Paper # 070IC-0320 Topic: Internal Combustion and Gas Turbine Engines

8th

U. S. National Combustion Meeting

Organized by the Western States Section of the Combustion Institute

and hosted by the University of Utah

May 19-22, 2013

Experimental Study of Traversing Hot-Jet Ignition of Lean

Hydrocarbon-Air Mixtures in a Constant-Volume Combustor

Prasanna Chinnathambi Abdullah Karimi Manikanda Rajagopal Razi Nalim

Mechanical Engineering, Indiana University-Purdue University, Indianapolis, IN 46202, USA

A constant-volume combustor is used to investigate ignition by a traversing jet of reactive hot gas, in support of combustion engine

applications that include novel constant-volume wave rotor, combustion gas turbines and pre-chamber IC engines. The combustor rig

consists of two combustion chambers: a rotating pre-chamber that issues a traversing hot jet, and a stationary main chamber with long

aspect ratio similar to a wave rotor combustor. In the pre-chamber, a fuel-rich mixture is ignited using an electric spark, generating

high pressure that results in breaking a diaphragm and discharge of a hot gas jet through a nozzle. The traversing nozzle sweeps the jet

of hot gases across one end of the main chamber containing fuel-lean mixture at room temperature and pressure, with complex jet

penetration, entrainment, and vortex dynamics. Two different jet traverse speeds were used, with traverse time fixed at 40.6 ms and

8.1 ms, which correspond to pre-chamber rotation rate of 150 rpm and 750 rpm, respectively. The overall equivalence ratio of the

main chamber mixture was maintained in the lean range at equivalence ratios of 0.4, 0.6 and 0.8. The pre-chamber uses rich ethylene-

air mixtures with equivalence ratio of 1.1 for all the test cases. Ignition delay is measured by high-speed videography and image

processing. Ignition delay time for methane-air mixtures is typically a few milliseconds, while ethylene-air mixtures exhibited less

than a millisecond delay time. Blended methane-hydrogen fuel exhibited shorter ignition delay than methane, as well as faster

apparent flame propagation speed. Supporting computational studies suggest that both chemical kinetics and mixing between the

transient jet and the combustible mixture affect ignition delay.

1. Introduction

A jet of hot reactive gas employed as an ignition source finds application in internal combustion engines

implementing lean burn technology (Toulson et al., 2010, Attard et al., 2012, Dober and Watson, 1999), pulsed

detonation engines (Lieberman et al., 2002) and wave rotor combustors (Wijeyakulasuriya et al., 2010, Matsutomi et al.,

2010, Perera, 2010). Chemically active radicals and fast turbulent mixing in the jet create an explosion that is more

energetic than a spark (Dober and Watson, 1999), allowing rapid ignition of lean mixtures. Further, the penetrating and

distributed nature of ignition can overcome mixture non-uniformity and accelerate combustion. By enabling lean

stratified mixtures, peak gas temperature is constrained, heat losses to the walls are reduced, and pollutant emissions can

be mitigated. Such ignition is of particular interest for wave rotor combustors, for which a similar constant-volume

combustor (CVC) experiment has been previously reported (Perera, 2010, Bilgin et al., 1998). In this CVC, the premixed

lean mixture is ignited using a hot jet produced by combustion of fuel-rich mixture ignited using a spark in a pre-

chamber.

A combustible mixture can be ignited by an inert gas jet or reactive gas from another combustion source. Hot-jet

ignition involves complex flow phenomena such as vortex evolution, jet mixing, and turbulence generation. Much of the

classical literature on jet ignition was concerned with avoiding ignition in mines, and typical experiments used inert hot

jets, at low velocities into quiescent mixtures. Wolfhard (1958) observed that nitrogen and carbon dioxide have similar

minimum jet temperature for ignition, while argon and helium require higher jet temperature. Fink and Vanpée (1975)

developed an overall rate expression for ignition of methane- and ethane-air mixtures by low-velocity hot inert gas jets.

