C OMPARISON WITH N UMERICAL S IMULATION

The corresponding simulation employs two parameters, inflow velocity and fuel ejection area (S=180°, 270°, and 360°), to elucidate flame lift-off. The result shows that a greater fuel ejection area yields a

wider range of flame lift-off inflow velocity. The experiment determines the same results (Figs. 33 and 38). Figure 41 (Uin = 1.03m/sec, v_w = 1.12cm/sec, and S=360°) shows oscillation of the lift-off flame, with a lift-off height of around 0.55D. As time passes, the wake flame and the lift-off flame appear in turn. Similar results can be clarified in Fig. 41, which is the corresponding transient simulation of Fig.

41. Besides, the oscillation period of the numerical calculation is 0.01sec, which is shorter than that, 0.11sec, experimentally observed.

The lift-off height in Fig. 42 is approximately 0.15D, lower than the experimental observation. The predicted phenomena are qualitatively consistent with the observations, but quantitative discrepancies exist between the predicted and observed flame lift-off heights and oscillation frequencies, because the simulation assumes laminar flow, whereas the lift-off flame exhibits turbulent characteristics.

CHAPTER 5 CONCLUSIONS

This study modifies the combustion model developed by Chen and Weng (1990), using a four-step chemical reaction mechanism instead of one-step overall kinetics and a finer distribution of grid cells to catch up the flame lift-off phenomena over a Tsuji burner. Besides, the corresponding experimental apparatus consists of a wind tunnel and a porous sintered cylindrical burner. The wind tunnel is open-circuit and orientated vertically upwards. It is designed to provide a laminar, uniform oxidizing flow over the porous cylindrical burner, from the surface of which fuel is ejected. The burner is designed with inner and outer parts. A digital video, fixed at an appropriate position, records the various flame profiles. The parameters of interest are the inflow air velocity (Uin) and fuel-ejection area (S). This report emphasizes occurrence of the lift-off flame, which was unidentified in Chen and Weng (1990) but observed in the experiments of Wang (1998).

The modified combustion model is validated first by comparing the predicted results with the corresponding measurements of Tsuji (1982) and the simulation results of Chen and Weng (1990). Then, it is compared with the related measurements and calculations of Dreier et al.

(1986). Generally, the present simulation yields a much better prediction than that of Chen and Weng (1990), implying that the prediction obtained using a four-step chemical mechanism is indeed better than that obtained using a one-step overall chemical mechanism.

Also, the proposed combustion model can reproduce the data measured

experimentally by Dreier et al. (1986): the agreement is much better than that of their own numerical results.

In the simulation, as the inflow velocity increases, the envelope, wake, lift-off, and wake flame appear in that order before the flame is completely extinguished. The two wake flames have similar structures but different transformation processes: one is transformation from the envelope flame and the other is transformation from lift-off flame.

Envelope flame, which is diffusion flame, exists when the inflow velocity is less than 0.9m/sec. Above that velocity, the flame front breaks away from the front stagnation streamline and retreats along the surface until a certain condition is met that it can be stabilized on the rear part of the cylinder. The flame then becomes a wake flame, whose flame front shows the feature of a premixed flame and which is positioned ahead of the rear stagnation point.

When the inflow velocity increases further to 1.05m/sec, the wake flame is abruptly transformed into a lift-off flame, whose flame front is not attached to but far from the rear surface of the cylindrical burner.

The maximum lift-off height is found to be 1.7D when the inflow velocity (U_in) is 1.05 m/sec. This height is maintained up to U_in = 1.09 m/sec. Then, the height declines gradually as the inflow velocity is increased. No recirculation flow occurs behind the cylindrical burner for these lift-off flames, unlike for the envelope and wake flames. When U_in reaches 1.13 m/sec, the vortex starts to reappear. However, the flame front remains behind the rear stagnation point with a lift-off height of 0.6D. The transition process from the lift-off to the wake flame occurs from 1.13 to 1.15 m/sec. The flame during the transition exhibits

some of the features of both flames. Finally, when the inflow velocity reaches 1.16 m/sec, the wake flame fully reappears. Eventually, the flame is completely extinguished at U_in > 2.12 m/sec. The entire process from the envelope to wake, then lift-off, and back to wake flame is verified by this experimental observation, made using a flow visualization technique.

