Period 2 Progress

©Copyright photo

Period 2 Progess


  • D1.3 – SOPRANO First Periodic Report

SOPRANO aims at developing clean and efficient combustion technologies for aircraft engines to low-medium TRL (up to TRL5), able to satisfy simultaneously the following challenges:

– Meet the ambitious ACARE NOx and CO2 objectives,
– Enhance the combustor’s liner durability,
– Control the emitted PM in terms of mass, number and morphology.

This can be broken down into the following scientific and industrial objectives. The SOPRANO project’s main scientific objective is to make a breakthrough in the overall investigation efforts in the field of soot particles chemistry, particles size distribution (PSD), and their radiative effect on combustors typical of aero-engines. SOPRANO aims at a qualitative shift in the knowledge and experimental and numerical approaches related to the characterization and prediction of soot emission and interaction with radiative Low NOx combustor environment. The main industrial objective of SOPRANO is to carry out an in-depth characterization of soot particles emitted by a modern combustor at engine relevant operating conditions and at increased pressures to pave the way for the future design of high-performance combustors: a more accurate evaluation of the radiation effect and, therefore, a more reliable liner temperature prediction, will drive a review of the design criteria in terms of combustor air distribution and will improve durability of some key modules, e.g. the combustor’s liners.


  • D2.4 – Experimental characterization of the flow and mixing field inside the KIT combustor for two operating conditions

In order to aid the design process towards low emissions’ combustion chambers, detailed measurements of combustion related emissions under realistic test conditions are essential. Subtask 2.4 in WP2 of the SOPRANO project focuses on soot emissions of aero engine combustors. In order to study the formation of soot particles, an RQL (Rich-Quench-Lean) combustor capable of burning kerosene was set up and atmospheric tests were performed. The combustor has realistic dimensions, pressure drop and mixing air injection as well as realistic air inlet temperatures. First steps were taken to characterize soot particle formation and oxidation inside the combustion chamber. In this context, flow and mixing field were characterized using a combination of PIV and OH Chemiluminescence measurements, while temperature and major chemical species were measured at different positions within the chamber.

  • D2.8 – Report on gas-soot interactions, soot inception and PSD dynamics

The need to establish the dynamics of particle size distributions (PSDs) of soot emissions from the nanoscale upwards, along with the current global indicators based on soot mass, stems from increasingly strict regulatory demands. In the current deliverable, a mass and number density preserving sectional model is coupled with a transported probability density function (PDF) method to study the evolution of soot PSDs in two non-premixed turbulent jet flames at Reynolds numbers of 10,000 and 20,000. The transported PDF approach is closed at joint-scalar level and includes mass fractions of 29 (14 steady state) gas phase species, 62 soot sections and enthalpy, leading to a fully coupled 78-dimensional joint-scalar space, treating interactions between turbulence and gas phase/soot chemistry as well as radiation without approximation. The systematically reduced gas phase chemistry is combined with a revised acetylene-based soot inception model calibrated using comprehensive detailed chemistry up to pyrene and applied to a well-stirred/plug flow reactor configuration for validation of PSD predictions. Soot surface growth is treated via a PAH analogy and oxidation via O, OH and O2 is accounted for. Particle sizes in the range 0.38 nm ≤ dp ≤ 100 μm are covered and the collision efficiency of small particles (≤ 10 nm) is treated via a Lennard-Jones potential based model. An extensive sensitivity analysis is also presented for key model components. Significantly, it is shown for the first time that the evolution of the PSDs of turbulent flames and be reproduced with an accuracy similar to state of the art modelling of laminar flames.

  • D2.9 – Report on model developments of the lumped sectional approach with respect to complex fuels

This report describes the development and validation of a soot model which is able to treat complex fuels. After an overview of the newly developed gas-phase mechanism, a detailed description of the sectional PAH and the soot model is given. For its validation, the model is tested against different flame configurations with increasing complexity. Starting with shock-tube pyrolysis and oxidation experiments of C2 fuels, complex fuels like jet-A1 are included in the validation. Different operating conditions with different pressures and stoichiometries are considered as well. Finally, a calculation of a laminar jet-A1 flame is performed. These comprehensive tests will proof the ability of the model to perform well under a large number of different operating conditions and fuels without modification of the model constants. All calculations are performed with the DLR in-house code THETA.

  • D2.14 – Numerical evaluation of methods of moment for the description of soot

The accurate description and robust simulation, at relatively low cost, of number density, volume fraction and also size distribution of soot particles in flames is still a major challenge. From their statistical modeling through the population balance equation, different kinds of models can be developed, like sectional and moment methods. However, the cost of sectional methods due to the large number of sections needed for a good accuracy and the complexity of moment method due to closure choice and complex realizability constraints on high order moments make the use of alternative method interesting. In this work, a recently developed two size moment (TSM) method is extended to the soot description, integrating a physico-chemical soot model. It is an hybrid method between sectional and moments methods, considering a size discretization in sections and two moments per section, which represent two physically important variables: the number density and the volume fraction. This leads to a modeling framework for the simulation of coupled inception, coagulation, condensation, growth and oxidation of soot particles in flames. The integrated model is validated by comparisons with reference simulations using the Monte-Carlo method and with experimental data for sooting premixed laminar flames. Moreover, its performances are compared to the ones of a sectional method and of a moment method, showing its great interest in term of accuracy and cost.