Cato and Kuchta (1966) experimented with laminar hot air jets and identified jet base temperature, jet dimensions,

composition of the combustible mixture, and jet velocity, as ignition determining factors. Vanpée and Wolfhard (1959)

found correlations between the “ignition temperature” of the jet (below which ignition fails) and the “limit flame

temperature” of the successful flame, for several fuels. The presence of reactive species in the jet influences ignition

kinetics. In the present study, the composition of the chemically active jet is kept fixed for all cases, using ethylene and

2

air of equivalence ratio (ϕ) 1.1 for the pre-chamber mixture. A high-speed compressible transient jet in a confined

volume is usually accompanied by shock wave formation due to both jet initiation and combustion pressure rise, leading

to subsequent interactions that change the flame shapes. The ignition delay time for a jet-ignited CVC may be defined as

the time from jet initiation to the occurrence of rapid, visible, and pressure-generating heat release (Perera, 2010). This

includes time for physical and chemical processes, unlike purely chemical ignition delay time reported in shock tube and

rapid compression studies (Davidson and Hanson, 2004).

Bilgin (1998) developed a CVC with long aspect ratio and square cross-section, representing a wave rotor channel. The

CVC chamber mixture is ignited by a jet of hot combustion products from a separately fueled pre-chamber that could be

spun to cause the jet to traverse across one end of the CVC (Bilgin et al., 1998). The relative motion reproduces the

action of a channel in a wave-rotor combustor and the pre-chamber may be representative of a previously combusted

channel supplying hot gas. Bilgin proposed a correlation between the Damköhler number and ignition of a fuel-air

mixture in the CVC.

Perera (2010) carried out experiments on the same CVC test rig for three different fuels – methane, ethylene, and

propane by varying the equivalence ratios in the pre-chamber and the CVC chamber. The pre-chamber was set stationary

and centered on the channel cross-section in these tests. The ignition delay time and the ignitability limits for both lean

and rich mixtures were investigated for all the three fuels in the CVC chamber, for fixed operating conditions. Variation

in ignition delay time was observed for fuels with different pre-chamber equivalence ratios and nozzle geometry.

Expectedly, methane exhibited the highest ignition delay time while ethylene mixtures had the lowest ignition delay

time. The dependence of ignition delay time on CVC equivalence ratio was interesting and as yet not fully explained.

The laminar flame speed of methane (major constituent in natural gas) is lower than that of the other hydrocarbon fuels

due it its higher activation energy and this effect is most significant under lean conditions (Ren et al., 2001). The

combustion characteristics of lean methane mixtures can be enhanced by blending with a highly reactive gas such as

hydrogen (Dagaut and Nicolle, 2005). At high temperature, the concentration of H radicals due to presence of H2 induces

a fast initiation reaction: H + CH4 → CH3 + H2 (Ju and T, 1995 ).

Maxson et al. (1991) investigated a pulsed jet combustion (PJC) system by introducing hot jets as a single stream or

multiple streams into a quiescent mixture by evaluating its ability to reliably ignite and sustain rapid combustion in

otherwise slow-burning lean mixtures. Attard et al. (2010) reviewed turbulent jet ignition systems for pre-chamber spark-

ignition engines. The pre-chamber mixture is well controlled and reliably spark-ignited, producing a reactive hot jet that

acts as a distributed ignition source for the main CVC mixture. Boretti and Watson (2009 ) studied jet ignition in

conjunction with direct injection to ignite the main chamber stratified mixtures. The pre-chamber is either spark ignited

or autoignited by contact with a hot glow plug. Efficient combustion was achieved for different fuel mixtures ranging

from near stoichiometric to extremely lean.

Tarzhanov et al. (2006) investigated hot detonation products to detonate stagnant propane-air mixtures and found that

detonation initiation depends on the initial volume concentrations, mass fraction of hot detonation products, and the

energy deposited from the detonation products. Mayinger et al. (1999) derived correlations between the induction time

(ignition delay time), the mixing time of the jet, and the adiabatic auto ignition time for the fuel-air mixtures.

The main objective of the current study is to understand the process of ignition and behavior of the traversing hot jet in a

confined volume containing lean fuel-air mixture. The combustion chamber employs an optically accessible rectangular

window with high-speed imaging to capture luminosity emitted in the wavelength range of 400-1000 nm during the

combustion process. The videography sheds light on the fluid dynamic behavior of the jet, its penetration and ignition

zones as the jet traverses through and across the quiescent mixture. Two traverse rates are used, 0.983 m/s and 4.91 m/s,

corresponding to traverse times of 40.6 and 8.1 ms, and rotation rates of 150 rpm and 750 rpm. The study further reports

the lean ignitability and ignition delay times for three gaseous fuels; methane, hydrogen-enriched methane, and ethylene.