In the experiment, the flame behaviors are separated into four regions (I-IV) at S=180°, whereas for S=360°, they are categorized into two regions (V and VI). In regions I, II, and III, at an initial inflow velocity of 0.41 m/s, an envelope diffusion flame is generated around the porous cylinder. For a fixed fuel ejection velocity, the flame stand-off distance decreases and the flame length increases as the inflow velocity increases, due to the enhanced flame stretch effect. The flame thickness of the envelope flame is found to be almost constant. Increasing the inflow velocity to a critical value causes the envelope flame to be blown-off and transformed into a wake flame, whose flame front exhibits premixed flame characteristics and whose downstream part exhibits features of a diffusion flame. The flame length is shortened and flame attached angle increased as inflow velocity increases. The major difference among these three regions is the color of the flames.

In region IV, the envelope flame turns into a wake flame as the inflow velocity increases. Maintaining the same inflow velocity for a short period allows the flame to be lifted away from the rear surface of the cylinder. As the inflow velocity increases, the attached angle of the wake flame increases, moving the two flame fronts closer together. The high pressures generated at these two flame fronts depress the vortices,

and eventually destroy them. Then, the lift-off flame appears. It can maintain a lift-off height over a porous cylinder because of the balance between the speed of the flame toward the rear surface of the cylinder and the velocity of the local fresh mixture in air flow direction. In the experiment, some back-and-forth oscillations of the lift-off flame were observed because the local balance position changes continually. When the inflow velocity exceeds the critical value, the vortices are again present behind the cylinder, and the flame lift-off height gradually declines. Finally, the lift-off flame drops back to a wake flame (late wake flame) again.

In full cylinder fuel-ejection (S=360°), the transformation from the envelope to the wake flame in region V is similar to that in front half cylinder fuel-ejection (S=180°). However, fuel downstream of the envelope flame is not expected to mix well with air, which results in fuel-rich burning in the downstream tail. As inflow velocity increases further, the lift-off flame is generated. The base of lift-off flame can stay longer above the cylinder surface than that in region IV.

In region VI, no wake flame is observed between the envelope and lift-off flames. Another difference between region V and VI is that the critical velocity to transform into the lift-off flame in region V is decreased as the fuel ejection velocity increases, but it follows the opposite trend in region VI. The fuel supply can be directly ejected into the lift-off flame in the case of full cylinder fuel-ejection; consequently, the survival domain is much greater than that of the front half cylinder fuel-ejection.

The formation of a lift-off flame is described briefly. When the inflow velocity exceeds the local flame speed, the wake flame front must

retreat downstream to a new stable position. However, it cannot move inward into the recirculation zone since this zone is full of combustion products. Consequently, the flame front must then leave the surface and shift further downstream. At this moment, no recirculation flow exists.

A reaction-frozen zone now exists between the burner and the flame front.

When the inflow velocity increases, more oxidizer is supplied to mix with the un-reacted fuel in the reaction-frozen zone to form a flammable mixture in front of the flame front. Therefore, the flame front can propagate upstream with a higher flame speed. The reduction of lift-off height, or flashback, is not so abrupt because it results from a stronger opposed flow. The flashback process continues as the inflow velocity increases until the lift-off flame front reaches the rear surface of the burner to form the wake flame again.

Finally, some suggestions are offered for future extensions of this study. The flow pattern of the flame could be observed by introducing particles (magnesium oxide particles) into the uniform air stream and incorporating a LASER system. This approach would help to visualize the flow field behind the cylinder and confirm the controlling mechanism described above. The temperature distribution should also be measured.

One more porous cylinder would be inserted into the test section to study flame interference/interaction phenomena. The flame strength could be quantified by heat release rate, according to the oxygen consumption calorimetry principle.

References

Andrews, J. R. and Biblarz, O. (1981) Temperature Dependence of Gas Properties in Polynomial Form, NPS67-81-001, Naval Postgraduate School, Monterey, California.

Bilger, R. W., Esler, M. B., and Starner, S. H. (1991) On Reduced Mechanisms for Methane-air Combustion, in Reduced Kinetic Mechanisms and Asymptotic Approximations for Methane-air Flames, Chapter 5, p. 86, Springer-Verlag, Berlin.

