In closed-session presentations to owner/operators of LM6000 and LM2500 engines at the 27th Annual Conference & Exhibition of the Western Turbine Users Inc (Las Vegas, March 19 – 22, 2017), engineers from Charlotte-based Strategic Power Systems Inc (SPS) reviewed last year’s fleet performance (November 2015 – October 2016) based on ORAP® data. They then identified the top contributors to forced-outage incidents for 2016.
The Operational Reliability Analysis Program developed by SPS provides the company’s engineers and analysts the data and tools to develop and track performance metrics for aeroderivative and frame gas turbines across all OEMs and a wide range of global owner/operators.
LM6000
The SPS presentation began with a statistical analysis of fleet performance for simple-cycle units based on duty cycle: peak, cycling, and base load. The data presented below for 2016 were compiled from information submitted to ORAP by users and represent about one-quarter of the LM6000 fleet of 1250 units, in round numbers.
For peaking machines, defined as those engines having a service factor of less than 10%:
-
-
-
Forced outage factor (percentage of time the unit is in a forced outage), 2.9%.
-
Maintenance outage factor (percentage of time the unit is in a maintenance outage), 1.0%.
-
Planned outage factor (percentage of time the unit is in a planned outage), 4.7%.
-
Service factor (percentage of time a given unit is producing power at any level), 9%.
-
Availability (percentage of time the equipment is capable of operating), 91.4%.
-
Service hours per start, 5.5.
-
-
For cycling units, defined as those engines having a service factor of from 10% to less than 50%:
-
-
-
Forced outage factor, 5.7%.
-
Maintenance outage factor, 0.9%.
-
Planned outage factor, 1.7%.
-
Service factor, 32.7%.
-
Availability, 91.8%.
-
Service hours per start, 17.4.
-
-
For engines in baseload operation (service factor of 50% or more):
-
-
-
Forced outage factor, 1.9%.
-
Maintenance outage factor, 0.7%.
-
Planned outage factor, 3.7%.
-
Service factor, 81.2%.
-
Availability, 93.7%.
-
Service hours per start, 102.8.
-
-
Analysis of forced-outage incidents, hours
Engine performance review complete, the SPS speakers identified the leading causes of forced outages, reporting both numbers of occurrences and their duration in hours. This is a valuable service to the user community because it points to areas owner/operators should focus on during engine inspections, as well as to engine spares they should have on hand at their plants.
Top contributors to forced-outage incidents in 2016 were split this way: aero engine, 24%; package, 32%; electric generator, 9%; balance-of-plant (BOP) station equipment, 35%. Combustion issues were at the top of the list, with 25 sites reporting them. Flameouts, emissions exceedances, and high acoustics were among the leading causes. Gas fuel control/metering valve problems were cited by 28 plants with driver faults/failures causing more than 600 hours of downtime. Control-card failures occurred at 14 sites, causing the third-most number of forced outages; however, they were of relatively short duration, averaging less than six hours per incident.
Flame-detector problems were reported by 18 sites, communications issues at 19 sites, spurious engine trips (causes all unidentified) at five locations. Compressor variable bleed valves were fingered in 22 incidents, with nearly half caused by actuator issues. Software/logic issues totaled 19 at eight sites—including unit stuck in warm-up, corrupt PLC, servo suicide, etc.
Eighteen incidents were reported against gas detectors at 11 sites with nearly three-quarters of the outages attributed to false indications. Tenth on the list of forced outages concerned gas fuel staging valves—17 incidents at five sites.
Top contributors to forced-outage hours in 2016 were split this way: aero engine, 31%; package, 9%; electric generator, 11%; BOP station equipment, 49%. Rotating air/oil seal issues in the high-pressure compressors (HPC) at two sites finished No. 1 with nearly 2800 forced-outage hours. However, this was by choice of the owner/operators who opted not to install lease engines when their units were sent to the shop for repair.
In the No. 2 spot was a single incident of HPC stator-vane cracking that kept the host engine on the sidelines for more than 2150 hours. In that case, also, the owner/operator opted not to install a lease engine. Same was true for a gearbox failure and a lube-oil drain piping leak that each rang up more than 1700 forced-outage hours. If you have no plans to run the engine during the time it would be out of service, install of a lease engine would benefit only the lessor.
No. 5 on the list of contributors to forced-outage hours was HPC vibration/stall at six sites. Two locations, which could have installed lease engines, accounted for 99% of the outage hours. For the engine suffering stall, blade/vane damage was found in compressor stages 8 – 14; vibration on the other unit was traced to damage in the HPC’s sixth stage and the first stage of the low-pressure turbine (LPT).