The effect of slow and moderate jet traverse speed was compared for different fuel mixtures, while keeping the nozzle

geometry and hot-jet composition fixed.

2. Experiment Design

The experiment to examine the stationary and traversing hot jet ignition processes in a long rectangular constant-volume

combustor (CVC) used a rig with rotating and structural components inherited from the work of Bilgin (1998) and Perera

3

(2010). The rig includes the main CVC chamber and a pre-chamber (Fig. 1). The CVC is a stationary rectangular channel

of square cross-section with side dimensions of 39.9 mm (1.57 inches) and 406 mm (16.0 inches) long. The pre-chamber

is a cylindrical chamber with an internal volume of 8.36×10-4

m3

(51 cubic inches), mounted on a shaft that can be

rotated by an electric motor via a belt drive. A nozzle delivers hot gas from the pre-chamber to one end of the main

chamber. The mixtures in the respective chambers are isolated during filling by an aluminum diaphragm of thickness

0.0762 mm (0.003 inches), which is ruptured during jet delivery (Fig. 1). The diaphragm is scored to facilitate and

precisely control rupture. The main chamber is placed flush against the pre-chamber before the start of each experiment

with the help of a sealing ring assembly provided near the entrance of the main chamber (Fig. 1). Details of the setup and

preparation for a stationary jet experiment are explained by Perera (2010). Fueling of the chambers with the desired

amount of fuel-air mixture was performed using the partial pressure method and the fueling system described by Perera

(2010).

The main chamber and pre-chamber are each filled with quiescent fuel-air mixtures at room temperature and atmospheric

pressure. For the current study ethylene of 99.5% purity supplied by Praxair was used to produce ethylene-air mixtures

for fuelling the pre-chamber. A conventional spark-ignition system is employed in the pre-chamber to develop the

combustion torch jet. A capacitive discharge ignition (MSD 6AL) system is used in conjunction with an ignition coil

(MSD Blaster 2) to produce the high voltage and current required for the spark plug. The pressure rise in the pre-

chamber causes rupture of the scored aluminum diaphragm delivering the jet through the convergent nozzle having exit

diameter of 6.0 mm. The hot jet enters the main chamber at one end, traversing across the open end from the upper wall

to the lower wall.

The main chamber is equipped with dynamic pressure transducers and also provides visual access through optical

windows comprising the two long sides of the main chamber. For structural reasons the leading edge of the main

chamber window is located 47.0 mm (1.85 inches) away from the entrance. Videographic data recorded using Vision

Research Phantom v9.0 high-speed camera was used to study the transient jet and complex ignition and combustion

phenomena in the main chamber. Operating at 10,000 frames per second the camera sensor is exposed for a duration of

96.75 μs every 100 μs. The reason for using a high exposure time was due to the importance of exposing the sensor to

the slightest illumination possible.

(a) Hot-jet ignition rig (b) Pre-chamber (c) Nozzle insert and

Diaphragm assembly

Figure 1: Major components of hot-jet ignition experimental rig

3. Measurements and Discussion

Diaphragm Rupture Time Analysis

In order to ensure that the jet flow is initiated immediately before or near the start of traverse of the CVC, it is necessary

to trigger the pre-chamber spark at a precisely known instant during a cycle of rotation, and to know the precise duration

from spark ignition to diaphragm rupture in the pre-chamber. This diaphragm rupture time is important in this study as it

4

defines incipient jet injection, and is the start of the ignition delay period. Diaphragm rupture time reported by Perera (

2010) for stationary pre-chamber was re-evaluated with ethylene mixture at ϕ = 1.1 while the pre-chamber was spinning

at 150 rpm. The camera was placed at a safe distance of about 5 m in front of the rotating pre-chamber, and the images

were recorded at a 480 × 480 resolution that encompasses the entire pre-chamber within the frame. Perpendicular laser

targets were used to identify the axis of the pre-chamber, as the experiments were performed in dark conditions. The

video was captured at the frame rate of 6400 frames per second (limited by the set resolution) with the lens aperture set

to an f-number of 1:1.8 and the exposure time of 153 μs. The diaphragm rupture time is defined as the elapsed time

between the trigger pulse that corresponds to initiation of spark in the pre-chamber and the first appearance of luminous

zone in the captured images. A series of tests were performed and a mean value of the diaphragm rupture time was found

as 15.4 ms with a standard deviation of ±0.3ms. This corresponds to 13.2 degrees of rotation at 150 rpm, and 69.1

degrees of rotation at 750 rpm.