Chen, C.-H. and Weng, F.-B. (1990) Flame Stabilization and Blowoff Over a Porous Cylinder, Combustion Science and Technology, 73, 427-446.

Chen, W.-S. (1993) Interphase Exchange Phenomena of a Convective Liquid Droplet (Canonical Droplet Theory), Ph. D. Dissertation, National Cheng Kung University, Taiwan.

Chiu, H. H. and Huang, J. S. (1996) Multiple-state Phenomena and Hysteresis of a Combusting Isolated Droplet, Atomization and Sprays, 6, 1-26.

Dixon-Lewis, G., David, T., Gaskell, P. H., Fukutani, S., Jinno, H., Miller, J. A., Kee, R. J., Smooke, M. D., Peters, N., Effelsberg, E., Warnatz, J., and Behrendt, F. (1984) Calculation of the Structure and Extinction Limit of a Methane-Air Counterflow Diffusion Flame in the Forward Stagnation Region of a Porous Cylinder, Twentieth Symposium (International) on Combustion, The Combustion Institute, Pittsburgh, p.

1893-1904.

Dreier, T., Lange, B., Wolfrum, J., Zahn, M., Behrendt, F., and Warnatz, J. (1986) Comparison of CARS Measurements and Calculations of the Structure of Laminar Methane-Air Counterflow Diffusion Flames, Ber.

Bunsenges. Phys. Chem., 90, 1010-1015.

Figliola, R. S., and Beasley, D. E. (1995) Theory and Design for Mechanical Measurements, 2^nd Ed., John Wiley and Sons, Canada.

Fox, R. W., and McDonald, A. T. (1994) Introduction to Fluid Mechanics, John Wiley and Sons, Canada.

Gollahalli, S. R. and Brzustowski, T. A. (1973) Experimental Studies on the Flame Structure in the Wake of a Burning Droplet, Fourteenth Symposium (International) on Combustion, The Combustion Institute, Pittsburgh, p. 1333.

Holman, J. P. (1989) Experimental Methods for Engineers, 5th Ed., McGraw-Hill, New York.

Huang, J.-S. (1995) Canonical and Renormalized Theory of Droplets:

States, Structures and Laws of Isolated and Interacting Droplets, Ph. D.

Dissertation, National Cheng Kung University, Taiwan.

Huang, J.-S. and Chiu, H.-H. (1997) Multistate Behavior of a Droplet in Dilute Sprays, Atomization and Sprays, 7, 479-506.

Jiang, T. L., Chen, W. S., Tsai, M. J., and Chiu, H. H. (1995) A Numerical Investigation of Multiple Flame Configurations in Convective Droplet Gasification, Combustion and Flame, 103, 221-238.

Kalghatgi, G. T. (1984) Lift-off Heights and Visible Lengths of Vertical Turbulent Jet Diffusion Flames in Still Air, Combustion Science and

Technology, 41, 17-29.

Kee, R. J., Rupley, F. M., Miller, J. A., Coltrin, M. E., Grcar, J. F., Meeks, E., Moffat, H. K., Lutz, A. E., Dixon-Lewis, G., Smooke, M. D., Warnatz, J., Evans, G. H., Larson, R. S., Mitchell, R. E., Petzold, L. R., Reynolds, W. C., Caracotsios, M., Stewart, W. E., and Glarborg, P. (1999A) Chemkin-III: A Software Package for the Analysis of Gas -phase Chemical and Plasma Kinetics, Reaction Design, San Diego.

Kee, R. J., Rupley, F. M., Miller, J. A., Coltrin, M. E., Grcar, J. F., Meeks, E., Moffat, H. K., Lutz, A. E., Dixon-Lewis, G., Smooke, M. D., Warnatz, J., Evans, G. H., Larson, R. S., Mitchell, R. E., Petzold, L. R., Reynolds, W. C., Caracotsios, M., Stewart, W. E., and Glarborg, P. (1999B) The Chemkin Thermodynamic Database, Reaction Design, San Diego.

