4 th International Conference of Hydrogen Safety, September 12-14, 2011, SAN FRANCISCO, USA Pro-Science EXPERIMENTAL STUDY OF IGNITED UNSTEADY HYDROGEN RELEASES FROM A HIGH PRESSURE RESERVOIR Grune, J. 1 , Sempert, K. 1 , Kuznetsov, M. 2 , Jordan, T. 2 1 Pro-Science GmbH, Germany 2 Karlsruhe Institute of Technology, Germany
17
Embed
Pro-Science 4 th International Conference of Hydrogen Safety, September 12-14, 2011, SAN FRANCISCO, USA EXPERIMENTAL STUDY OF IGNITED UNSTEADY HYDROGEN.
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
4th International Conference of Hydrogen Safety, September 12-14, 2011, SAN FRANCISCO, USA
Pro-Science
EXPERIMENTAL STUDY OF IGNITED UNSTEADY HYDROGEN RELEASES FROM A HIGH PRESSURE
RESERVOIR
Grune, J. 1, Sempert, K.1, Kuznetsov, M. 2, Jordan, T.2
1Pro-Science GmbH, Germany
2Karlsruhe Institute of Technology, Germany
4th International Conference of Hydrogen Safety, September 12-14, 2011, SAN FRANCISCO, USA
Pro-Science
Structure of the presentation
Introduction
Objectives
Experimental
Test results
Summary
4th International Conference of Hydrogen Safety, September 12-14, 2011, SAN FRANCISCO, USA
Pro-Science
Hydrogen is successfully used as energy carrier in many different applications.
Introduction
Accidental hydrogen releases from pipe systems are one of the main hazards that occur in the handling of pressurized hydrogen.
The accidental H2 release from the system should be detected fast and as safety
consequence the main supply tank should close. So the released amount of hydrogen is limited and the release conditions are unsteady.
The generated hydrogen cloud can be ignited subsequently by an external ignition source or in case of sudden hydrogen release from high pressure state the possibility of a self ignited jet fire is present.
Currently no systematic studies are available on the hazard potential of transient releases of small amounts of hydrogen from a high pressure reservoir.
4th International Conference of Hydrogen Safety, September 12-14, 2011, SAN FRANCISCO, USA
Pro-Science
Objectives
Goal of this work is to quantify the possible hazards arising from spark ignited or auto ignited free transient hydrogen jets.
To study the nature of transient hydrogen jets and their combustion behaviour a simulation of an accidental hydrogen release from a high pressure pipe system was performed with two different start-up release conditions.
I. A fast valve opening to produce the free gas jet. This jet is ignited with a forced spark.
II. Additionally a rupture disc is installed in the exhaust pipe, which leads to a sudden hydrogen release with the possibility of an auto ignited jet fire.
- Circular release opening: 3 mm, 4 mm and 10 mm (forced ignition test). 4 mm (auto ignition test)- Initial reservoir pressures of up to 200 bar. - Reservoir volume of 0.37 dm3.
The main parameters in these experiments are:
4th International Conference of Hydrogen Safety, September 12-14, 2011, SAN FRANCISCO, USA
Pro-Science
ab c
d e
fg
h
P
H2
a c3.70 dm3
tube nozzle
tube nozzle with rupture disc
Experimental set-up
da
b cd e
fg
h
P
H2
a c3.70 dm3
tube nozzle
tube nozzle with rupture disc
Experimental set-up
d
Experimental set-up
(pneumatic needle valve DN 3 mm, driven with pressurized helium, valve opening time < 2 ms)
The small volume between the rupture disc and the leak valve was filled with H2 at 1 bar.
The nozzle downstream of the rupture disc is open to the atmosphere.
The burst pressure of the used aluminium rupture disc was clearly lower than the overpressure in the reservoir.
Schematic of the experimental set-up.
Forced ignition test:
Auto ignition test:
Fast leak valve opening produced the unsteady free jet.
Additionally a rupture disc holder is installed behind the fast leak valve.
4th International Conference of Hydrogen Safety, September 12-14, 2011, SAN FRANCISCO, USA
Pro-Science
Ignition source Positioned along the jet axis on a lanceHigh frequency electric spark (20 kHz, ~ 60 kV) with an electrode distance of ~ 6 mm.
Five dynamic pressure sensors: PCB (Type 113A31)
Pressure sensor line was used in parallel (50 cm) to the jet axis.
(reflected pressure measurement)
Integral heat flux sensorsPositioned on lance along the jet axis.
Unobstructed length in space of 4 m. Distance to the floor and the nearest wall was 1.6 m.
Jet axis
Experimental set-up
Experimental set-up
Optical observation By high speed camera in different configurations.
4th International Conference of Hydrogen Safety, September 12-14, 2011, SAN FRANCISCO, USA
Pro-ScienceDynamics of H2-releases from pressurized vessels
Example for the combustion pressure amplitude of a spontaneous ignition test.
4 mm nozzle, 200 bar
The small amount of burned mixture released from the nozzle ignites the following hydrogen flow jet fire.