Detection of Ignition and Ignition Delay Time

Ignition of a combustible mixture has been defined in various ways by different researchers. Radiation of visible light

was observed as ignition in other hot jet ignition research conducted by Wolfhard (1958) , Vanpée and Wolfhard (1959),

Fink and Vanpée (1975) and Cato and Kuchta (1966). Among various experiments, the definition of ignition delay time

varies as much as the definition of ignition. Some of the variations are due to the measuring equipment and methods

used, and some due to the physical variations of the experimental facility. In shock-ignition experiments the ignition time

is measured from the instant when the shock wave reflects at the closed end of the driven section to the instant when

combustible mixture appears to ignite, with shock arrival measured by pressure transducers (Baker and Skinner, 1972,

Tarzhanov et al., 2006) and ignition determined using photomultipliers (Baker and Skinner, 1972, Davidson and Hanson,

2004) or the emission of specific species (Davidson and Hanson, 2004, Tarzhanov et al., 2006). Radiation from

combustion products including the species and radicals from oxidation of methane, ethylene, and propane in shock tube

experiments is emitted in the wavelength range of 185-800 nm. Soot radiation is emitted in the visible and infrared

wavelength ranges (Grosshandler, 1993, Baukal, 2000). For the current study ignition delay time is defined as elapsed

time between the diaphragm rupture event and the detection of luminous region having intensity above a specified

threshold limit in main chamber (Perera, 2010). The high-speed camera captures luminosity emitted in the wavelength

range of 400-1000 nm which corresponds to emissions by soot radiation during the combustion process.

Image Processing

Video images of the jet in the main CVC chamber taken through a side window were captured at a resolution of 1632 ×

104 at frame rate of 10,000 per second (time interval between frames is 100 μs) with the lens aperture set to an f-number

of 1:1.8 and the exposure time of 96.75 μs and EDR (Extreme Dynamic Range TM

) exposure of 0 μs. EDR can be used to

adjust exposure on pixel level to reduce over-exposure at certain regions due to bright spots. Setting EDR exposure time

to 0 μs, the exposure level at the time of triggering the camera remains unchanged for the entire series of captured

images. An image processing code was written and applied using the Sobel edge-detection technique (Parker, 1997) to

provide optimal edge detection for the current study. The high-speed images from each test were analyzed frame-by-

frame for luminosity value at each pixel of the frame. The pixel values of the 8-bit monochrome camera output vary

from 0 to 255. The threshold value for identification of ignition was set at 50% of the full-scale signal. The threshold

value was chosen due to increased chemical activity for traversing cases near the nozzle region. The pixel luminosity

value, number of pixels above the threshold per frame, and the location of pixels above the threshold were tracked on

each frame. The number of pixels and growth of the pixel area is tracked for subsequent frames to detect ignition.

Two cases of jet ignition with significantly different jet structure and behavior will be presented first to illustrate the

range of ignition activity. The original grey scale images as captured using the high-speed camera and binary images

processed using the code for ignition detection is compared for each case. In Fig. 2 images are presented for methane

mixture at ϕ = 0.6 and traverse time of 40.6 ms. Ignition zones were identified at three different locations visible in the

4.8 ms frame of the processed image. It is observed that these ignition zones continue to grow rapidly over the next 0.3

ms. This distributed source of ignition at multiple regions in the main chamber is an important characteristic of jet

ignition and has implications for the rate of combustion (Attard et al., 2012). A second set of images corresponding to

ethylene mixture at ϕ = 0.4 and a traverse time of 40.6 ms is presented in Fig. 3. the maximum luminosity observed in

the images is lower than that of methane mixture with ϕ = 0.6 (Fig. 2). The existence of ignition regions above the

threshold limit can be better discerned in the processed images compared to the original images. As the ignition delay

time is significantly different for the two fuels, the images have been presented over different time scales. The default

criterion for successful ignition was set at 50% threshold limit in this study but the outcome of successful ignition was

5

not changed even though the threshold level was increased to 80%. However, ignition delay time measurements were

made using a 10% threshold as chosen by (Perera, 2010) to enable comparison with other stationary pre-chamber cases.

