COMBUSTOR DEVELOPMENT IS EASY (PART TWO)

Published on: 

Part One (Nov/Dec13, p.40) introduced some problems involved in combustor development. This theme is continued in Part Two.

Most operators require that a gas turbine can operate over a reasonably wide range of fuels. Specifically, in pipeline and gas field duty, there is an expectation that a gas turbine can operate over a significant Wobbe index range. The Wobbe index is a number that relates the fuel’s heating value to its specific volume.

But there are some misconceptions about this: Wobbe index only dictates the fuel system’s pressure drop to the chemical energy input rate into the gas turbine for a specific fuel. Effectively, if two different fuels have the same Wobbe index, they will have the same pressure drop or volume flow through the fuel control system.

This is important since all components in the fuel system (pipes, valves, injectors, orifices, and so forth) must be sized for a specific volume flow and pressure drop. If the Wobbe Index of a given fuel changes drastically, the fuel system may not be able to handle the required volume flow to maintain the energy input into the combustor required to reach the gas turbine’s design output power (low Wobbe index) or the fuel control valve cannot provide adequate control at part load (high Wobbe index).

The Wobbe index relates two basic physical properties of the fuel but does not provide any information on combustion kinetic properties such as flame speed, reaction rate and auto-ignition delay time. Thus, when determining whether a specific fuel can be used in a gas turbine, one cannot simply rely on Wobbe index but must look at all the combustion kinetic properties.

Reaction rate

Another combustion property that impacts the combustor design and sizing is the reaction rate of the fuel-air mixture. Several hundred different reaction and sub-reactions occur when natural gas and air combust. The reaction rate characterizes how quickly these reactions are completed and the heat from the combustion reactions has been fully released.

All combustion reactions must be finished before the mixture exits the combustor and enters the turbine section. Consequently, the reaction rate sets the mixture’s required residence time in the combustor. Further, depending on the amount of mixing, turbulence intensity and overall aerodynamic flow structure in the combustor, this usually then determines the physical volume of the combustor or the combustor transition piece.

For gas fuels this is normally not a difficult design challenge as most reactions complete relatively quickly. But when liquid fuels are used that must be atomized and evaporated first or, even worse, when heavy hydrocarbon liquid drop-outs occur in gas fuels, the transient time is also related to the liquid droplet size.

Liquid droplets must fully evaporate before the fuel can react with the oxygen in the air. Thus poor atomization or liquid streaks in the gas fuel will not completely combust in the combustor and carry into the turbine section where they can cause localized combustion hot spots and severe damage.

Oscillations

Advertisement

The phenomenon of combustor pressure oscillations, often also called combustor “humming,” “rumble,” or “pulsations,” is well documented for gas turbines with dry low emissions (lean premixed) combustion systems. Combustor oscillations can lead to metal fretting of casing and support structures, premature wear of sheet metal structures such as combustor liners, performance deterioration and limitations, and sometimes premature equipment failure. It is characterized by pressure pulsations with rounded but distinct frequency peaks in the combustor and can be identified using acoustic or dynamic pressure measurements.

Although this phenomenon is primarily thought to only be a problem with lean premixed combustors, conventional diffusion flame combustors can sometimes also experience oscillations. In lean premixed combustion systems, oscillations are usually caused by acoustic resonance, resulting from the time lag between the fuel or air release into the premixing zone and the heat release in the combustor.

An unstable flame front creates small pressure waves in the combustor. When the frequency of these waves coincides with an acoustic resonance frequency of the combustor geometry, significant pressure amplitude gain and strong harmonic pulsations can occur. Furthermore, if the frequencies of these combustion-excited harmonic pulsations coincide with a structural resonance frequency of the gas turbine, large vibrations are usually the result.

Because of the complex geometry of the combustor and the many independent structural parts in a gas turbine, many coincident acoustic-structural resonance frequencies are possible. Typically, most severe combustion oscillations problems are at low frequency (acoustic bulk longitudinal mode) instabilities (0-100 Hz) but higher 3-D mode frequencies up to 1,200 Hz can also be excited (acoustic combustor circumferential or transverse modes).

Often combustion oscillation problems in DLE systems are related to other issues:

• The Wobbe index is outside of the manufacturer’s allowable limits

• Liquid hydrocarbons or water form in the fuel supply system and are carried into the combustor causing a non-uniform fuel distribution

• The combustor nozzle is severely degraded because of corrosion or coking.

These are just a few of the many design challenges a manufacturer faces when designing a modern gas turbine combustor. Designing a combustor that meets strict emission standards, can handle a wide range of fuels, is efficient, operates for a wide range of steady state and transient operating conditions, and most important, operates reliably for several years, is extremely difficult. Buyers and operators of gas turbines should carefully evaluate and verify the actual field experience and fleet operating hours of any new combustion system they intend to use.

Unfortunately, in the past, some unproven designs have been introduced to the market, some of which have caused significant downtime and costs for operators. When accepting a new gas turbine design or upgrading to a novel combustor design it is imperative to obtain real references and perform the necessary engineering due diligence.

Authors

Klaus Brun is the Machinery Program Director at Southwest Research Institute in San Antonio, Texas. He is also the past Chair of the Board of Directors of the ASME International Gas Turbine Institute and the IGTI Oil & Gas applications committee.

Rainer Kurz is the Manager of Systems Analysis for Solar Turbines Incorporated in San Diego, CA. He is an ASME Fellow since 2003 and chair of the IGTI Oil & Gas applications committee.