Kline, S. J., and Mcclintock, F. A. (1953) Describing Uncertainties in Single-Sample Experiments, Mechanical Engineering, 75, 3-8.

Lee, B. J., Kim, J. S., and Chung, S. H. (1994) Effect of Dilution on the Liftoff of Non-premixed Jet Flames, Twenty-fifth Symposium (International) on Combustion, The Combustion Institute, Pittsburgh, p.

1175.

Miller, J. A., Kee, R. J., Smooke, M. D., and Grcar, J. F. (1984) The Combustion of the Structure and Extinction Limit of a Methane-Air Stagnation Point Diffusion Flame, Western States Section, The Combustion Institute, Paper WSS/CI 84-10.

Moffat, R. J. (1982) Contributions to the Theory of Single-Sample Uncertainty Analysis, Journal of Fluid Engineering, 104, 50-260.

Nakabe, K., Mcgrattan, K. B., Kashiwagi, T., Baum, H. R., Yamashita, H., and Kushida, G. (1994) Ignition and Transition to Flame Spread over

a Thermally Thin Cellulosic Sheet in a Microgravity Environment,

Combustion and Flame, 98, 361-374.

Paczko, G., Lefdal, P. M., and Peters, N. (1986) Reduced Reaction Schemes for Methane, Methanol and Propane Flames, Twenty-first Symposium (International) on Combustion, The Combustion Institute, Pittsburgh, p. 739-748.

Peters, N. and Kee, R. J. (1987) The Computation of Stretched Laminar Methane-air Diffusion Flames Using a Reduced Four-step Mechanism,

Combustion and Flame, 68, 17-29.

Rogg, B. (1991) Sensitivity Analysis of Laminar Premixed CH₄-air Flames Using Full and Reduced Kinetic Mechanisms, in Reduced Kinetic Mechanisms and Asymptotic Approximations for Methane-air Flames, Chapter 8, p. 159, Springer-Verlag, Berlin.

Rogg, B. (1993) Systematically Reduced Kinetic Mechanisms:

Sensitivity Analysis, in Dynamics of Gaseous Combustion, p. 202, American Institute of Aeronautics and Astronautics, Washington D.C..

Seshadri, K. and Peters, N. (1990) The Inner Structure of Methane-air Flames, Combustion and Flame, 81, 96-118.

Sick, V., Arnold, A., Diebel, E., Dreier, T., Ketterle, W., Lange, B., Wolfrum, J., Thiele, K. U., Behrendt, F., and Warnatz, J. (1990) Two-dimensional Laser Diagnostics and Modeling of Counterflow Diffusion Flames, Twenty-third Symposium (International) on Combustion, The Combustion Institute, Pittsburgh, p. 495-501.

Smooke, M. D. and Giovangigli, V. (1991) Formulation of the Premixed and Nonpremixed Test Problems, in Reduced Kinetic Mechanisms and Asymptotic Approximations for Methane-air Flames, Chapter 1, p. 1, Springer-Verlag, Berlin.

Sung, C. J., Liu, J. B., and Law, C. K. (1995) Structural Response of Counterflow Diffusion Flames to Strain Rate Variations, Combustion and

Flame, 102, 481-492.

Tsuji, H. and Yamaoka, I. (1967) The Counterflow Diffusion Flame in the Forward Stagnation Region of a Porous Cylinder, Eleventh Symposium (International) on Combustion, The Combustion Institute, Pittsburgh, p. 979.

Tsuji, H. and Yamaoka, I. (1969) The Structure of Counterflow Diffusion Flame in the Stagnation Region of a Porous Cylinder, Twelfth Symposium (International) on Combustion, The Combustion Institute, Pittsburgh, p. 997.

Tsuji, H. and Yamaoka, I. (1971) Structure Analysis of Counterflow Diffusion Flames in the Forward Stagnation Region of a Porous Cylinder, Thirteenth Symposium (International) on Combustion, p. 723.

Tsuji, H. (1982) Counterflow Diffusion Flame, Progress in Energy and

Combustion Science, 8, 93.

Vanquickenborne, L. and Van Tiggelen, A. (1966) The Stabilization Mechanism of Lifted Diffusion Flames, Combustion and Flame, 10, 59-69.

