Test Data AC2-11

From KBwiki
Jump to navigation Jump to search

Front Page


Test Data

CFD Simulations


Best Practice Advice

Delft-Jet-in-Hot-Coflow (DJHC) burner

Application Challenge AC2-11   © copyright ERCOFTAC 2021

Overview of Tests

The database is related to flames generated by a jet of Dutch natural gas injected along coflows generated by a secondary burner. The geometry can be found here. Different coflow composition and temperatures were explored, as well as diverse fuel jet Reynolds numbers.

Three cases were studied in more detail, and they differ in the coflow composition and temperature. These cases are denoted as DJHC-I, DJHC-V and DJHC-X. The first case had three different fuel jet Reynolds number settings, while the other two had only one.

The cases are denoted based on the estimated values for the Reynolds numbers (Re). The available measurement locations of velocities and temperatures of the studied cases are given below for each case. Here z denotes the height above the tip of the fuel nozzle.

  • DJHC-I-Re3K (Re ≈ 3000)
    • Velocities at centreline and z = 3, 30, 60 and 120 mm
    • Temperatures at z = 60, 90, 120 and 150 mm
  • DJHC-I-Re4K5 (Re ≈ 4500)
    • Velocities at centreline and z = 3, 15, 30, 60, 120 and 150 mm
    • Temperatures at centreline and z = 3, 15, 30, 60, 90, 120 and 150 mm
  • DJHC-I-Re8K5 (Re ≈ 8500)
    • Velocities at centreline and z = 3, 15, 30, 60 and 120 mm
    • Temperatures at centreline and z = 3, 30, 60, 90 and 120 mm
  • DJHC-V-Re4K5 (Re ≈ 4500)
    • Velocities at z = 3, 60 and 120 mm
    • Temperatures at z = 3, 90 and 120 mm
  • DJHC-X-Re4K5 (Re ≈ 4500)
    • Velocities at centreline and z = 3, 15, 30, 60, 120 and 150 mm
    • Temperatures at centreline and z = 3, 30, 60, 90, 120 and 150 mm

All cases have data for the composition at z = 3 mm. Values for O2, CO and NO obtained from probe measurements are provided.

Each case has data for radial profiles of velocity and temperature (both mean and variance values). The centreline data, present for some of the cases, was typically between z = 3 and 150 mm, with more measurement points at low z values. The radial coordinate range and specific locations varied from case to case. Measurements covered approximately 35 mm radius, in both directions (i.e. r = -35 to 35 mm). Details on the velocity measurements and temperature measurements are provided below. The files are ordered per case, and subdivided into temperature, velocity, flue-gas and OH-PLIF data:

  • Velocity data from LDA is in .csv format. The columns contain: r [mm], U , V ,u'u' ,v'v' ,u'v' (all [m/s] and [m2/s2]).
  • Temperature data is also in .csv format. The columns contain: r [mm], T [K] and rms(T) [K] along with fitting data( columns 4 and 5).
  • Flue gas data is also in .csv format. The columns contain: r [mm], O2 vol % and measured CO and NO, at z = 3 mm for the different cases.

Initial and Boundary Conditions

The reported composition of Dutch natural gas employed as fuel is 81% methane, 4% ethane, 14% nitrogen and 1% higher alkanes (values by volume). The volumetric flow rates for each case of the fuel jet, fuel employed in the secondary burner and air are displayed in Table 1.

Table 1 — Volumetric flows of the different studied cases.
Case Fuel Jet ((dm)3n/min) Fuel Secondary Burner ((dm)3n/min) Air ((dm)3n/min)
DJHC-I 10.7 16.1 224
DJHC-V 16.1 15.3 231
DJHC-X 30.0 14.2 239

Experimental data is available at an axial position z = 3 mm, serving as a guideline for boundary conditions. The profiles are not flat and not completely symmetric, as shown in Figs. 5 to 8, which are for DJHC-I-Re4K5. Notably, O2 concentration displays significant asymmetry.

AC2-11 fig5.png
Figure 5: Mean axial velocity profile at an axial position z = 3 mm for case for DJHC-I-Re4K5.

AC2-11 fig6.png
Figure 6: Mean temperature profile at an axial position z = 3 mm for case for DJHC-I-Re4K5.

AC2-11 fig7.png
Figure 7: Turbulence kinetic energy profile at an axial position z = 3 mm for case for DJHC-I-Re4K5.

AC2-11 fig8.png
Figure 8: Oxygen mole fraction profile at an axial position z = 3 mm for all cases.

Accuracy Considerations

The reported experimental accuracies are presented in Table 2.