The failure of a LPC shaft knocked an LM6000 out of service, adding 1200 forced-outage hours to the fleet total; the penalty could have been reduced dramatically with a lease engine. Damage to HP turbine nozzle segments at another site contributed another 968 forced-outage hours. A cracked oil band for the inlet gearbox rang up 689 more hours.
Interestingly, eight of the 10 top contributors to fleet forced-outage hours could have installed lease engines to reduce downtime. Had they done that, total fleet forced-outage hours would have been reduced by at least 75% in 2016.
LM2500
The SPS presentation began with a statistical analysis of fleet performance for simple-cycle units based on duty cycle: peak, cycling, and base load, as it did for the LM6000 fleet above. The data presented below for 2016 were compiled from information submitted to ORAP by users.
For peaking machines, defined as those engines having a service factor of less than 10%:
-
-
-
Forced outage factor (percentage of time the unit is in a forced outage), 4.3%.
-
Maintenance outage factor (percentage of time the unit is in a maintenance outage), 1.7%.
-
Planned outage factor (percentage of time the unit is in a planned outage), 1.2%.
-
Service factor (percentage of time a given unit is producing power at any level), 7.4%.
-
Availability (percentage of time the equipment is capable of operating), 92.7%.
-
Service hours per start, 15.6.
-
-
For cycling units, defined as those engines having a service factor of from 10% to less than 50%:
-
-
-
Forced outage factor, 2.8%.
-
Maintenance outage factor, 0.2%.
-
Planned outage factor, 4.3%.
-
Service factor, 28.8%.
-
Availability, 92.7%.
-
Service hours per start, 60.1.
-
-
For engines in baseload operation (service factor of 50% or more):
-
-
-
Forced outage factor, 2.1%.
-
Maintenance outage factor, 0.6%.
-
Planned outage factor, 2.9%.
-
Service factor, 90.2%.
-
Availability, 94.4%.
-
Service hours per start, 379.
-
-
Analysis of forced-outage incidents, hours
Next, the SPS speakers identified the leading causes of forced outages, reporting both numbers of occurrences and their duration in hours.
Top contributors to forced-outage incidents in 2016 were split this way: aero engine, 22%; power turbine, 35%; package, 37%; electric generator, 3%; BOP station equipment, 3%. Controls, including controllers and software, were at the top of the list, with users reporting 55 incidents at 19 sites. Last year (2015) this category ranked second with 61 incidents. Causes of the forced outages were wide-ranging: analog hardware failure, erratic communication, false indication of over-speed, software failure, spurious signals, maintenance errors, etc.
Lubrication issues placed second with 16 sites reporting 34 incidents—half having to do with leaking piping, seven more with instrumentation. Problems with gas fuel control and regulating valves were reported by nine sites, with four experiencing 19 malfunctions of shut-off valves attributed to closed position and feedback error. Combustion systems accounted for 13 incidents at eight sites with flame-detector and flameout/failed-to-fire issues the major causes.
Fifth among the top contributors to forced-outage incidents, power-turbine wiring and instrumentation, related primarily to issues with speed and wheel-space temperature sensors. One site took the sixth position all by itself, reporting 11 incidents with package liquid-fuel instrumentation, all of which resulted in failure-to-start events. Package fire protection at seven sites contributed a dozen incidents, all related to instrumentation malfunction (gas detector, fire detection, limit switch).
HPC pressure, temperature, and speed sensor issues at seven sites (10 incidents) were next on the list, followed by nine starting-system incidents at five sites and problems with the water injection system at one site.
Top contributors to forced-outage hours in 2016 were split this way: aero engine, 36%; power turbine, 17%; package, 43%; electric generator, 3%; BOP station equipment, 1%. Four sites each reported trip events traced to inadequate equipment specifications for the auxiliary power supply system. Total forced-outage hours: 4696.
The second-most number of forced-outage hours recorded last year by the ORAP system for the LM2500 fleet, 3807, was recorded by a unit that experienced an inlet gearbox spline failure. Forced-outage hours attributed to gas fuel control and regulating valves, mentioned above in the section ranking issues by number of incidents, came next at 848 hours. Right behind it was a site recording an 837-hr outage attributed to failure of HPT bearings.
While controls, including controllers and software, led in number of incidents, it placed fifth in forced-outage hours because the longest outages lasted an average of only two days. Lubrication systems contributed 670 outage hours, most relatively short-lived to fix leaks. One site reported five incidents totaling 475 outage hours because of problems created by out-of-spec distillate fuel. Generator exciter issues included two incidents traced to the automatic voltage regulator because of unstable excitation. Erratic excitation also caused another unit to trip.
Failure of package ventilation caused a unit trip, putting that plant out of service for two weeks. Issues with the steam injection system for another engine forced a 10-day outage. All other incidents described by the SPS engineers were relatively minor in terms of outage time.