Figure 4 shows high speed video images that are processed for edge detection indicating the boundary of the reaction

zone. Edge detection is required to precisely define the boundary in the original images and can be further used to extract

information on flame structures and apparent flame velocity from the captured images.

Original Images (Grey Scale) Processed Images (Binary)

Figure 3: Identification of ignition zones and subsequent growth

(fuel: ethylene, = 0.4, hot-jet traverse time 40.6 ms)

Original Images (Grey Scale) Processed Images (Binary)

Figure 2: Identification of ignition zones and subsequent growth (fuel: methane, = 0.6, hot-jet traverse

time 40.6 ms). Ignition zones are circled at t=4.8ms

6

Effect of Jet Traverse Speed on Ignition

The objective of the current study was to identify ignition locations and to examine the effect of hot-jet traverse speeds

on the ignition delay time for lean fuel-air mixtures in the CVC. The ignition delay in the CVC chamber can be affected

by several parameters. The mass injected into the CVC chamber before the ignition would be lower for low jet traverse

time. This could affect the ignition delay in the CVC chamber. However, the jet traverse speed time for the tests

discussed in this work is much higher than the ignition delay time in the CVC chamber and hence the difference in

injected mass would not have any significant effect for speeds considered. The jet penetration and the fluid dynamics of

mixing are different for different initial jet locations. For low traverse speed (40.5ms traverse time), the jet remains

attached to the wall of the CVC chamber and later impinges on the lower wall producing vortices.

A range of ignition delay times between 0.6 and 4.8 ms was observed. Methane-air mixtures exhibited the highest

ignition delay time while ethylene mixtures had the lowest ignition delay time. Hydrogen-enriched methane mixtures

shows ignition delay times that falls between the other two fuels. For mixtures with ϕ = 0.4, ignition onset was identified

close to the nozzle location irrespective of the type of fuel accompanied with longer ignition delay times.

The ignition delay time obtained from the experimental study is listed in Table 1 for various fuels. The data of Perera

(2010) is included for convenience, but may not be directly compared due to differences in mass injection rates and

injection conditions. It is generally observed that faster jet traverse (in 8.1 ms) results in more rapid ignition than the

slower jet traverse (40.6 ms). The ignition jet images for ethylene mixture (ϕ = 0.8) for the two traverse speeds are

presented in Fig. 5. It is observed that for fixed equivalence ratio the location of the ignition zones for the two cases

depends upon the jet penetration at different traverse speeds. A similar trend has been observed for all the fuels

considered in the study. Due to the extended ignition delay time for ϕ = 0.4 and increased exothermic activity along the

Original Images (Grey Scale) Processed Images (Binary)

Figure 4: Flame edge detection by image processing for hydrogen-enriched methane mixture, = 0.8 for

pre-chamber traverse time of 40.6 ms

7

jet for ethylene air mixtures the images in Fig. 6 reveals the nature of jet penetration as it traverses through the main

chamber at different speeds. It can be observed that 40.6 ms, the jet barely moves from its initial location and behaves

much like a stationary offset jet, which initially travels along the wall and later impinges on the bottom wall. At a fixed

time (e.g., t = 0.5 ms), the jet traversing at high speed penetrates twice as far as the jet traversing at low speed.