Wang, J.-Y. (1998) The Effect of Interaction Between Two Porous Flame Mechanism with Nitrogen Ejector, M. S. Thesis, National Taiwan University, Taiwan.

Warnatz, J. (1984) Rate Coefficients in the C/H/O System, in Combustion Chemistry, Chapter 5, p. 197, Springer-Verlag, New York.

Weng, F.-B. (1989) Diffusion Flame Stabilization and Blowoff over a Porous Cylinder, M. S. Thesis, National Chiao Tung University, Taiwan.

Wohl, K., Gazley, C., and Kapp, N. (1949) Diffusion Flames, Third Symposium on Combustion, Flame, and Explosion Phenomena, p. 288.

Yang, J. C., Donnelly, M. K., Prive, N. C., and Grosshandler, W. L.

(1999) Dispersed Liquid Agent Fire Suppression Screen Apparatus, NISTIR 6319, National Institute of Standards and Technology.

Zhao, J., Li, C., and Yang, S. (1997) Measurements of Temperature Distribution in a Counterflow Diffusion Flame by USED CARS, Laser

Technology, 21, 4, 218.

Table I

Transformed governing equations

)

Table II

Rate coefficient parameters for methane oxidation reactions

No Reaction

_{B n}

E

1 CH4+H→CH3+H2 2.2×10 3.0 36676.4

2 CH4+OH→CH3+H2O 1.6×10³ 2.1 10257.7

6 CHO+H→CO+H2 2.0×10¹¹ 0.0 0.0

7 CHO+M_T→CO+H+MT 7.14×10¹¹ 0.0 70338.2

8 CHO+O₂→CO+HO2 3.0×10⁹ 0.0 0.0

9 CO+OH CO₂+H 4.4×10³ 1.5 -3098.2

10 H+O₂ OH+O 1.2×10¹⁴ -0.91 69165.9

14 H+O₂+M_T→HO2+M_T 2.0×10¹⁵ -0.80 0.0 15 H+OH+M_T→H2O+M_T 2.15×10¹⁹ -2.0 0.0

16 H+HO2→2OH 1.5×10¹¹ 0.0 4186.8

17 H+HO2→H2+O₂ 2.5×10¹⁰ 0.0 2888.9

18 OH+HO2→H2O+O₂ 2.0×10¹⁰ 0.0 0.0





− °

= R T

T E B k ⁿexp

Table III Grid test results

The peak temperature in the whole computational domain (unit: K)

62×27 164×85 218×115 402×221 864×501 Case B1 1989 1992 1948 1947 1947 Case B2 1899 1918 1902 1899 1900 Case B3 1870 1877 1895 1894 1895 Case B4 1872 1861 1888 1890 1885 Case B5 1859 1862 1867 1867 1867 Case B6 1858 1850 1863 1854 1853 Case B7 1884 1844 1856 1851 1850 Case B8 1877 1822 1849 1851 1848 Case B9 1875 1822 1837 1843 1844

Table IV

Summary of uncertainty analysis

Parameters Uncertainty

D

i,

D

_o,

L

,

a

,

b

±0.5 mm

A

±1.267%

Burner

A

±2.084%

ν

±0.09%

ρ

air ±0.201%

T

±0.5 ℃

P

±1 torr

Q

air ±2.2%

Q

fuel ±1%

V

air ±2.54%

V

fuel ±2.31%

R

e ±3.04%

Table V

The experimental repeatibility

Fuel

Table VI Property values

Name Symbol Value Unit

Ambient Temperature

T

a 300 K

Reference Temperature

T^* 1250 K

Density (reference) ρ^* 0.2835 Kg/m³

Kinematic Viscosity (reference) υ^* 1.69E-4 m²/sec Thermal Diffusivity (reference) α^* 2.36E-4 m²/sec Specific Heat (reference) Cp

* 1.351 KJ/(Kg×K)

Cylinder surface temperature

T

w 400 K

Oxidizer velocity U_in variable m/sec

Fuel-ejection velocity v_w 0.065 m/sec

Cylinder radius R 0.015 m

Air molecular weight (reference) M_air 28.97 Kg/Kmole Atmospheric pressure at STP condition

P

rc 101325 Pa

Table VII

Comparison of inflow velocity regions for various flame appearances (unit: m/sec)

