This paper reports on a series of large-scale flame-jet ignition tests conducted at the Defense Research Establishment Suffield in Canada. The experimental configuration consisted of a steel tube connected to a large plastic bag simulating an unconfined region. The test volumes were filled with fuel-air mixtures employing acetylene, ethylene, propane and vinyl chloride monomer fuels. The nature of the flame jet was controlled by an array of obstacles in the tube and, in many of the tests, by a central circular blockage installed in the exit plane of the tube. Transition to detonation was observed in the bag for acetylene-air mixtures under a variety of geometric conditions. In the case of a flame jet emerging from an open tube, high-speed cinematography and numerical calculations have shown that the entrainment of hot combustion products into the vortex established by the shock-induced flow ahead of the flame is responsible for a sequence of localized explosions which result in shock wave amplification sufficient to initiate detonation. Other modes of transition were observed with the acetylene-air system, including transition downstream of a central blockage, transition due to colliding flame tongues in an asymmetric jet, and transition due to flame/boundary interactions. When the data from all of these experiments are plotted in terms of flame-jet velocity versus the sensitivity of the mixture to detonation, they describe a well-defined relationship. Successful correlation of this diversity of data would suggest that the critical conditions are established by the flame jet and that nearby perturbing boundaries merely trigger the onset of detonation. Although transition to detonation was not observed with the other fuel-air systems, it was found that a significant explosion in the bag could nonetheless occur for the relatively sensitive ethylene-air system, generating pressures approaching 6 bar. Cinematographic coverage has revealed that this is likely the result of a series of “hot spots” or regions of high reactivity developing downstream of the tube exit. Although these mini-explosions do not persist long enough for the critical size of a detonation kernel to be exceeded, the resulting reactive blast wave takes considerable time to decay.
[1]
Bjørn H. Hjertager,et al.
Transition to detonation in a flame jet
,
1989
.
[2]
John H. S. Lee,et al.
Dynamic Parameters of Gaseous Detonations
,
1984
.
[3]
A. Jenssen,et al.
Transition to detonation in a large fuelair cloud
,
1985
.
[4]
I. Moen,et al.
The mechans of transition from deflagration to detonation in vapor cloud explosions
,
1980
.
[5]
R. Knystautas,et al.
Photochemical initiation of gaseous detonations
,
1978
.
[6]
I. Moen,et al.
Detonation Length Scales for Fuel-Air Explosives
,
1983
.
[7]
Stephen A. Ward,et al.
A Brief Description of the DRES Fuel-Air Explosives Testing Facility and Current Research Program,
,
1982
.
[8]
Y. Zeldovich,et al.
Regime classification of an exothermic reaction with nonuniform initial conditions
,
1980
.
[9]
P. J. Shuff,et al.
Deflagration to Detonation Transition from a Venting Pipe
,
1989
.