The 2nd PRESLHY Workshop of Experimental was organised on 23 July 2020, 14:30 – 16:30 CEST online via GotoMeeting
The presented experiments were done by partner ProScience.


1) Short motivation/intent of the experiments
2) Description of the set-up, sensors etc.
3) Experimental matrix
4) Results (measurement data, curves, correlations, videos including post-processing steps)
5) Discussion of exploitation – all (modelling, benchmarking, scientific publication,…..)


TJ: Thomas Jordan KIT
PS:         Pratap.Sathiah Shell
MV:       Molkov, Vladimir UU
CD :       Cirrone, Donatella UU
FA :       Friedrich, Andreas PS
AV :       Alexandros Venetsanos (
SG :       Stella Giannissi
AM:       Akiko Matsuo (
WJ :       Wen, Jennifer (
KM:       Kuznetsov, Mike (ITES) (
NG:       Necker, Gottfried (ITES) (
KN :       Kotchourko, Natalie (ITES) (
DH: Deborah Houssin AL
DM: Dmitriy Makarov UU
EH: Ethan Hecht SNL
ES: Etienne Studer CEA
AG: Alberto Garcia FCH JU
JF: Jens Franzen Daimler
IT: Ilias Tolias NCSRD
LP: Lee Phillips Shell
MJ: Michael Johnson DNV GL
PB: Peter Barker
SC: Simon Coldrick HSE
ZR: Zhaoxin Ren UWAR
CL: Chris LaFleur SNL
SK: Sanjay Kumar Shell

Documentation of the event

The presentation is available via

Loader Loading…
EAD Logo Taking too long?

Reload Reload document
| Open Open in new tab

The videos used in the presentation are:

The following discussion started

DISCHA facility (nonignited jet)

  1. AV: To specify geometry of the nozzle in more details for numerical simulations: length, tube diameter, valve cross-section, …
    FA: The drawing will be presented in more technical details.
  2. AV: Arrival time for hydrogen concentration in a jet is not consistent to position of the sensors.
    FA: The reason could be the widening of rectangular input signal to gaussian output signal due to transport of sample through pipe to sensor. Given time delay is for first reaction of sensor, not for maximum value.
  3. AV: Are there some jet effects or shock waves due to fast valve opening affected the sampling probes measurements?
    FA: The opening was very accurate without a shock wave or any disturbance of gas flow. The process (transport of sample to sensor) is relatively slow and takes about 2 s.
  4. AV: The temperature measurements look more reliable than the concentration ones. If we can adjust the maximum of temperature and the maximum of concentration to shift the concentration – time dependence?
    KM: Yes, it will be a good idea to improve the concentration wave arrival time by comparing them with temperature history because it should be a similarity of temperature and concentration profiles in the jet.
  5. AV: In the data file, the initial temperature is 15C.What was the real temperature?
    FA: Since the work was done from winter to spring time, the ambient temperature can be changed from 10 to 20 deg C. The real temperature can be recovered from weather history record in Karlsruhe.
  6. AV: Do you have a video to figure out the distance for buoyancy effect?
    KN: We don’t have BOS photos for very far zone. The only possibility is to process Panasonic records to extract such information. The accuracy will be very bad because of camera set-up in front a jet (not a side).
  7. AM/SG: Measured temperature at LN2 state has 7K difference compared to boiling temperature of 77. Is it regular temperature shift for the full temperature range?
    FA: The thermocouples T1-T3 were calibrated for ambient temperature of 20C, LN2 temperature of 77K and LH2 temperature of 20K. the only difference was at LN2 state (+7K) and at LH2 state (+6K). T4 was also calibrated but the difference should be checked.
  8. AM/SG: How was the sampling probes calibrated against real hydrogen concentration?
    FA: Hydrogen sensors were checked against 100% H2 and test gas 0,25% H2 in N2. All sensors have sampling lines of same length and diameter and are connected to one pump. Response time was checked by applying pure H2 from a balloon cautiously (avoiding flow effects) to tip of sampling tube. Valve of balloon was controlled remotely, opening time was recorded and compared with first reaction of sensor. Time delay of first sensor reaction was 2 s for all sensors.
  9. AM/SG: How were the thermocouples positioned: only axially or also a side direction?
    FA: Thermocouples were all in vertical plane of jet (no horizontal distance to jet axis). Thermocouples T8 and T9 have vertical distances to jet axis (as shown on slide 20), T5, T6 and T7 are on jet axis.
  10. PS: Did you make the flow velocity measurements?
    KM: It wasn’t measured. It needs PIV equipment.
    KN: It could also be done from BOS images but only with high speed camera Photron. We hope, we can process videos within ignited jet experiments to extract flow velocity before the ignition moment.
  11. PS: Was it at least once an auto-ignition of hydrogen jet release?
    It was no one self-ignition of jet release because we used quite slow ball-valve to open very carefully without auto-ignition. Of course, it should be an auto-ignition in case of very quick valve (<5ms opening time) or membrane rupture.
  12. PS: If static ignition is possible or not?
    FA: We did no efforts to prevent static ignition. In the contrary we were interested in measuring electrostatic field build-up due to the release. No spontaneous ignition was observed in all tests.
    KM: In my opinion up to now it is only a guess about LH2 ignition by the static discharge. There is a static charge but its capability to ignite LH2-jet should be experimentally proven.
  13. AV: How the existence of condensed and solid particles affect the cryogenic hydrogen release?
    FA: We observed ice at the nozzle after the experiments, but in the video records we did not observe any significant changes in the direction of the jet that might be due to a partial blockage of the opening.
  14. AV: Does it block the nozzle?
    FA: A complete blockage of the nozzle opening was not observed in the tests.

