By W. Cary Campbell
During the initial commissioning and routing performance testing of several simple cycle and combined cycle gas-fired power plants, several problems limiting plant performance have occurred, including high compressor inlet temperatures. Photo courtesy: True North Consulting
Ratings, capacity payments anddispatch rankings of gas turbine and combined cycle units are often determined by periodic maximum capacity demonstration tests. For the plant owner or operator, these tests may establish the potential revenue and profitability of a unit. For the power purchaser or utility the tests should demonstrate what that the unit can reliably produce when called upon for operation.
To achieve maximum plant output and receive credit for actual dependable capacity, the designers, owners, and operators of a gas turbine or combined cycle power plant must properly consider all environmental, permitting, and operational limitations of their equipment. During initial commissioning and routine performance testing of numerous simple cycle and combined cycle gas turbine power plants over the past 25 years, the author has encountered several design, analytical and operational problems which limited the output and demonstrated capacity of these units. This article describes many of these problems and methods used to resolve them.
The design capability of a unit is a function of the equipment utilized, expected load profile, site elevation, prevailing wind direction, ambient temperature and humidity, barometric pressure, cooling water source, fuel properties and air permit requirements. The actual demonstrated plant capability may not live up to these expectation due to initial deficiencies in equipment capabilities, degradation, instrumentation problems, control issues, inefficient operation, or improper corrections for test conditions.
During the design phase of a new plant or retrofit, opportunities for producing maximum output should be considered. These may include the following:
- Gas turbine inlet air cooling
- Steam or water injection into gas turbines
- HRSG duct burner heat input
- Elevated steam turbine throttle pressures
- Minimal auxiliary power requirements
The base load capability of a gas turbine or combined cycle unit is primarily effected by the site elevation, location, orientation and prevailing ambient conditions. A plant located at a high elevation will always be limited by the atmospheric pressure at that altitude. When performing a capacity demonstration test, the measured output should be corrected to the standard atmospheric pressure for that altitude. For example, correcting the output of a unit located at 1000 ft. to the atmospheric pressure at sea level does not indicate the realistic capability of the unit at its actual elevation. Likewise, the atmospheric pressure used for correcting output to reference conditions should be based on the actual pressure at the site, not the pressure corrected to sea level reported at a nearby airport.
The corrections for ambient temperature and humidity should be based on the actual conditions entering a control volume surrounding the entire unit. Corrected gas turbine and unit output experienced during a test with the wind direction from one direction may not represent the actual capability of the unit during different conditions.
The heat and humidity from the steam turbine, HRSG and cooling tower should be blown away from the gas turbine inlets when the wind is from the northwest. However, when the wind is from the east, heat and humidity from the HRSGs and cooling tower may be blown toward the gas turbine inlets. Likewise, any heat from equipment located near the gas turbine inlets such as generator coolers, lube oil coolers, transformers, control compartment air conditioners, and even heat from asphalt pavement can get ingested into gas turbine inlets. These heat influences have been indicated during tests when gas turbine inlet air temperatures were found to vary on different levels or sides of gas turbine inlets. These heat sources can also affect the measurements of ambient conditions used in the corrections of unit capability.
In one example, corrected unit output for a unit was inconsistent. After investigation, the station instrument being used for ambient temperature corrections was found to be located above an MCC building air conditioning unit. Whenever the air conditioner turned on, the apparent ambient temperature would go up, as would the corrected output. This problem was eliminated by the installation of a dedicated station weather station upwind of the unit.
The ability of a unit to produce maximum output depends on the maximum capabilities of the major equipment, power augmentation modes available and how close the operator or control system runs to these limits. Individual components may be limited due to pressure, flow, temperature or cooling requirements. Overall unit capabilities may be limited by environmental, emissions, or even residential noise restrictions.
For example, the following limits applied to a 650 MW 2-on-1 F-class combined cycle unit:
- Evaporative coolers ON only when ambient temp. >= 60 F
- GT steam injection only when comp. inlet temp. >= 59 F
- HP Superheater outlet header pressure
- Ductburner heat input
- Ductburner outlet (SH tube surface) temperatures
- HP steam flowrate
- Steam turbine generator output
- Generator cooler service water temperatures
- Main Transformer MVA
Any one of these parameters may be the limiting factor to maximum unit output on a given day. If a unit is running at base load, ambient conditions may limit the output based on whether evaporative coolers can be operated. If a unit running in full pressure mode, duct burner heat input may be the limit. All of these limits should be monitored while the unit is loaded until one of them becomes the limiting factor.
Figure 1 illustrates the potential benefit and full capability with the various power augmentation systems. In this example, a maximum output of 650 MW is expected at 75 Deg. F and 45% RH with ductburners, steam injection and evaporative coolers in operation. At the same ambient conditions without any power augmentation, expected output is only 492 MW.
Even though the tested performance of the individual pieces of major equipment may have meet their guarantees (when corrected to their particular guaranteed conditions), the net output of the entire unit may be less than expected. What the HRSG or steam turbine can produce at its reference conditions is different from what it can produce at the actual gas turbine and HRSG outlet conditions.