Present Study Chen and Weng (1990)

Envelope flame < 0.9 < 1.07

Side flame



1.07~1.30

Wake flame 0.9~1.04 1.31~1.99

Lift-off flame 1.05~1.15



Late Wake flame 1.16~2.12



Extinction 2.13 2.00

Table VIII

The surviving range of lift-off flame

The inflow air velocity that caused the wake flame to lift (Unit:

m/sec)

The inflow air velocity that caused the lift-off flame to drop back (Unit: m/sec)

Front half side cylinder surface fuel-ejection

case (S=180°)

1.05 1.14

Front three quarter cylinder surface fuel-ejection case

(S=270°)

1.03 1.31

Full cylinder surface fuel-ejection case

(S=360°)

1.01 1.39

Table IX

Flame lift-off height at various fuel-ejection area (U

in

=1.05 m/sec)

Flame lift-off height Front half side cylinder surface

fuel-ejection case (S=180°)

1.7D

Front three quarter cylinder surface fuel-ejection case (S=270°)

1.3D

Full cylinder surface fuel-ejection case (S=360°)

1.3D

Table X

The characteristics of each kind of flame for S=180°

(a) v

w

= 1.12 cm/s

(c) v

w

= 2.46 cm/s

Table XI

The characteristics of each kind of flame for S=360°

(a) v

w

= 1.23 cm/s

Table XII

Comparisons with Tsuji’s flame blow-off study (1982)

Present study Tsuji’s study Flame stretch rate

(sec^-1)

-fw -fw Difference (%)

141.33 0.1307 0.1565 16.49

145.33 0.1359 0.1581 14.04

149.33 0.1410 0.1597 11.71

153.33 0.1460 0.1613 9.49

160 0.1496 0.164 8.78

165.33 0.1543 0.1661 7.10

168 0.1596 0.1672 4.55

172 0.1642 0.1688 2.73

Flame stretch rate =

R U

_in

2

Nondimensional fuel-ejection rate, -f_w=

2

 Re

 



in w

U

v

,

ν R U

_in

=

Re

FIGURE 1 Schematic drawing of overall experimental system

X

FIGURE 2 Boundary conditions of the physical domain

FIGURE 3 Schema of the wind tunnel

FIGURE 4 The picture of AMCA 210-85 standard in wind tunnel

FIGURE 5 The design figure of AMCA 210-85 standard

0.00 10.00 20.00 30.00 40.00 50.00

Blower Frequency (Hz)

0.00 1.00 2.00 3.00 4.00

A ir fl o w V e lo c it y ( m /s e c )

FIGURE 6 The relation figure of blower frequency and airflow velocity

FIGURE 7 The connecting of blower and tunnel

FIGURE 8 The picture of cooling system

FIGURE 9 The pitot tube in test section

FIGURE 10 The position of pitot tube in test section

0.00 5.00 10.00 15.00 20.00

Pitot-tube Position (cm)

0.00 0.40 0.80 1.20

P re s s u re D if fe re n c e ( m m A q )

Airflow Velocity (m/sec) 0.21

0.58 1.14 1.39 1.82 2.20 2.54 3.38 3.75 4.0

FIGURE 11 Pressure difference at different position in test section

FIGURE 12 Porous sintered stainless steel cylinder

FIGURE 13 The picture of burner

FIGURE 14 Cylindrical brass rod

FIGURE 15 The digital mass flow controller

FIGURE 16 The design figure of bi-directional pitot tube

FIGURE 17 The picture of bi-directional pitot tube

FIGURE 18 The picture of O

2

analyzer

FIGURE 19 The connecting path in the pretreatment system

FIGURE 20 The picture of the pretreatment system

FIGURE 21 The measuring probes in the vent

FIGURE 2 2 Schema of instruments building

2 U in / R

-f w

1 0⁰ 1 0¹ 1 0² 1 0³ 1 0⁴

0 0 .5 1 1 .5 2

T s u ji's E xp e rim e n t s T h is S tu d y

C h e n & W e n g 's S im u la tio n s

E n ve lo p e F lam e

B lo w - o ff

FIGURE 23 Flame blow-off curves for counterflow diffusion flame in the forward

stagnation region of a porous cylinder (R=1.5cm, and the fuel is methane)