Static electric charge (nonignited jet)+ Cryostat.

  1. KN: We can try to process jet ignition experiments to extract the flow velocity.
  2. PS: How the wind speed affect the jet behaviour?.
    FA: We used ultrasonic thermo-anemometers to measure direction of the wind and velocity.
  3. PS: What is the nature of ice-“fingers” formed at the nozzle? CO2, H2O, O2, N2?
    FA: It might be gaseous substances CO2, O2, N2 because it was no humidity at the floor after the fingers dropped and evaporated without a liquid water.
  4. AV: Did you make humidity measurements?
    FA: We did not measure the humidity. There is a weather conditions record on KIT web-site to be used to extract the humidity data.
  5. AV: We are waiting for the deliverable report to start numerical calculations.
    FA: We will deliver the report within 2-3 weeks.
  6. MV: What is the nature of ice-“fingers? CO2 concentration is too small compared to steam. Probably it does not participate in “fingers”.
    KM: it could be comparable because CO2 concentration is 0.041% and water concentration changes from zero to 2.6% depending on humidity.
  7. PS: Please, upload the Cryostat video as quick as possible to process the data.
    FA: Yes, we will upload at least the raw data to the web-site.

Cryogenic Jet ignition.

  • WJ: We need the experimental layout, experimental conditions in details to start the numerical simulations for jet ignition experiments.
  • CD: Did you make the heat flux measurements for jet ignition experiments?
    KM: We did not directly measure the heat flux. The only possibility to evaluate the heat flux from ignited jet is to process thermo-vision camera data.
  • CD: Did you figure out at what pressure and distance the effect of buoyancy for ignited jet occurs?
    KM: Yes, we can distinguish only a rough distance to buoyancy dominated jet using Panasonic camera.
  • WJ: Please include all details for experimental layout, experimental conditions for numerical simulations of jet ignition experiments.
    FA: Soma details are already in the presentation. More details will be in our report.
  • CD: Can we see the border between the reaction and combustion products using ignited jet BOS images?
    KM: It is quite difficult to distinguish the reaction and the products using BOS because it has no spectral resolution. The only temperature threshold obtained by thermo-camera can be used in current tests. But the temperature threshold will be very conventional.
  • CD: (Slide 60) How the ignition time delay was chosen?
    FA: In order to reduce test matrix, it was chosen for 4-mm nozzle with regard to the maximum combustion pressure for different initial pressures and ignition location from the nozzle. Maximum loads were observed for time delays from 80 to 160 ms depending on the ignition location, so a delay of 120 ms was chosen for all tests of the second (main) series.
  • MV: It was found the flush-back concentration of 11% for hydrogen jet-ignition at ambient temperature hydrogen release. Was it found now for cryogenic test?
    KM: Previously it was found within the icefuel project. After video data processing, we also hope to get such a threshold for cryogenic jet as well.
  • CD: What is the location for pressure sensor P1?
    FA: The location is 50 cm from the nozzle and 110 cm below the jet axis (slide 53)

WJ: What is the distance from the nozzle to the explosion?
KN: The distance to ignition point is 62.5 cm, nozzle size is 4 mm, initial pressure is 200 bar, initial temperature 293K.
KM: We have analysed the high speed movie. As we found the local flame velocity was changing from 130 to 270 m/s in radial direction and from 160 to 310 in axial (co-flow) direction. The shock wave was produced by strong deflagration process. The shock wave velocity was varied from 500 to 380 m/s depending on distance to the centre of explosion. Our rough evaluation gives that the energy of explosion corresponds to 0.5-1.5 g of hydrogen compared to the amount of hydrogen 3.7 corresponding to the volume of burnable cloud.