GAS TURBINE PROBLEMS
Low unit output and low heat supplied to an HRSG may start with the gas turbine. Problems seen with gas turbine performance include:
- Dirty inlet filters
- Dry evaporative coolers
- Recirculation or ingestion of hot or humid air
- IGV position indications
- Fouled compressors
- Broken inlet total pressure probes
- Exhaust temperatures
- Compressor air leakages
- Low steam injection flow
Stream injection has the potential to increase F-class gas turbine outputs by up to 15 MW each. Problems with pressure regulation, flow metering, and combustor flame stability have been found to limit steam flowrate and reduce total unit output.
It should be noted that using steam from the HRSG reduces flow to the steam turbine. However, in all cases tested, it was found that more net output was achieved with maximum steam injection to the gas turbine, even at the expense of less flow to the steam turbine.
Limits in HRSG steam production and performance have been found due to:
- Low ductburner heat input
- Improper relief valve setpoints
- HP, IP and LP bypass valve leakages
- HRSG tube leaks
- Cycle isolation problems
One of the most common limitations in HRSG capability is ductburner heat input. Since the ductburners are a source of air pollutants, the maximum heat input to the ductburners is regulated by the air permit. These limits may be stated in MMBtu/hr based on either HHV or LHV.
The flowrates are typically measured using orifice meters which produce a differential pressure proportional to volumetric flow. Typical units of ductburner flow are in standard cubic feet per minute (SCFM). This requires the conversion of actual volumetric flow at the actual pressure and temperature of the gas flowing through the meter to what the equivalent volumetric flowrate would be at standard pressure and temperature (typically 14.696 PSIA and 60 Deg. F). In order to convert the flows from actual to standard conditions, a detailed analysis of the fuel properties is required so that the density of the gas at actual and standard conditions can be calculated.
Many plant do not have on-line gas chromatographs so the fuel properties are assumed to be constant from periodic fuel samples or gas company reports so they may not know the current density or heating value of the fuel. This results in incorrect calculations of ductburner heat input and the maximum allowable ductburner heat input may not be utilized.
In order to simplify the process, many plants assume the relationship between ductburner heat input and flowrate (in SCFM) is directly proportional, so they set their ductburner heat input strictly by adjusting the indicated flow to the expected SCFM at full heat input. This often results in the maximum allowable heat input not being supplied to the ductburners, resulting in less steam production and unit MW output.
LHV or HHV?
Another common problem found is a lack of understanding of whether the ductburner limits are based on LHV or HHV. In one example, ductburners were sized according to an expected heat input of 488 MMBtu/hr LHV which is equivalent to 541.7 MMBtu/hr HHV. However, the plant computer was programmed to limit the ductburner flowrate to an SCFM flowrate equivalent to 488 MMBtu/hr HHV – which is actually only 90% of the permit limit. This resulted in an HP steam flow approximately 4% low. After correcting the heat input limit to the HHV value, unit output increased 12 MW.
Ductburner Flame Distribution
It should be noted that both the HRSG and steam turbine must be sized to handle the increased tube surface temperatures, superheater outlet pressures and temperatures, superheat spray requirements and steam flows produced by maximum duct burner heat input. For units tested with a maximum tube surface temperature of 1500 Deg. F, it only takes one hot spot indication to limit the total heat input to the entire set of ductburners. This hot spot may be due to abnormally long flame lengths due to improper sizing or spacing of burner holes, or tube surface temperature instrumentation errors. Improper distribution of burner heat can also be detected by an imbalance in superheater outlet header steam temperatures.
For example, during ductburner testing of a large 2-on-1 G-class unit, superheater outlet steam temperatures on the far end of the header were approaching design limits which required limiting ductburner heat input. During an outage, the vendor drilled out or inserted orifices along the length of ductburner burner holes to produce proper distribution of the gas. Upon startup ductburner flame lengths and superheater header temperature were more uniform and duct burner heat input could be increased to the maximum allowed limit.
Ductburner Metering Errors
Other limitations in ductburner heat input result from metering errors of ductburner fuel flowrate. Since natural gas may come from a pipeline at pressures as high as 6000 PSIG, it gets extremely cold when expanded to the pressure required in a combined cycle power plant (usually around 350 PSIG). Many plant have fuel heaters which warm the gas to at least 280 Deg. F. Typical fuel temperatures required by the ductburners is 80 Deg. F.
In one plant tested, the fuel temperature entering the ductburners was regulated by mixing the heated gas (at 280 Deg. F) with unheated gas (sometimes as low as 24 Deg. F) as shown in Figure 2. The resulting mixed gas temperature was monitored downstream of the orifice plate. During testing the indicated flowrates appeared unrealistic. Upon inspection it was noticed that condensation was present on the top of the pipe just upstream of the orifice while the bottom on the pipe was almost too hot to touch. This showed that the gas had not fully mixed before entering the orifice and the flowrate calculated using the downstream temperature was not reliable.
In order to resolve this issue, the orifice plate was moved farther downstream to allow more time for mixing as shown a similar meter in Figure 3.
