Accurate fuel-nozzle flow testing critical to optimal DLN2.6 operation

Sometimes you think you know something until you listen to someone who really knows that something, making you aware of how little you really know. Fuel-nozzle flow testing may be one of those things. Mitch Cohen of Orlando-based Turbine Technology Services Corp (TTS), a respected combustion-system expert, presented work on that subject developed by the Electric Power Research Institute (EPRI) during the vendor presentation portion of the 7F Users Group annual conference, last week, in Greenville, SC. Judging from the questions, there were some bona fide subject-matter experts at the 45-min SRO breakout session for owner/operators, but there also were a few others at the opposite end of the knowledge spectrum.

Flow testing is a longtime key component of both the fuel-nozzle manufacturing and repair processes, even for diffusion and early DLN systems having NOx targets of 25 ppm or higher. For these older systems, acceptable combustion performance could be achieved with a higher degree of combustor-to-combustor variation in the fuel/air ratio than today’s more sophisticated combustor designs require. Consequently, methods used to flow test older model fuel nozzles did not have such stringent requirements for accuracy.

However, the evolution to advanced low-NOx combustion systems—such as the DLN2.6—demands more accurate, reproducible, and repeatable measurements and tighter control of a nozzle’s effective flow area. This accuracy and tighter control extends both to the absolute area of nozzles, which impacts the dynamic characteristics of the combustion process, and to the variation in area among sets of fuel nozzles and components, which impacts—most importantly—margin from lean blowout (LBO). Emissions and dynamics also are affected.

Cohen’s presentation announced the availability of the EPRI technical update 1023970, “Fuel Nozzle Flow Testing Guideline for Gas Turbine Low-NOx Combustion System,” that he, Leonard Angello of EPRI, and Hans Van Esch of Turbine End-user Services co-authored to accomplish the following:

• Assist operators in the selection and qualification of fuel-nozzle repair/flow-test vendors.

• Provide a detailed methodology for obtaining accurate flow-test results, which discusses both equipment and flow-test procedures.

• Identify the steps involved in validating flow-test results.

• Provide guidance on both how to interpret flow test data and how to use that information to establish criteria for returning nozzles to service.

Cohen told attendees that optimal operation of 7FA DLN2.6 combustion systems requires the balancing of several competing requirements, including these:

• Maintaining NOx emissions at less than 9 ppm.

Preventing LBO trips.

• Maintaining the amplitudes of hot, cold, and chug tones within acceptable limits.

• Achieving the required load turndown while holding CO emissions within permit limits.

Tight control of fuel/air ratio is critical to these goals, he said, adding that exacting fuel-nozzle flow testing is required to assure this level of control can be achieved. However, the experience of TTS in troubleshooting DLN operational problems, Cohen noted, is that operations personnel sometimes are unaware of the basic requirements of flow testing or how to evaluate data they receive. In many instances, he added, operators are not even sure if they have received flow-test data for nozzles that are in service, and/or they are unable to locate this information. These findings helped provide the incentive to develop the EPRI guideline and the accompanying EPRI calibrated-fuel-nozzle testing kit (Fig 1).

1. EPRI calibrated-fuel-nozzle testing kit ensures accurate results by incorporating orifice plates calibrated to NIST standards as well as detailed test procedures

1. EPRI calibrated-fuel-nozzle testing kit ensures accurate results by incorporating orifice plates calibrated to NIST standards as well as detailed test procedures

What is flow testing? Cohen began his tutorial by answering this question. He said, “It is the measurement of the effective flow area, or simply the ‘effective area’ of a flow passage, usually expressed in square inches.” For a standard venturi nozzle measurement is relatively simple, but for more complicated geometries—such as a fuel nozzle—the effective area must be determined in a flow test rig (Fig 2). Cohen continued: “In a flow test rig, if we know precisely the mass flow and gas composition (dry air) we can calculate the effective area by measuring the upstream and downstream pressures and the temperature.”

2. Flow testing of a DLN2.6 end-cover assembly

2. Flow testing of a DLN2.6 end-cover assembly

Flow testing is important, he said, to achieve equal (that is, as close to equal as technology allows) fuel flows through the respective fuel passages (PM1, PM2, PM3, Quat) of each combustor by minimizing the variation in effective area from can-to-can. Cohen pointed out that, for a given fuel manifold, each combustor sees the same pressure and temperature conditions and that any difference in fuel flow among the chambers is due solely to differences in effective area.

Another reason flow testing is important is that failure to operate at the design pressure ratio across fuel nozzles is conducive to combustor dynamics issues. Achieving the correct pressure ratio demands high accuracy in measuring the targeted effective area of the fuel nozzles. Cohen used this as a segue to discuss how the reliability of flow test results are validated. Accuracy, repeatability, and reproducibility of the data are vital to enable proper tuning of the engine. He defined those terms this way:

• Accuracy is the deviation of the measured effective area from the actual effective area.

• Repeatability is the ability of a measurement system—a flow test rig in this case—to reproduce a measurement while operating under an unchanging set of conditions.

• Reproducibility is the ability to reproduce a measurement under different conditions of pressure, temperature, and humidity, and at different points in time. It presumes the test rig has been broken down and rebuilt between measurements. Reproducibility also is an indicator of the consistency of the setup and testing procedure used by different, or the same, technicians responsible for measurements.

The EPRI guideline recommends these quantitative limits for each validation parameter:

• Accuracy within plus/minus 0.75% of a calibrated master standard part—defined as one whose effective area has been measured by an independent calibration lab using instrumentation traceable to the National Institute of Standards and Technology.

• Repeatability: Two standard deviations; less than 0.5% of the average measurement.

• Reproducibility: Two standard deviations; less than 0.75% of the average measurement.

Having valid measurements is only table stakes, however. How this information is used to develop test specifications and to qualify/reject fuel nozzles is the owner/operator’s challenge. First step, Cohen said, is for the operator and vendor to discuss and agree on the limits for can-to-can variation in effective area for each flow passage (PM1, PM2, etc). This typically is expressed as a limit of the percentage calculated by subtracting the minimum effective area from the maximum and dividing that by the average effective area. He added that many vendors offering flow-test services use this as the only qualifying criterion.

But that is a mistake, Cohen believes. For each gas passage, the average effective area also should be within a certain percentage of a specified target value, calculated by subtracting the targeted effective area from the average and dividing the result by the targeted effective area.

He went on to illustrate, by use of a series of charts which will be available on the 7F Users Group website by mid-June, how numbers written into a specification can appear correct but may be unsatisfactory for assuring proper tuning of advanced DLN combustion systems. The example presented showed two sets of PM3 assembly flow test data having the same average effective area and the same max-to-min range, 4.0% in this case. For the first data set, the variation from the average effective area was +2.1%, -1.9%; for the second set, +0.6%, -3.4%. Both sets of numbers satisfy the spec, but the second set does not meet the intent of the spec.

Here’s one reason why: One combustor always will be the lowest flowing of the 14 and the first to flame out if operated too lean. Minimizing the variation in average effective area—at the low end in particular—will provide more margin from potential LBO. Another reason: High- or low-flowing cans can result in excessively high dynamics in only one or two combustion chambers and this can lead to high NOx emissions when the entire combustion system is tuned to reduce the dynamics in the one or two affected cans.

For more information on the EPRI guideline and calibrated-fuel-nozzle testing kit, contact Leonard Angello at 650-855-7939 or langello@epri.com.

 

Posted in 7F Users Group |

Comments are closed.