Table 2 — Technique and reported accuracy of the measured variables.
Variable Technique Reported Accuracy
Velocity Laser Doppler Anemometry 2-8%
Temperature CARS 20 K
Oxygen volume fraction Gas Analyser ±0.20%

The overall uncertainty of the LDA measurements is about 2 % of the bulk velocity for the mean velocity and 8 % of the local maximum for the Reynolds stress (Absil, 1995)

Velocity Measurements (Oldenhof, 2012)

Laser Doppler Anemometry

LDA measurements were performed with a two-component, dual beam TSI-system. The system employed a green line (514.5 nm) and a blue line (488 nm) of a 10 W Continuum Argon-ion laser to directly measure the axial and radial velocity components. Two of the incident beams (one of each color) were frequency pre-shifted over 40 MHz by a Bragg cell to enable the detection of instantaneous flow reversals and stagnant flow. The focusing lens had a 82 mm aperture and a focal length of 250 mm. The length and diameter of the measurement volume were 1.7 mm and 0.12 mm, respectively. The fringe distances for the green and blue channel were 2.6 μm and 2.5 μm. Alumina (Al2O3) particles with an average size of about 1 μm were used as seeding particles. Two cyclone-type particle generators were used to seed the air and fuel separately. The generators have a provision that enables the control of the seed density in both the coflow and in the fuel jet. This provision was used to equalise the seeding density in the both flows, thereby minimizing errors related to so-called conditional seeding. The data rate (number of bursts per second) was used as an indicator, as it is proportional to the fluid density, the velocity magnitude and the seeding density. This proportionality was therefore used to equalize the seeding densities: the ratio of the data rates of the unmixed fuel- and coflow streams was made approximately equal to the ratio of their products of axial velocity and fluid density. The light scattered by the seeding particles was collected in back-scatter mode. The photomultiplier output signals were electronically down-mixed, and subsequently fed to a FSA-3000 signal processor to determine the instantaneous velocity of light-scattering particles. All statistics were computed as transit-time weighted results to eliminate the effects of the velocity bias. Autocorrelation functions of the axial velocity component were constructed from time series with 4×105 velocity samples that were acquired at a mean data rate of approximately 500 Hz by using the slotting method with local normalisation (Tummers & Passchier, 2001).

Temperature Measurements (Oldenhof, 2012)

Temperatures were determined with a CARS system. It is based on an injection-seeded, frequency-doubled Nd:YAG laser (Spectron SL805 SLM), which yields 500 mJ per pulse at 532 nm with a pulse duration of 12 ns at 10 Hz repetition rate. About 80% of the radiation is used to pump a modeless Stokes dye laser (Mode-X ML-3), emitting a broadband profile around 607 nm for Rhodamine 640 in methanol. The remaining 20% of the pump laser travels along a delay line and is split into two beams with equal intensity. The Stokes beam and the two beams at 532 nm are focused by an aplanat lens with a focal length of 300 mm in a planar-boxcars phase-matching configuration. With this configuration, a CARS probe volume of 700 μm length and 35 μm diameter is obtained. The generated CARS radiation is recollimated and combined with an attenuated sample of the Stokes beam on a dichroic beam splitter. The beams are focused onto the entrance of an echelle spectrometer, and their spectra are dispersed on a CCD detector with 1100 × 330 pixels. The spectra are contained in two strips of 1100 intensity values, which are digitized by an 18-bit AD converter and stored. The CARS spectrum is referenced to the simultaneously measured Stokes excitation profile and fitted to a library of theoretical, temperature-dependent spectra. The single-shot imprecision of the system is 1% - 4% over a range from 2000 K to 300 K. The inaccuracy is estimated to be 20 K. For each point in space, mean temperatures were determined from the results of 1000 single-shot CARS spectra.

Flue Gas Measurements (Oldenhof, 2012)

Oxygen measurements were performed with a Testo 335 flue-gas analyser, with a specified inaccuracy of ≈ 0.20%. The measured oxygen volume fractions in the coflow were converted to mass fractions using the results from equilibrium chemistry calculations with the species of the detailed Warnatz-mechanism that includes C-2 chemistry.

Measured data

The experimental data (in ASCII format) can be found below as a ZIP file, together with its README file. Further information can be obtained by contacting Prof. dr. Dirk Roekaerts, Delft University of Technology (d.j.e.m.roekaerts@tudelft.nl)

AC2-11_Data.zip README.pdf


  • L.H.J. Absil, Analysis of the Laser Doppler Measurement Technique for Application in Turbulent Flows, Doctoral Thesis, Delft University of Technology, The Netherlands (1995). uuid:deec2bf1-e25d-458b-b3c9-35c2810b22a5
  • Oldenhof E, ‘Auto-ignition and flame stabilisation processes in non-premixed turbulent hot coflow flames’, PhD Thesis, Delft University of Technology, March 20, 2012 Available from http://repository.tudelft.nl/
  • Oldenhof E, Tummers MJ, van Veen EH, Roekaerts DJEM. Role of entrainment in the stabilisation of jet-in-hot-coflow flames. Combustion and Flame 2011;158:1553-1563
  • Oldenhof E, Tummers MJ, van Veen EH, Roekaerts DJEM. Ignition kernel formation and lift-off behaviour of jet-in-hot-coflow flames. Combustion and Flame 2010;157:1167-1178.
  • Tummers MJ, Passchier DM. Spectral analysis of biased LDA data. Measurement Science and Technology 2001; 12:1641?1650.

Contributed by: André Perpignan, Dirk Roekaerts, E. Oldenhof, E.H. van Veen, M.J. Tummers, Hesheng Bao, Xu Huang — TU Delft

Contributed by: Jeffrey W. Labahn — Stanford University

Front Page


Test Data

CFD Simulations


Best Practice Advice

© copyright ERCOFTAC 2018