The ignition zones for pure methane and hydrogen-enriched methane have been compared for mixture ratio of ϕ = 0.8

and jet traversing in 40.6 ms in Fig. 7. It is noticed from figure that for both the fuels, the ignition zone tends to move

towards the bottom wall of the CVC chamber, where ignition occurs. For the case of methane mixture with ϕ = 0.6 and

traverse duration of 40.6, ignition region is located at the center-axis of the chamber (Fig. 2). Fig. 7 also indicates that the

flame propagation is faster in the hydrogen-enriched methane mixture as compared to pure methane-air combustion. The

results corresponding to pure methane and hydrogen-enriched methane mixture (ϕ = 0.8 and spin rate at 750 rpm) are

Table 1: Measured Ignition Delay Time

Fuel Equivalence

Ratio,

Ignition Delay Time (ms)

Slower traverse

(40.6 ms)

Faster traverse

(8.1 ms)

Stationary

(Perera 2010)

C2H4

0.4 1.8 0.6 1.9

0.6 1.8 0.9 0.7

0.8 1.9 1.9 0.6

CH4

0.4 3.2 2.8 4.4

0.6 4.8 2.4 3.0

0.8 4.1 2.6 2.8

H2-CH4

0.4 2.4 2.3 -

0.6 2.6 1.9 -

0.8 3.6 2.3 -

Figure 5: Ignition for ethylene, =0.8 mixtures, for hot-jet traverse time of

40.6 ms (left) and 8.1 ms (right). Red border indicates frame of ignition detection.

8

presented in Fig. 8. It is observed that the location of the ignition zone was consistent for pure methane and hydrogen

blended methane mixtures for fixed equivalence ratio and traverse speed, but the ignition delay time was lower

Figure 6: Traversing jet penetration before ignition for ethylene =0.4 for hot-jet traverse time of 40.6

ms (left) and 8.1 ms (right)

Figure 7: Traversing jet ignition of methane =0.8 (left) and hydrogen-enriched methane =0.8 (right) for

hot-jet traverse time of 40.6ms

9

Shock-Flame Interaction

In the confined volume of the CVC, the injection of the hot jet and rapid combustion gives rise pressure waves. The

resulting shock wave reflects at the opposite end of the chamber and returns to interact with the incipient flame. The

process of shock interaction with a propagating premixed flame is encountered in various physical processes ranging

from deflagration to detonation transition to supernovas. The understanding of shock-flame interaction is essential for

promoting faster combustion reactions in novel combustion devices such as steady or pulsed detonation engines

(Kailasanath, 2003, Akbari and Nalim, 2006) or wave rotor combustors (Nalim and Pekkan, 2003, Nalim, 1999). In

wave-rotor combustor the flame is expected to be within the corrugated flame regime (Warnatz et al., 1996), where the

major effect is produced by the flame area increase following the shock-flame interaction. Shock-flame interaction leads

to a significant increase in total energy release rates; as a consequence, the overall reaction rate increases due to the

baroclinic vorticity production (Kilchyk, 2009, Kilchyk et al., 2011). The high-speed video images from experiments

corresponding to combustion of ϕ = 0.8 ethylene mixture and jet traversing speed of (8.1 ms) is presented in Fig. 9. Due

to the interaction of the propagating flame with the shock wave returning from the main CVC chamber end, the flame

front deforms into a mushroom like shape. Numerical analysis of the shock-flame interaction and the reaction rate

increase for different traverse speeds has been carried out in another work (Karimi et al., 2013). Predicted results indicate

that shock-flame interaction causes significant increase in reaction rates. The reaction rate increase has been observed to

be caused by both flame length/surface increase due to deformation and kinetic amplification (Kilchyk, 2009).

Figure 8: Comparison of ignition and subsequent flame propagation for methane =0.8 (left) and

hydrogen-enriched methane =0.8 (right) for hot-jet traverse time of 8.1 ms

10

3. Conclusions

Traversing jet ignition tests were carried out for ethylene, methane and hydrogen-enriched methane mixtures with

the jet traversing the end of a long constant-volume combustor in 40.6 ms and 8.1 ms. The fuel-air mixtures were

successfully ignited for a range of lean equivalence ratios (ϕ = 0.4, 0.6 and 0.8) in the CVC chamber, while keeping the

pre-chamber fuel-air ratio fixed (ethylene at ϕ = 1.1). It is observed that for the same fuel-air mixtures, the lean

ignitability limit measured using a hot-jet is well below the ignitibility limit by spark ignition methods. By using a

quiescent mixture in the main CVC chamber, the fluid dynamic behavior of the traversing jet was observed clearly. For

