Ultra-low-NOx combustors use reduced peak flame temperatures within the combustor to limit the formation of thermal NOx through staging strategies such as Lean Premixed combustion. As combustion temperatures are decreased in low NOx applications, designers must address other undesirable combustion phenomena, in particular the onset of combustion instability in the form of Lean Blow Off (LBO). Experimental methods to analyze LBO within the design cycle are too costly and time consuming given today’s time-to-market pressures. Using advanced combustion simulation tools instead, designers can leverage existing CFD data to initiate a simulation and evaluate how close the flame is to LBO. The speed and accuracy of this approach makes it easier for combustor designers to evaluate LBO behavior for different designs and operating conditions.

1. Using Advanced Simulation for Faster, Less Costly
Lean Blow Off Analysis
By Reaction Design
6440 Lusk Blvd, Suite D205
San Diego, CA 92121
Phone: 858-550-1920
www.reactiondesign.com

2. Reaction Design White Paper
INTRODUCTION
Ultra-low-NOx combustors use reduced peak flame temperatures within the
combustor to limit the formation of thermal NOx through staging strategies such as Lean
Premixed combustion. As combustion temperatures are decreased in low NOx
applications, designers must address other undesirable combustion phenomena, in
particular the onset of combustion instability in the form of Lean Blow Off (LBO).
Experimental methods to analyze LBO within the design cycle are too costly and time
consuming given today’s time-to-market pressures. Using advanced combustion
simulation tools instead, designers can leverage existing CFD data to initiate a
simulation and evaluate how close the flame is to LBO. The speed and accuracy of this
approach makes it easier for combustor designers to evaluate LBO behavior for
different designs and operating conditions.
LEAN COMBUSTION MAY LEAD TO LBO
In recent years, combustor designers have applied ultra-low-NOx combustion
design techniques to successfully reduce emissions from gas turbines, burners and
boilers. The reductions in NOx emissions result predominantly from lean combustion
techniques in which the flame burns at lower temperatures in a fuel-lean condition and
forms less NOx. But, if you take too much fuel away from the flame, it blows out. This
LBO point is one of the key factors limiting further reductions in NOx for devices such as
gas turbines.
A related factor that limits reductions in NOx is combustion acoustics, in which
the flame-attachment point becomes unstable and develops a resonant frequency with
vibrations strong enough to potentially lead to catastrophic damage. Combustion
acoustics can also be linked to parts of the flame experiencing LBO and then reigniting
in a repetitive pattern. So, LBO underlies two of the key challenges facing lean
premixed combustion systems.
2 of 8

3. Reaction Design White Paper
EXPERIMENTAL LBO ANALYSIS IS COSTLY AND TIME CONSUMING
Commercial combustion equipment designers have been measuring LBO
experimentally for many years. It is common to use experiments to determine the
stability loop of combustion devices by determining the extinction points with the fuel-
lean and fuel-rich regions of operation ( Figure 1). These experiments are quite simple,
yet expensive, to carry out – decreasing the fuel at a specified airflow condition to the
point where the flame blows out or combustion acoustics drive potentially harmful
vibrations. In this way, an air/fuel ratio at LBO can be determined (Figure 2).
Figure 1: Combustion Stability Loop a) Constant Pressure and b) with Pressure Variation
1
(Lefebvre, 1999) .
While this global method of determining LBO is valuable to the combustion
system operator, it has limited value to the combustion system designer, who may
understand when the combustor will blow out but doesn’t understand why, where or
how it blows out. In other words, the combustor is treated as a black box and thereby
provides none of the information that a designer needs to improve the design. The
designer requires local, or spatially resolved, information on the stability of the flame to
see where the flame is struggling and then be able to perform what is needed to
improve stability. Understanding must be achieved locally in order to improve stability
globally. Using experiments to achieve that understanding is expensive and time
1
Lefebvre, A. H. (1999). Gas Turbine Combustion. New York, Taylor & Francis Group, LLC.
3 of 8

4. Reaction Design White Paper
consuming, so designers turn to accurate and efficient simulation tools to minimize the
need for experimental resources.
Figure 2: Typical Method of Evaluating LBO.
LBO SIMULATION MUST FOCUS ON LOCAL EFFECTS
Assessing LBO requires an analysis of the combined effects of fluid dynamic
mixing and fuel reaction rates. This analysis must focus on the specific areas in the flow
field that represent the portions of the flame critical for combustion stability. These
areas are usually found in the flame front or where a recirculation zone impacts the
downstream portion of the flame. An assessment of the ratio of the characteristic
chemical reaction time scale to the characteristic mixing or transport time scale, referred
to as the Damköhler Number (Da), is often used to predict LBO in experimental and
commercial combustion systems.
The Da is typically defined as
Analysis of the Da number provides a measure of combustion stability, as
follows:
• If the chemistry time scale, τchem, is much faster than the mixing time scale, τmix,
(i.e., Da << 1), then combustion is presumed to be stable.
4 of 8

