Clement Joly’s numerical simulation (3-month) internship done at ONERA ‘The French Aerospace Lab’ in Palaiseau (France).
Original title of the internship: “Essais en soufflerie hypersonique hyperenthalpique. Simulation numérique”
Translated title of the internship: “Numerical simulation of tests in a hypersonic hyperenthalpic blowdown wind tunnel”
Aim: Understand the supersonic combustion of air and hydrogen over a flate plate for different experimental conditions to study its aerothermochemical phenomena for a non-vitiated and a vitiated heater.
Dates:05/30/2011 – 08/26/2011
Definition: a supersonic flow or vehicle is a flow, or a vehicle, respectively, that travels at a speed faster than the speed of sound (~340 m/s at sea level). An hypersonic flow is a particular supersonic flow; it travels at a speed at least 5 times faster than the speed of sound. The distinction between supersonic and hypersonic application is made because of the fact that around hypersonic speeds the rarefaction, the ionization and the molecular dissociation of the surrounding gas (air around the vehicle for example) are usually not negligible.
Moreover, due to the very high compression of the gas (creating the shock) in front of the moving body (scramjet, rocket, etc) the shock becomes closer to the body as the travel speed, – or Mach number -, increases. This increase in Mach number also increases the internal energy of the fluid from the important kinetic energy due to viscous effects. This leads to an increased temperature of the fluid after the shock and the moving body does not only have to resist the structural stress due to the compression of the gas but also to the increase in temperature. Therefore it is important to know these quantities to use the appropriate materials.
Abstract: This document is about the numerical study of the H2/Air supersonic combustion process, reproducing the F4 wind tunnel conditions (viciated air due to the arc-heating process) in order to determine if whether or not F4 was able to study such a phenomenon. Our calculations completed, we were able to conclude that the wind tunnel is indeed able to study supersonic combustion. This will help us understand better this process still not masterized at the present time, while using a clean fuel.
Keywords: Supersonic combustion, hydrogen, RANS, F4 wind tunnel, vitiated air, NO contamination, Takeno index, reaction mechanisms
Résumé: Ce document est un rapport concernant le stage effectué à l’ONERA durant l’été 2011. Il concerne l’étude numérique de la combustion supersonique de l’hydrogène dans l’air, dans les conditions de la soufflerie F4 (air vicié par la mise aux conditions génératrices grâce à un arc électrique) dans le but de savoir si elle était adaptée à l’étude de ce genre de phénomène. Après avoir réalisé nos calculs, nous avons conclu que la soufflerie était en mesure de pouvoir étudier cette combustion et ainsi continuer de comprendre ce phénomène loin d’être totalement maitrisé, tout en utilisant un combustible propre.
Mots-clés:Combustion supersonique, hydrogène, RANS, soufflerie F4, air vicié, contamination au NO, critère de Takeno, mécanismes réactionnels
Experiments were performed at the F4 ONERA wind tunnel and the results were compared to the ones from the numerical simulations. The F4 wind tunnel is a blowdown wind tunnel featuring bursts of 50-200 ms that travel at a speed up to 6 km/s. This facility was used purely for atmospheric reentry tests before its potential for supersonic combustion was put to use.
The initial pressure and temperature conditions for a given test are obtained thanks to an electric arc. The one disadvantage of this type of heating process is the dissociation of the gas molecules of air in our case leading to a NO (Nitric Oxyde) vitiation in the 3m diameter round test section. Computational experiments were realized with and without a NO vitiation to take the effect of the species into account for the auto-ignition delay.
The image underneath shows the flow topology close to the transverse injection. The contour colors represent the flow Mach number and the streamlines colors are representative of the water molar percentage – the more water the more has combusted. The flow is running from left to right at hypersonic speed and hits the penetrating jet to form different structures; a boundary layer shock close to the flat plate before the injection (first zone of recirculation on the left), a bow shock as the interaction between the external flow and the hydrogen injection (on top of the boundary layer shock) and a Mach disk where the Mach number is the highest – symbol of an underexpanded flow after the fuel injection.
A second high-vorticity zone is featured for this flow, as it can be seen on the image below. Here the external flow comes around the tridimensional bow shock to move downstream.
6 calculations were realized for different configurations of pressure (external and injection), temperature and velocity. The results were then compared in terms of combustion efficiency and jet penetration height. The results for the jet penetration as a function of the ratio of the injection pressure over the total pressure followed the trend of Schetz although the results obtained slightly underestimated this trend. The combustion efficiency was determined by calculating the integral mass flux of the resulting combustion steps’ products at the outlet of the calculation domain which was of the size of the plate used in the wind tunnel.
Moreover beside a parametric study of the flow another aspect was analyzed; the kinetics due to the combustion. In order to do so an auto-ignition delay investigation was undertaken. The delay was calculated for different temperatures and different NO vitiation concentrations (0%, 0.5%, 2.5% and 5%) for 3 reaction mechanisms; Eklund (7 equations with 6 species), Jachimowski’s simple model (19 equations with 8 species) and Jachimowski’s complex model (33 equations, 13 species). Eklund’s mechanism however does not include the NO species so the vitiation study was only possible with Jachimowski’s models. The auto-ignition delay versus the temperature for these three mechanisms was compared to experimental results. It was found that for higher temperatures the vitiation has less effect on the auto-ignition delay than it has for lower temperatures. The range of temperatures used was 900K – 1100K.
Finally, a study of the type of flame was realized using Takeno’s index. Takeno’s index is a qualitative index to determine the type of flame if there is combustion for a 1-step reaction. Since the main reaction that occurred in the experiment was H2 + O2 = 2OH it is this reaction that was mostly studied to determine the type of flame present in the flow.
Takeno’s index is defined as a dot product of the concentration gradient of a species (H2) and the concentration gradient of the other species (O2). Therefore if the result is a positive number then the species (H2 & O2) are losing concentration in the same direction (downstream) and if the result is negative then the species are diffusing within each other. In the first case the flame is a premixed flame while in the second case it is a diffusion flame. It is however important to note that Takeno’s index gives the type of the flame even though a combustion does not necessarily occur at this location. For this reason it was normalized by the reaction heat to only consider the places where a combustion actually takes place.