D is ta n ce F rom B urne r H e a d ( m m )

T e m p e ra tu re (K )

0 2 4 6 8

5 0 0 1 0 0 0 1 5 0 0 2 0 0 0

D re i e r e t a l. ' s E x p e ri m e n ts D re i e r e t a l. ' s ca l cu l a tio n s T h i s S t u d y

FIGURE 24 Temperature distributions through the flame front of a Tsuji burner

with R=0.02m, U

in

=0.15m/sec, and -f

w

=0.318. The solid line and its corresponding

squares are the CARS measurements of Dreier et al. (1986), the dash-dot-dot line and

its corresponding triangles are the numerical results of Dreier et al. (1986), and the

dashed line and its corresponding circles are the numerical results of the current study.

-5 0 5 1 0

- 5 0 5 1 0

FIGURE 25 Series of temperature contours and streamlines

(The orange, light green, dark green, light blue, and dark blue lines represent 1800,

1500, 1200, 900, and 600K temperature contours, respectively.)

0 . 2

0 .2

FIGURE 26 Series of methane (solid lines) and oxygen (dashed lines) mass fraction

contours

- 5 0 5 1 0

- 5 0 5 1 0

FIGURE 27 Series of 10

^-4

g/(cm

³

×sec) methane reaction rate contours

5 . 5 8 E - 0 5

2 . 7 9 E -0 5

8 . 3 7 E - 0 5

- 5 0 5 1 0

0 1 2 3 4

(B4) U

in

= 1.05m/sec (k

s

=140 sec

^-1

)

FIGURE 28 The mass fraction contours of hydrogen for case B4

(C1) Envelope flame (U

in

=1.0m/sec) (C2) Wake flame (U

in

=1.2m/sec)

(C3) Lift-off flame (U

in

=1.39m/sec) (C4) Late wake flame (U

in

=1.43m/sec)

(Night shot photograph)

FIGURE 29 The flame configurations for the experimental visualization (-f

w

=0.201)

- 5 0 5 1 0 0

1 2 3 4

(C1) U

in

= 0.75m/sec (k

s

=100 sec

^-1

)

- 5 0 5 1 0

0 1 2 3 4

(C2) U

in

= 1.0m/sec (k

s

=133.33 sec

^-1

)

- 5 0 5 1 0

0 1 2 3 4

(C3) U

in

= 1.05m/sec (k

s

=140 sec

^-1

)

- 5 0 5 1 0

0 1 2 3 4

(C4) U

in

= 1.32m/sec (k

s

=176 sec

^-1

)

FIGURE 30 Series of temperature contours and streamlines in the front three

quarter side cylinder surface fuel-ejection condition (S=270°)

- 5 0 5 1 0

FIGURE 31 Series of temperature contours and streamlines in the full cylinder

surface fuel-ejection condition (S=360°)

(a) Flame stand-off distance and flame thickness

(b) Flame attached angle (c) Flame length

(d) Flame lift-off height (H)

FIGURE 32 Definitions of flame stand-off distance, flame thickness, flame attached

angle, flame length, and flame lift-off height for each kind of flame

1.00 2.00 3.00 Fuel Ejection Velocity (cm/s)

0.00 1.00 2.00 3.00 4.00

Airflow Velocity (m/s)

Half Fuel-Ejection Area Blue Envelope to Blue Wake Blue Envelope to Yellow Wake Yellow Wake to Blue Wake Yellow Envelope to Yellow Wake Yellow Wake to Blue Wake Extinction

Lift-off Flame After Wake Flame (the same velocity) Late Wake Flame

Extinction

Wake Flame

Ι

ΙΙ ΙΙΙ IV

Envelope Flame

FIGURE 33 Various flame stabilization regions over a burner (S=180°)

FIGURE 34 Series of flame configurations as a function of inflow velocity (v

w

= 1.12cm/s and S = 180°), (a) U

in

= 0.41m/s, (b) U

in

= 0.51m/s, (c) U

in

= 0.66m/s, (d) U

in

= 1.00m/s, and (e) U

in

= 2.06 m/s

(a)

(b)