the fuels and equivalence ratios considered in the study, the ignition delay time generally was lower for the faster

traverse jet (8.1 ms) compared to the slower traverse jet (40.6 ms), the latter behaving essentially as an offset stationary

jet.. For methane at ϕ = 0.8, the difference was as much as 2.5ms. The observation supports the fact that ignition was

controlled by the traversing nature of the jet across the chamber. Hydrogen enrichment of the methane mixture

expectedly decreased the ignition delay times and increased the consumption rate of the CVC mixture compared to pure

methane. Criterion for detection of ignition and measurement of ignition delay times can further be improved by

gathering dynamic pressure data in the main CVC chamber. The flame deformation caused by shock-flame interaction is

also clearly observed from the high speed video images.

Acknowledgement:

The authors would like to thank Mr. Kevin Murphy for fabrication and assembly of the main chamber, fueling system,

and other rig components, Mr. Indika Perera for design of rig components and instrumentation, Mr. Michael David for

assessing and programming the edge detection technique, and the following for improvements to the rig electrical

systems and safety:.Mr. Kenneth Lee, Mr Kyung Hoon Bang and Ms. Golnaz Mortazavi.

Figure 9: shock flame interaction observed for ethylene =0.8 (right) for hot-jet traverse time of

8.1 ms.

11

References

Akbari, P. & Nalim, M. R. Analysis of Flow Processes in Detonative Wave Rotors and Pulse Detonation Engines .

AIAA 44th Aerospace Sciences Meeting, AIAA Paper 2006-1236, 2006.

Attard, W. P., Toulson, E., Huisjen, A., Chen, X., Zhu, G. & Schock, H. Spark Ignition and Pre-Chamber Turbulent Jet

Ignition Combustion Visualization. SAE Paper 2012-01-0823, 2012.

Baker, J. A. & Skinner, G. B. Shock-Tube Studies on the Ignition of Ethylene-Oxygen-Argon Mixtures. Combustion and

Flame, 19, 347-350, 1972.

Baukal, C. E. Heat Transfer Modes. Heat Transfer in Industrial Combustion, CRC press, 2000.

Bilgin, M. Stationary and Rotating Hot Jet Ignition and Flame Propagation in a Premixed Cell. Ph D Ph.D. Dissertation,

University of Washington, 1998.

Bilgin, M., Keller, J. J. & Breidenthal, R. E. Ignition and Flame Propagation with Rotating Hot jets in a Simulated Wave

Engine Test Cell. AIAA Paper 98- 3399, 1998.

Boretti, A. A. & Watson, H. C. The Lean Burn Direct Injection Jet Ignition Gas Engine. International Journal of

Hydrogen Energy, 3 4, 7835 –7841, 2009.

Cato, R. J. & Kuchta, J. M. Hot Gas Ignition Temperatures of Hydrocarbon Fuel Vapor-Air Mixtures. Bureau of Mines,

AD0643518, 1966.

Dagaut, P. & Nicolle, A. Experimental and Detailed Kinetic Modeling Study of Hydrogen-Enriched Natural Gas Blend

Oxidation Over Extended Temperature And Equivalence Ratio Ranges. Proceedings of the Combustion Institute 30,

2631–2638, 2005.

Davidson, F. D. & Hanson, K. R. Interpreting Shock Tube Ignition Data. International Journal of Chemical Kinetics, 36,

510-523, 2004.

Dober, G. G. & Watson, H. C. Modelling the Flame Enhancement of a HAJI Equipped Spark Ignition Engine. IPC-10

Proceedings Paper Number 99091, 1999.

Fink, Z. J. & Vanpée, M. Overall Kinetics of Hot Gas Ignition. Combustion Science and Technology, 11, 1975, 229-238.

Grosshandler, W. L. Radcal: A Narrow-Band Model for Radiation Calculations in a Combustion Environment. NIST

Technical Note 1402, 1993.

Ju, Y. & T, N. Ignition Simulation of Methane/Hydrogen Mixtures in a Supersonic Mixing Layer, Combustion and

Flame, 102, 462–470, 1995.

Kailasanath, K. Recent Developments in the Research on Pulse Detonation Engines. AIAA Journal, 41, 2003, 145-159.