-Fist: shock wave from the ruptured disc leaves the nozzle exit.
-Second: combustion of the turbulent free jet.
Two pressure peaks were observed:
4th International Conference of Hydrogen Safety, September 12-14, 2011, SAN FRANCISCO, USA
Pro-Science
P0 = 200bar; 4mm nozzle
Ignition distance 50 cm
0
20
40
60
80
100
120
140
160
180
200
20 30 40 50 60 70 80ignition delay time / ms
Pm
ax /
mb
ar
gauge P3
Ignition distance vs. maximum combustion overpressure
0
50
100
150
200
250
300
0 50 100 150 200
ignition distance / cm
P /
mb
ar 4 mm [100 bar] 3 mm [100 bar] 3 mm [200 bar]
4 mm [200 bar]10 mm [200 bar]
time
high reactive mixture cloud
jet
time
high reactive mixture cloud
jet
Spark ignition: maximum pressure
Due to the start-up process of the head of the flow in combination with the transient release conditions a highly reactive hydrogen / air cloud inside the turbulent jet is formed.
The volumetric size and the position of this highly reactive hydrogen / air cloud are changing during the transient release.
For every configuration a distinct ignition time and ignition position exists for the generation of a maximum pressure wave due to a "local explosion" in the free jet.
-Variation of the ignition time. -Variation of the ignition distance.
4th International Conference of Hydrogen Safety, September 12-14, 2011, SAN FRANCISCO, USA
Pro-Science
Nozzle: 4 mm, P0 = 200 bar, ignition time and distance optimized.
Example "local explosion"Spark ignition:
0.6 m
Integral heat flux sensors Nozzle
Ignition source
H2 jet
Real and shadow image
Digitally processed(pressure wave visualization)
4th International Conference of Hydrogen Safety, September 12-14, 2011, SAN FRANCISCO, USA
Pro-Science
Maximum peak overpressure, auto and spark ignition: nozzle 4mm
0
50
100
150
200
250
0 50 100 150 200 250
P0 / bar
Pm
ax /
mb
ar
25 mm, no ignition
25 mm, ignition
42 mm, no ignition
42 mm, ignition
60 mm, no ignition
60 mm, ignition
60 mm, ignition
80 mm, no ignition
80 mm, ignition
120 mm, no ignition
120 mm, ignition
subsequent spark ignition
posibility of eardrum rupture [7]
Hazard potential of pressure loads.
Maximum measured overpressure amplitudes from the 4 mm nozzle experiments, measured in reflected orientation in 50 cm distance to the jet axis.
For 200 bar of initial release overpressure reflected amplitudes of 220 mbar in a distance of 50 cm to the jet axis were measured, this value reaches the upper limit for possible irreparable ear injuries.
4th International Conference of Hydrogen Safety, September 12-14, 2011, SAN FRANCISCO, USA
An early ignition of the released hydrogen close to the nozzle produces the highest thermal loads.
Measured maximum integral heat fluxes along the jet axis.
The maximum measured thermal loads decrease almost linearly with increasing distance to the nozzle.
Due to the fast process: Integral heat flux sensors were used.
Measured integral heat fluxes along the jet axis (auto ignition).
The case of the spontaneous ignition of the first released hydrogen inside the nozzle pipe is the earliest possible ignition time.
A conservative linearization of the maximum thermal loads along the jet axis is possible.
4th International Conference of Hydrogen Safety, September 12-14, 2011, SAN FRANCISCO, USA
Pro-Science
Critical distance to the nozzle for possible second degree burns of human skin
0
50
100
150
200
250
300
0 50 100 150 200
initial release overpressure / bar
dis
tan
ce
to
th
e n
ozz
le /
cm
4 mm nozzle
Assumption: The characteristic release time of the transient hydrogen jets is the conservative exposure time for the integral thermal loads.
Hazard potential of thermal loads
Energy threshold: A.M. Stoll et al exposure time = characteristic release time
4th International Conference of Hydrogen Safety, September 12-14, 2011, SAN FRANCISCO, USA
Pro-Science
In this study an accidental release of pressurized H2 from a small reservoir of 0.37 dm3 was investigated experimentally with two different procedures.
In the first procedure a fast valve opens and tubular release nozzles with diameters of 3, 4, and 10 mm produce a hydrogen free jet in air with subsequent spark ignition.
In the second release application an additional rupture disc was installed in the 4 mm exhaust pipe and the sudden release conditions initiated a spontaneous ignition of the jet.
A minimum reservoir overpressure of 25 bar for a spontaneous ignition event was observed.
The transition from the auto ignition inside the exhaust nozzle pipe to the full jet fire in the ambience generates a detectable overpressure blast wave due to a "local explosion" in the free jet.
For every configuration a distinct ignition time and ignition position exists for the generation of a maximum pressure wave due to a "local explosion" in the free jet.
The generated thermal loads were investigated systematically for both ignition scenarios.A free jet that is ignited early and close to the nozzle generates the maximum thermal load for its ambience. The spontaneous ignition of the first released H2 inside the tube behind the rupture disc is the earliest ignition situation of the released jet.