M.S. Kuznetsov

Evaluation of limits for effective flame acceleration in hydrogen mixtures

Abstract Results of experiments and data analysis on turbulent flame propagation in obstructed channels are presented. The data cover a wide range of mixtures: H 2 /air, H 2 /air/steam (from lean to rich) at normal and elevated initial temperatures (from 298 to 650 K) and pressures (from 1 to 3 bar); and stoichiometric H 2 /O 2 mixtures diluted with N 2 , Ar, He and CO 2 at normal initial conditions. The dataset chosen also covers a wide range of scales exceeding two orders of magnitude. It is shown that basic flame parameters, such as mixture expansion ratio σ , Zeldovich number β and Lewis number Le , can be used to estimate a priori a potential for effective flame acceleration for a given mixture. Critical conditions for effective flame acceleration are suggested in the form of correlations of critical expansion ratio σ ∗ versus dimensionless effective activation energy. On this basis, limits for effective flame acceleration for hydrogen combustibles can be estimated. Uncertainties in determination of critical σ ∗ values are discussed.

Slow and fast deflagrations in hydrocarbon-air mixtures

Results of experiments are presented on the behavior of turbulent flames in mixtures of methane, ethane, and propane with air. Tests were performed in two explosion tubes, one with an inner-diameter of 174 mm and the other of 520 mm. It was found that (similar to hydrogen combustibles) the flame acceleration can be weak or strong resulting either in slow subsonic flames, or in fast supersonic combustion regimes. Critical mixture compositions for the threshold between slow and fast combustion regimes were determined in the tests. In lean mixtures of hydrocarbon fuels, the critical compositions were not dependent on the tube size. On the rich side, a significant scale effect was observed. Effect of the basic flame properties, such as mixture expansion ratio, laminar flame thickness, and effective, activation energy on the critical conditions for strong flame acceleration is discussed.

Experimental study of flame acceleration and the deflagration-to-detonation transition under conditions of transverse venting

Abstract Results of experiments on critical conditions for flame acceleration and the deflagration-to-detonation transition in tubes with transverse venting are presented. Tests were made with hydrogen mixtures in two tubes (inner diameter of 46 and 92 mm) with obstacles. Ratios of vent area to total tube area were 0.2 and 0.4. Venting was shown to influence flame acceleration significantly. The greater the vent ratio, the more reactive the mixture necessary for development of fast flames. Critical conditions for flame acceleration in tubes with venting, expressed through a critical mixture expansion ratio σ cr , were found to be σ cr / σ 0 ∼1+2 α , where σ 0 is the critical value for a closed tube. Critical conditions for detonation onset in a vented tube were found to be very close to those in a closed tube with similar configuration of obstacles.

Detonation propagation, decay, and reinitiation in nonuniform gaseous mixtures

Experiments on the behavior of detonation waves in nonuniform mixtures are presented. The situation studied was the propagation of a detonation wave from a driver mixture of variable length through a concentration gradient of variable width into a less reactive acceptor mixture. The effect of the gradient on the transmission process were studied. A detonation tube of 174 mm id. was used. The tube was initially divided by a fast-opening, stretched rubber diaphragm. Stoichiometric hydrogen/air mixture was used in the driver section. Hydrogen/air mixtures (14.0–19.0% H 2 ) were used as acceptor mixtures. Natural diffusion was used to create a concentration gradient between two mixtures. It was shown that the behavior of detonations at concentration gradients depends significantly on the sharpness of the gradient. For relatively sharp gradients a detonation always decays in the nonuniform region. It can be reinitiated downstream in the acceptor mixture, if the driver length is large enough for a particular acceptor mixture. For relatively smooth gradients, detonation is able to propagate through without decay. The boundary between these cases is defined only by the value of sensitivity gradient for a particular pair of driver and acceptor mixtures. The critical value of the gradient depends strongly on the difference in energy content of driver and acceptor mixtures. The more overdriven is the detonation in the driver mixture compared to that in the acceptor, the sharper gradient is necessary for detonation decay. The order of magnitude of critical values of the gradient shows that evolution of the cellular structure may play a role effecting conditions for detonation decay at concentration gradients.

Air blast and heat radiation from fuel-rich mixture detonations

Large scale experiments were carried out to study the effect fuel concentration on air blast parameters and heart radiation from gaseous detonations. Hemispheric plastic envelope (4 meters in radius) was used with propane-air mixtures containing from 4 to 7 vol. % of fuel. The expressions for overpressures and impulses were determined in Sachs variables. The effect of fuel concentration on blast parameters is shown to be insignificant for the same amount of oxygen in the mixture volume. Thus the blast wave parameters can be described as for stoichiometric mixtures using additional scaling for the explosion energy according to oxygen content (cloud volume). The results of large scale experiments with fuel spray clouds containing 0.16–100 tons of fuel with mean concentration from stoichiometric (C0) up to 3C0 are reconsidered. These results confirm the proposed scaling of air blast parameters for a wide range of fuel types, cloud volumes and fuel concentrations. Detonations of fuel rich gaseous mixtures result in a strong heat radiation. Heat radiation energy, time and size of the fireball formed are studied as a function of fuel concentration.

Heat loss rates from Hydrogen-Air turbulent flames in tubes

Experiments were performed to study heat fluxes to tube walls caused by propagating turbulent flames and detonations. Tests were made in cylindrical explosion tubes of 174- and 520-mm i.d. with obstacles using hydrogen-air mixtures (10%, 11.5%, and 13% H 2 ) and stoichiometric hydrogen-oxygen mixtures diluted with nitrogen (dilution coefficient g in the range from 0.5 to 5.7). Three different flame-propagation regimes were studied: slow (subsonic) combustion, sonic flames, and detonations. Fast thermal gauges (time resolution <4 ms) were constructed and thoroughly calibrated for the tests. The rate of energy loss from the combustion products was shown to depend significantly on the speed of the flame propagation. The tube diameter did not influence the values of heat fluxes, while the characteristic duration of the process, in addition to the total energy absorbed by the tube walls, differed greatly at two different scales. The main mechanism for the heat losses in the present tests was shown to be the convective heat transfer. Results suggest that qualitative and quantitative accounting for heat losses from the combustion products are necessary for turbulent combustion models to increase the reliability of their predictions.