(c)

(d)

(e)

FIGURE 35 Series of flame configurations as a function of inflow velocity (v

w

= 1.23cm/s and S = 180°), (a) U

in

= 0.41m/s, (b) U

in

= 0.62m/s, (c) U

in

= 0.76m/s, (d) U

in

= 0.89m/s, (e) U

in

= 1.04 m/s, (f) U

in

= 1.28m/s, and (g) U

in

= 2.35m/s

(b) (a)

(c)

(d)

(g)

(f)

(e)

FIGURE 36 Series of flame configurations as a function of inflow velocity (v

w

= 2.46cm/s and S=180°), (a) U

in

= 0.41m/s, (b) U

in

= 1.00m/s, (c) U

in

= 1.2m/s, (d) U

in

= 1.25m/s, (e) U

in

= 1.58m/s, and (f) U

in

= 3.10m/s

(a)

(b)

(c)

(d)

(e)

(f)

(1) t = 0 sec

(2) t = 4 ms

(3) t = 7 ms

(4) t = 11ms

(a)

(b)

(c)

(d)

(e)

(e)

(e)

(e)

(5) t = 35 ms (9) t = 95 ms

(6) t = 42 ms (10) t = 1 sec

(7) t = 63 ms

(8) t = 77 ms

FIGURE 37 Series of flame configurations as a function of inflow velocity (v

w

= 3.02cm/s and S = 180°), (a) U

in

= 0.41m/s, (b) U

in

= 0.62m/s, (c) U

in

= 0.71m/s, (d) U

in

= 1.00m/s, (e) U

in

= 1.39m/s (Night shot photos), (f) U

in

= 1.43m/s, and (g) U

in

= 3.00m/s

(e)

(e)

(e)

(e) (g)

(f) (e)

(e)

FIGURE 38 Various flame stabilization regions over a burner (S=360°)

1.00 1.20 1.40 1.60 1.80

Fuel Ejection Velocity (cm/sec)

0.00 0.40 0.80 1.20 1.60 2.00

A ir fl o w V e lo c it y ( m /s e c )

Envelope Flame (Yellow Flame in the Downstream, Initial) wake flame (blue and yellow)

lift-off flame (blue and yellow) lift-off flame (blue)

late wake flame

V VI

Envelope Flame Wake Flame

Yellow Lift-off Flame Blue Lift-off Flame Late Wake Flame

(1) t = 0 sec

(2) t = 4 ms

(3) t = 28 ms

(d) (c)

(e) (b)

(a) (e)

(e)

(e)

(4) t = 77 ms

(5) t = 109 ms Night shot photo

(6) t = 116 ms

(7) t = 123 ms

FIGURE 39 Series of flame configurations as a function of inflow velocity (v

w

= 1.23cm/s and S=360°), (a) U

in

= 0.41m/s, (b) U

in

= 0.51m/s, (c) U

in

= 0.80m/s, (d) U

in

= 1.00m/s, (e) U

in

= 1.05m/s (Night shot photos), (f) U

in

= 1.21m/s and, (g) U

in

= 1.63m/s

(e)

(e) (e)

(e)

(e)

(f)

(f)

(g)

FIGURE 40 Series of flame configurations as a function of inflow velocity (v

w

= 1.4cm/s and S=360°), (a) U

in

= 0.41m/s, (b) U

in

= 0.51m/s, (c) U

in

= 0.84m/s, (d) U

in

= 1.06m/s (Night shot photo), (e) U

in

= 1.24m/s (Night shot photo), and (f) U

in

= 1.63m/s

(a) (e)

(b) (f)

(c)

(d)

(1)

(2)

(3)

(4)

FIGURE 41 The transient oscillation photos of lift-off flame in U

in

= 1.03m/sec and

v

w

= 1.12cm/sec (The number 1, 2, 3, and 4 represent the time sequence.) (Left photos

are normal ones and right photos are night shot ones.)

6 0 0

FIGURE 42 The predicted temperature contours of transient lift-off flame in U

in

=

1.03m/sec and v

w

= 1.12cm/sec (Unit: K)

在文檔中 雙Tsuji燃燒器火燄交互作用之實驗與數值分析(I) (頁 61-0)