  • D3.5 – Ignition measurements w/o cooling at atmospheric and relight conditions

The ability to re-ignite at high altitudes after a flame out event is critical for flight safety. One reason that makes the relight process of the engine difficult is the low temperature and low pressure, which lead to poor atomization, slow reaction rate and evaporation of the atomized fuel. The characterization of the ignition process for kerosene(Jet A-1)-air mixtures has been conducted with a rectangular, one sector combustion chamber, which has been developed by Engler-Bunte Institute at KIT in cooperation with GE AVIO. The fuel injection system and the ignition system are also provided by GE AVIO. Ignition capability for different realistic conditions in terms of probability and minimum FAR for a successful ignition event, was determined, indicating that pressure is the dominant parameter which affects the ignition process. In addition, the time interval between the first spark until the onset of the flame was measured. Finally, high-speed recording of the ignition event in the visible wavelength was acquired. The generation and the propagation of the flame kernel were captured.

  • D3.7 – Flowfield measurements at atmospheric conditions w and w/o cooling

Film cooling jets in a combustor chamber are deeply affected by swirling flow interactions and unsteadiness; moreover they have a direct impact on different phenomena such as cooling capabilities and ignition. For these reasons, an in-depth characterization of the film-cooling flows in the presence of a swirling mainflow, demands dedicated time-resolved analyses. The experimental setup consists of a non-reactive single-sector linear combustor simulator installed in an open loop wind tunnel. It is equipped with a swirler and a multiperforated plate to simulate the effusion cooling system. The rig is scaled with respect to the engine configuration, to increase spatial resolution and to reduce the characteristic frequencies of the unsteady phenomena. Different values of the pressure drop across the liner plate were investigated to assess its effect on the results applying Time-Resolved Particle Image Velocimetry (TRPIV) technique. Oscillations of the jets and unsteady interactions of the mainstream with the wall of the liner have been investigated in details.

  • D3.10 – CFD results and analysis

The layouts of gas turbine combustors will change in the coming years to meet the increasingly strict regulations on pollutant emissions as well as new market targets. This re-design process goes through the development of advanced cooling systems. The complexity of the turbulent reacting flow field in the primary zone of a combustor leads the experimental campaign unfeasible and standard CFD-RANS simulations inaccurate for the prediction of metal temperature. In this context unsteady CHT tools have a great potential in the optimization of computational resources for this multi-physics problem. In this work, the multi-physics THERM3D approach is validated and applied on a Scale Resolving Simulation of a double swirled burner under non-reacting conditions to include unsteady effects on the metal temperature distribution.


  • D4.3 – High-pressure non-reactive LES computation of the MICADO geometry

This document describes the numerical LES setup with the AVBP code for the non-reactive two-phase flow simulation applied to the MICADO high pressure test rig (ONERA).  The numerical methodology used for the LES non-reactive computations is explained in details, with a particular attention to the liquid injection modelling. The results are then presented and analyzed in terms of swirl motion and interaction with the fuel spray for the three operating points simulated in this study. The non-reactive fields have highlighted a strong interaction between the PVC and the spray, which is typical of the swirled injection configurations. Once the reactive computations performed, this interaction is expected to have a very strong impact on the flame structure, with a flame front moving accordingly to the PVC motion.

  • D4.6 – Numerical assessment of primary zone mixedness

This document describes the numerical CFD simulations of two combustor geometries. RANS simulations were performed to identify the main features of the flow field and assess the mixedness of these combustor flow fields as represented by a passive scalar. LES calculations have been conducted to better resolve the unsteady flow and obtain statistics in time. Analysis was also carried out on both the RANS and LES results and the statistics associated with the passive scalar mixing in the primary zone were calculated. Based on the analysis from RANS and LES results, simplified geometries were proposed, on which mainly RANS simulations were performed to investigate some of the main geometric effects that impact on the combustor mixedness.

  • D4.9 – Results from first measurement campaign available

Results from the first experimental campaign at the modified High Pressure Optical Test facility (HiPOT) at DLR are presented and discussed in this work. Preliminary shadowgraph images of kerosene sprays injected via an in-house designed plain orifice nozzle are obtained under different temperatures and pressures. The nozzles are shown to perform per design and provide potential flexibilities in future experiments in terms of precise control over spray properties. Due to an unexpected shutdown at HiPOT caused by technical difficulties unrelated to the Soprano project, the first measurement campaign is terminated without detailed characterization of fuel injections under superheated conditions by laser shadowgraphy. Instead, the optical diagnostic system is rebuilt in a laboratory for spray imaging in an atmospheric flow channel to assess its quality and capability. By implementing either a regular objective or a long distance microscope, a large-scale overview of the spray