Kilchyk, V. Pressure-Wave Amplification of Flame Area in Wave Rotor Channels. PhD Thesis, Purdue University,

2009.

Kilchyk, V., Nalim, M. & Merkle, C. Laminar Premixed Flame Fuel Consumption Rate Modulation by Shocks and

Expansion Waves. Combustion and Flame, 158, 140-1148, 2011.

Lieberman, D. H., Parkin, K. L. & Shepherd, J. E. Detonation Initiation by a Hot Turbulent Jet for use in Pulse

Detonation Engines, AIAA 38th Joint Propulsion Conference, Paper 2002-3909, 2002.

Matsutomi, Y., Meyer, S. E., Wijeyakulasuriya, S. D., Izzy, Z., Nalim, M. R., Shimo, M., Kowalkowski, M. & Snyder,

P. H. Experimental Investigation on the Wave Rotor Constant Volume Combustor, AIAA 46th Joint Propulsion

Conference, AIAA Paper 2010-7043, 2010.

12

Maxson, J. A., Hensinger, D. M., Hom, K. & Oppenheim, A. K. Performance of Multiple Stream Pulsed Jet Combustion

Systems. SAE International Congress and Exposition, SAE Paper 910565, 1991.

Mayinger, F., Jordan, M., Eder, A., Zaslonko, I. S., Karpov, V. P. & Frolov, S. M. Flame-Jet Ignition of fuel-Air

Mixtures. Experimental Findings and Modeling, 17th International Colloquium on the Dynamics of Explosions and

Reactive Systems, 1999.

Menter, F. R. Two-equation eddy-viscosity turbulence modeling for engineering applications. AIAA Journal, 32, 1598-

1605, 1994.

Nalim, M. R. Assessment of Combustion Modes for Internal Combustion Wave Rotors ASME J. of Engineering for Gas

Turbines & Power, 121, 265-271, 1999.

Nalim, M. R. & Pekkan, K. Internal Combustion Wave Rotors for Gas Turbine Engine Enhancement. ASME Paper

IGTC-2003-FR-303, 2003.

Parker, J. R. Algorithms for Image Processing and Computer Vision. New York, John Wiley & Sons, Inc, 23-29, 1997.

Perera, I. U. Experimental Investigation into Combustion Torch Jet Ignition of Methane-Air, Ethylene-Air, and Propane-

Air Mixtures. MS Thesis, Purdue University, 2010.

Ren, J. Y., Qin, W., Egolfopoulos, F. N., Mak, H. & Tsotsis, T. T. Methane reforming and its potential efect on the

efficiency and pollutant emissions of lean methane-air combustion, Chemical Engineering Science, 56, 1541-1549, 2001.

Tarzhanov, V. I., Telichko, I. V., Vil’danov, V. G., Sdobnov, V. I., Makarov, A. E., S L Mukhin, Koretskii, I. G.,

Ogarkov, V. A., Vlasov, V. V., Zinchenko, A. D., Vorob’ev, A. V., Grachev, A. N., Matkin, V. A. & Potashnikov, V. A.

Detonation of Propane-Air Mixtures under Injection of Hot Detonation Products. Combustion Explosion and Shock

Waves, 42, 336-345, 2006.

Toulson, E., Schock, H. J. & Attard, W. P. A Review of Pre-Chamber Initiated Jet Ignition Combustion Systems. SAE

Paper 2263, 2010.

Vanpée, M. & Wolfhard, G. H. Comparison between Hot Gas Ignition and Limit Flame Temperatures. ARS JOURNAL,

29, 517-519, 1959.

Warnatz, J., Maas, U. & Dibble, R. W. Combustion: Physical and Chemical Fundamentals, Modelling and Simulation,

Experiments, Pollutant Formation. Springer-Verlag, 1996.

Wijeyakulasuriya, S. D., Perera, I. U. & Nalim, M. R. Mixing and Ignition Potential of Transient Confined Turbulent Jet

in a Wave-Rotor Constant-Volume Combustor, AIAA 46th Joint Propulsion Conference, AIAA 2010-7042, 2010.

Wolfhard, H. G. The Ignition of Combustible Mixtures by Hot Gases. Jet Propulsion, 28, 798-804, 1958.