5. Reaction Design White Paper
• As the chemistry time scale approaches the mixing time scale (i.e., Da ~ 1), then
local extinction, and possibly LBO, is more likely to occur.
This analysis is often conducted using global estimates for the chemical and
mixing time scales. A local assessment of the chemical and mixing/transport time scales
yields a more accurate understanding of when the chemistry of the fuel reaction is
dominant and when it is slowed to the point that it can be limited by fluid dynamics.
Several simulation methods exist to evaluate LBO. These fall into two main
categories: (1) detailed combustion analysis with idealized reactors such as CHEMKIN,
and (2) Computational Fluid Dynamics (CFD). In the former, idealized reactors such as
Perfectly Stirred Reactors (PSR) are able to evaluate the overall reaction rate at specific
conditions that allow a designer the ability to determine the chemical time scale.
However, the mixing time scale must be estimated by other means. In the latter, CFD
allows the calculation of the local mixing in the combustor, but cannot handle the multi-
step chemical reaction mechanism required to account for the key reactions that lead to
obtaining an accurate reaction rate and determining the correct chemical time scale
[see Accurate Chemistry Simulation Enables Clean Combustion Design for Power
Generation].
LEAN BLOW OFF ANALYSIS WITH ACCURATE CHEMISTRY
Reaction Design’s ENERGICO™ Simulation Package provides an innovative,
patent-pending approach to defining Damköhler numbers for a combustor by taking
advantage of the strengths of both types of simulation: the local flow and
thermochemical properties extracted from a CFD solution and the detailed combustion
kinetics available in the full chemical mechanism. Rather than trying to define a pair of
“global” chemical and flow residence times of a combustor, the tool’s local LBO analysis
defines both chemical and residence times locally to account for the spatial variation of
mean flow, turbulence and gas-mixture propertiesi.
In ENERGICO, the simulation can define the local (CFD-cell) flow residence
times by either the mean cell velocity and volume, or by the cell turbulence kinetic
5 of 8

6. Reaction Design White Paper
energy and dissipation rate (or by the minimum of the two). The simulation determines
local chemical time scale, based on local cell temperature and gas composition, by
using a detailed combustion-chemistry calculation for each cell. The resulting LBO
analysis verifies the integrity of the flame locally and then provides an indication of the
overall soundness of the flame zone visually, as contours of the local Damköhler
number. The Damköhler number distribution exposes location and size of the stable
flame core in the combustor. This local Damköhler number data does not immediately
indicate whether or not the flame will blow out. It is only by examining the structure and
topology of the flame core and the integrity of key regions in the flame that the designer
can assess the likelihood of blow-off as well as the adequacy of the CFD mesh for flame
predictions in that area.
To perform an analysis of the local Damköhler number on the flame, the tool first
processes the flame region through a filter that identifies local areas of high fuel fraction
and high temperature gradients. This filter defines the flame region by determining all
the computational cells from the CFD simulation that meet these criteria. Once the
flame front has been captured in this way, the tool conducts an analysis of the
characteristic chemistry and mixing time scales on a cell–by-cell basis. These
calculations determine the local Damköhler number for each cell within the flame region.
For example, the simulation of a wall-jet combustor, shown in Figure 3, colors all of the
cells associated with the flame according to their specific Damköhler number. The
combustor designer can now see how viable the flame is in regions where the CFD
simulation asserts the flame to be stable. The designer can also assess how stable the
most critical flame attachment points are with this spatially-resolved information.
6 of 8

7. Reaction Design White Paper
Figure 3: Local Damköhler Number for a Wall Jet Combustor.
The usefulness of any simulation tool in a commercial environment is related to
computational time as well as accuracy. ENERGICO can complete calculations for a
local Damköhler number, within the flame zone, in minutes, with a fully detailed
chemical reaction mechanism.
In another example, the result of an analysis for a commercial gas-turbine
combustor fuel injector appears in Figure 4. Here, the periphery of the flame region is
characterized by a Damköhler number just below unity, indicating a stable-flame
condition. Note that the region in the center of the premixed fuel-air core indicates local
Damköhler numbers higher than unity. The temperatures in this region are low – i.e.,
close to the inlet temperature – and the local Damköhler number indicates that the fuel-
air mixture has not yet been heated to its ignition point. In this type of swirl-stabilized
flame, it is critical that the recirculation zone remains quite stable, so it can continue to
ignite the oncoming fuel-air mixture. These data show that the majority of the
recirculation zone and the region nearest the fuel injector have very low (stable)
Damköhler numbers.
7 of 8

8. Reaction Design White Paper
Figure 4: Local Damköhler Number Calculation for a GE Energy Fuel Injector (Drennan 2009).
SUMMARY
Using ENERGICO for advanced combustion simulation, designers can quickly
capture the combustion flame from CFD analysis, conduct detailed local analysis with
visualization, and evaluate how close the flame is to LBO. The data generated with this
analysis are critical to help a combustor designer understand how stable the most
critical regions in their flow field are. The computational speed of this approach makes
it easier for combustor designers to evaluate LBO behavior for different designs and
operating conditions.
i
Drennan, S.D, et al. “Flow field Derived Equivalent Reactor Networks for Accurate Chemistry Simulation
in Gas Turbine Combustors” GT2009-59861, Proceedings of the ASME 2009 Turbo Expo, 2009.
8 of 8