摘要:

This paper presents a numerical study of the transient operation of a pre-pressurized (augmented) airbag inflator. Augmented inflators dilute hot gaseous products of propellant combustion with ambient temperature, high-pressure stored gas before discharging the mixture into the airbag. The solid propellant selected for this study is a non-azide propellant composed of a mixture of azodicarbonirnide, potassium perchlorate, and cupric oxide, Predicted performance of the inflator is presented in terms of pressure, temperature and mass flow rate profiles in the inflator and discharge tank which is used to simulate an airbag. This work also predicts first-order estimates of gas-phase species exit concentrations and characteristic residence times in the inflator. Carbon monoxide. produced as a product of combustion from the high flame temperature propellant, is partially converted to COz as it flows from an internal combustionchamber to the pressurized plenum before being discharged into the airbag, Specifically. the production/destruction of CO is tracked using three different gas-phase reaction models: I) chemically frozen. 2) local (shifting) equilibrium. and 3) finite-rate elementary kinetics. Results presented in this paper demonstrate the necessity of an airbag combustion program that includes finite-rate, gas-phase kinetics. Results from the finite-rate CO chemistry model are qualitatively consistent with experimental results reported by others for the same propellant formulation in a similar operating environment.

ISSN:0010-2202    

DOI:10.1080/00102209708935613

出版商:Taylor & Francis Group    

年代:1997

数据来源: Taylor

摘要:

Numerical modeling results of M-and V-shaped methane/air flames with an equivalence ratio of 0.7 are presented and compared with experimental results in this paper. The numerical model uses a one-step chemistry model and a vorticity-streamfunction formulation for the flow field. The experimental results used to validate the model consist of flame shapes, stand-off distances, velocity profiles measured with Laser Doppler Velocimetry and critical transition and blowoff gradients. The flame shape and the stand-off distance of the M-shaped flame are reproduced well by the model. The values of the vertical velocity are. however, lower than the experimental values. The lower vertical velocities computed with the model are due to relatively small differences between the computed and the experimental flame shape. The shape of the V-shaped flame is also reproduced reasonably well by the model. The stand-off distance predicted by the model differs 0.8 mm from the experimental value. The vertical velocities predicted by the model are, as for the M-shaped flame, lower than the experimental values. This is also caused by flame shape differences. The small differences between the computed and experimental flame shapes is probably related with the absence of highly diffusive radicals in the one-step model. The model is also used to compute the critical transition (from M-to V-shaped flame) and blowoff (of the V-shaped flame) gradients. The values for the critical gradients predicted by the model differ no more than 10 % from the experimental values obtained with our burner.

ISSN:0010-2202    

DOI:10.1080/00102209708935614

出版商:Taylor & Francis Group    

年代:1997

数据来源: Taylor

摘要:

The oxidation of n-heptane and mixtures n-heptane-MTBE (50:50) and n-heptane-ETBE (50:50) has been studied experimentally in a high-pressure jet-stirred reactor in a wide range of conditions covering the low and high temperature oxidation regimes (570-1150 K, 10 atm, 0=1, 0.1% of fuel). The mole fractions of reactants, intermediates and final products have been measured. The influence of the additives on the formation of several pollutants has been addressed. The present results clearly show an influencing effect of MTBE and ETBE on the kinetics of n-heptane oxidation by a reduction of the mixture reactivity in the low temperature regime (570-800K). The results are interpreted in terms of knocking and non-knocking tendencies related to fuel structure and low temperature oxidation mechanism.

ISSN:0010-2202    

DOI:10.1080/00102209708935615

出版商:Taylor & Francis Group    

年代:1997

数据来源: Taylor

摘要:

Combustion of RDX was studied under self-deflagrating and CO2laser-assisted conditions at atmospheric pressure in air. Steady measurements included near-surface temperature (embedded microthermocouple), melt layer thickness, and sensitivity of burning rate to initial temperature and radiant flux. Unsteady measurements of oscillatory burning rate were also obtained using the laser-recoil technique. Thermocouple data showed a relatively thick (several hundred urn) melt layer, which increased in thickness with increasing laser flux but remained at a relatively constant temperature of about 650 K. The temporally fluctuating, spatially isothermal (time-averaged) nature of the melt layer suggest that a bubbling/mixing mechanism plays an important transport role in this zone. The temperature- and radiant flux-burning rate sensitivity data show that the equivalence principle is reasonably accurate for RDX under these conditions. The response function data agree qualitatively with those of Finlinson, et al. The classical, quasisteady ZN model does not fit RDX combustion, at least at atm, presumably due to conditions in the meh layer (e.g., bubble-induced mixing) which violate model assumptions. Nevertheless these unsteady combustion data should be useful for validating more comprehensive RDX combustion models.

ISSN:0010-2202    

DOI:10.1080/00102209708935616

出版商:Taylor & Francis Group    

年代:1997

数据来源: Taylor

摘要:

Chemical models using complex gas-phase chemistry in one-dimensional flows have predicted some acetylene/oxygen flame temperatures which exceed the adiabatic flame temperature by over 800 K under fuel-rich, high flow rate conditions. In this work, laser-induced fluorescence was applied to measure the actual temperatures in these flames and to compare them with the predictions of a 1-D model. While the model predicts maximum flame temperatures which are typically within 100 K of the experimentally determined temperatures under low flow rate conditions, at higher flow rates, the actual maximum flame temperature rises much more slowly with increasing flow rate than predicted by the model. The discrepancy between model and experiment reaches 700 K at the highest flow rates studied. In addition, the experiments suggest that there may be important structural differences between the model-predicted and the experimentally observed flames.

ISSN:0010-2202    

DOI:10.1080/00102209708935617

出版商:Taylor & Francis Group    

年代:1997

数据来源: Taylor

摘要:

The flame stretch concept is extended for the case of 3D instationary flames with finite flame front thickness. It is shown that additional contributions to the stretch rate appear apart from the terms which are usually used in flame studies. These extra terms are associated with variations in the mass density along the flame iso-contours and with variations in flame front thickness in time and space. It is finally shown that the following definition for the stretch rate is applicable: K = \/m (dm/dt), denoting the fractional change of mass in an infinitesimally small flame volume.

ISSN:0010-2202    

DOI:10.1080/00102209708935618

出版商:Taylor & Francis Group    

年代:1997

数据来源: Taylor