be serviceable at the next inspection. It is possible unit condition will progress in 4000 hours to be unserviceable, but unlikely the progression will cause an unscheduled event within the inspection period.

• Peak load refers to engines started whenever power is needed and shut down after the peak is over. Several starts per day are possible. It was said that peaking operation of a ground engine is harder on the machine than commercial aircraft service. The reason: The aircraft engine is at peak during takeoff and at 40% load the rest of the flight.

During fast starts and stops, the presenter said, virtually all engine parts can be affected by thermal distress and material fatigue. Oil leaks are common in peaking service because of the disproportional pressure ratio between seal air and sump air by fluctuating loads or long-term low-power operation. Small leaks are conducive to build-up of oil deposits, possibly causing bearing damage. Serrations on rotating seals and honeycombs of stationary seals suffer abnormal wear.

A series of photos illustrating the types of damage identified with peaking services included the following:

• Trailing-edge cracks on HPT second-stage nozzles.

• HPT shroud and blade-tip rubbing.

• Worn rotating air-seal serrations on the HPC rear shaft.

• Tip rubbing on HPC stator vanes and lands, leading to an increase in clearances.

• Pressurization of the oil sump, contributing to leakage.

The takeaway from this portion of the program was that peak-load operation can be more expensive than baseload operation. Reasons include the following: long-term operation as a peaker increases the wear of engine components, increases heat rate (decrease in efficiency), adversely impacts performance, and raises the cost of maintenance.

A user challenged the OEM, pointing out that “we’re in the digital age. Perhaps this could be used to determine the true impact of starts.” GE’s response was that it has done a preliminary analysis using ORAP® data and was expecting to do more work with Strategic Power Systems Inc on this.

A depot representative said you can’t always know the exact impact because it’s very site specific and operations specific. Example: When you operate with NOx water injection you take a hit; same when you run on Sprint™. The bottom line: Owner/operators should not expect to see the maintenance intervals of 25,000/50,000 hours often sited in promotional literature. An attendee contributed to the conversation by saying that with 16 years of experience behind him, he recommends planning based on 16,000-hr intervals.

LP Sprint™ damage. The subject of Sprint damage was revisited later in the program. The group was shown photos of abnormally heavy erosion on the leading edges of VIGVs and S0 blades, and on the rub lands of LPC stator cases. Operators were urged to monitor their water flow rates, keeping them within the OEM’s recommendations and using even less than that amount of water where possible. Maintaining spray water quality with spec also was stressed. Regular inspection of spray nozzles was stressed, with repair or replacement recommended when tips show signs of wear/erosion.