The location of these meters can also have a significant effect on the reliability of the flow calculations. Some ductburner flow meters are located downstream of pressure regulators where they operate at low pressures which vary with ductburner flow demand. The natural gas pressure can vary significantly at these location causing large swings in actual orifice meter volumetric flow, differential pressure and head loss. Therefore, it is preferable to locate ductburner flow meters upstream of the pressure regulators where they experience a more constant pressure.
MultivariaBLE Flow Transmitters
Due to the complexity of the conversions of volumetric flow indications from actual line to standard conditions, many plant now use multivariable flow transmitters for ductburner fuel and steam injection flow measurements. These instruments perform density and flow calculations internally so they require accurate flow element information including dimensions, calibration data, reference fluid properties and valid sensor input data.
The flow calculations performed by these instruments are often found to be more accurate than plant computer calculations; however, the data manually entered or input by attached sensors must be valid. One common problems is that these instruments often only output a signal for the calculated flowrate and they continue to do this even if the RTD input to the transmitter has failed. Therefore it is recommended that all sensor inputs including differential pressure, pressure and temperature as well as the calculate flowrate be output.
Since these transmitters may be the primary indication of the maximum heat input to the ductburners and steam injection to the gas turbines, it may be useful to call in an instrument specialist or test consultant to verify that these instruments are programmed correctly and working properly.
STEAM TURBINE PROBLEMS
Output from the steam turbine is generally determined by the steam flows and conditions provided by the HRSGs and the efficiency of the turbine. Problems found during testing include:
- HP and reheat temperature controls set to HRSG outlet conditions rather than turbine inlet
- HP steam pressure set to header indications rather than turbine inlet
- Throttle valves not wide open
- Condenser and cooling tower performance
- Cycle isolation
- Air in-leakage
- Vacuum pump problems
OTHER PLANT PROBLEMS
Net unit output can also be limited due to:
- Gas chromatograph errors
- Fuel heater problems
- Erroneous or unrealistic correction curves
- Watt metering errors
- Abnormal power factors during tests
- Transmission system losses
- Station service loads
Ideally all of the problems discussed herein would be found and resolved before starting a capacity demonstration test; however, many tests are conducted on short notice when ambient conditions or load conditions make it economical to test. These tests are sometimes conducted soon after a unit starts up from an outage or after a period of extended shutdown when many power augmentation systems may have not have been used. Many problems with these systems, such as a cracked evaporative cooler pipe, may not become apparent until the system is started just for the purposes of performing the capacity test.
MAXIMUM CAPABILITY CHECKLIST
In order to prepare and verify that the unit can produce its maximum output and demonstrate it during the test, the following activities are recommended:
- Check calibration of key control and test instrumentation
- Clean or replace inlet air filters
- Clean out evaporative cooler distribution headers and verify proper flow settings
- Verify that compressor inlet pressure probes are undamaged and properly oriented
- Check bleed heat, compressor blowoff and bypass valves for leakage
- Perform an off-line water wash
- Verify gas turbine control constants
- Inspect steam injection and duct burner flow orifice plates for wear or damage
- Verify proper measurements of steam injection and ductburner flows
- Check HP, IP and LP bypass valves for leakage
- Clean condenser cleanliness and air in-leakage
- Verify proper cycle isolation
- Validate gas turbine control inputs and constants
- Test inlet cooling systems
- Test steam injection systems
- Test ductburner systems, maximum flow and pressure capabilities
- Consider hiring a test consultant to evaluate preliminary performance and resolve any problems before the official test
KEY PLANT INSTRUMENTATION
Many problems are related to improper control due to bad feedback instrumentation. The following instruments are critical for accurate unit control and testing:
- Atmospheric pressure
- Compressor inlet pressure and temperature
- Compressor discharge pressure
- GT exhaust temperatures
- IGV position
- Steam injection flow DP, pressure and temperature
- Ductburner flow DP, pressure and temperature
- Gas chromatograph readings
- Superheater outlet header pressure and temperatures
- HP steam pressure and temperature (at turbine inlet)
- Hot reheat temperature (at turbine inlet)
- Revenue watt meters
- Plant ambient temperature and relative humidity (if used for the test)
During the test, the following operating modes and procedures are recommended:
- Allow ample time to start the unit, work out any problems, and allow the unit to stabilize
- Monitor HP steam pressure as close as possible to the steam turbine inlet
- Verify that all cooling water pumps and cooling tower fans are running
- Minimize the number of condensate and feedpumps running
- Adjust steam turbine generator power factor to near 1.0 to minimize MVA
- Leave auxiliary equipment and station service in routine conditions
- Measure atmospheric pressure on-site
- Utilize the same watt meters that will be used for revenue payments; however, compare and validate net output metering against individual generator output and auxiliary power measurements
- Consider hiring an independent test consultant to witness and supervise the tests
Lessons learned from capacity testing of numerous gas turbine and combined cycle units have shown many design deficiencies, equipment problems and operating conditions that can limit demonstrated maximum dependable capacity. Through diligent effort and cooperation of test consultants, plant engineers, unit designers, vendors, and plant operators, most of these problems can be overcome and the maximum capability of a combined cycle unit can be achieved and demonstrated.
W. Cary Campbell is senior consulting engineer for True North Consulting, LLC.