water screen simulation

[LT]

Hello Everyone,
I'm trying to perform some fds simulations about compartimentation with water curtain in road tunnel cases.
I learnt about the use of lagrangian particles in the FDS Users's Guide and I've created a  spray model with solid particles based on the follow parameters:

&PART ID='solid_particle',
      SURF_ID='SPHERE',
      RGB=51.0,102.0,255.0,
      AGE=3.0,
      SAMPLING_FACTOR=1,
      QUANTITIES='PARTICLE DIAMETER','PARTICLE VELOCITY'/

&SURF ID='SPHERE',
      GEOMETRY='SPHERICAL',
      RADIUS=5.0E-4,
      MATL_ID='water'/

&MATL ID='water',
      SPECIFIC_HEAT=4.186,
      CONDUCTIVITY=0.6,
      DENSITY=1000.0/

&PROP ID='SprayMod01',
      PART_ID='solid_particle',
      PARTICLES_PER_SECOND=32000,
      FLOW_RATE=81.0,
      PARTICLE_VELOCITY=14.0,
      SPRAY_ANGLE(1,1:2)=0.0,0.0,
      SPRAY_ANGLE(2,1:2)=5.0,85.0,
      SPRAY_PATTERN_SHAPE='UNIFORM'/

&DEVC ID='NOZZLE_nord', PROP_ID='SprayMod01', XYZ=39.0,7.0,8.0, QUANTITY='TIME', SETPOINT=0.0/
&DEVC ID='NOZZLE_ymax', PROP_ID='SprayMod01', XYZ=39.0,8.25,7.75, ORIENTATION=0.0,0.707107,-0.707107, ROTATION=90.0, QUANTITY='TIME', SETPOINT=0.0/
&DEVC ID='NOZZLE_ymin', PROP_ID='SprayMod01', XYZ=39.0,5.75,7.75, ORIENTATION=0.0,-0.707107,-0.707107, ROTATION=90.0, QUANTITY='TIME', SETPOINT=0.0/


I''ve applied this spray model to three nozzles with different orientation in order to fill the whole cross section of the tunnel (see image 1 and 2)  and then i evaluated the effect of the water curtain on the smoke relased from a fire vent (1 m^2,HRRPUA=1000 kW/m^2=constant).
The results that i've obtained are not so good (see  images 3,4,5,6): the smoke penetrates the water barrier,follows the field of velocity induced by the nozzles and when hits the tunnel surfaces it is separated:
-a portion  of smoke goes downstream in the fire zone ;
-a portion  of smoke goes upstream.

What could i do to improve the situation?

Best Regards










 

[Kevin]

This is not the best way to specify water droplets. Do something like this:

&SPEC ID='WATER VAPOR'/
&PART ID='water drops', SPEC_ID='WATER VAPOR', DIAMETER=750. /
&PROP ID='K-11', QUANTITY='SPRINKLER LINK TEMPERATURE', OFFSET=0.10, PART_ID='water drops', FLOW_RATE=180., PARTICLE_VELOCITY=10., SPRAY_ANGLE=30.,80., SMOKEVIEW_ID='sprinkler_upright' /
&DEVC XYZ=0.0,0.0,4.9, PROP_ID='K-11', ID='Spr_1', QUANTITY='TIME', SETPOINT=0.  /

[LT]

Il giorno venerdì 24 febbraio 2017 21:13:15 UTC+1, Kevin ha scritto:

This is not the best way to specify water droplets. Do something like this:

&SPEC ID='WATER VAPOR'/
&PART ID='water drops', SPEC_ID='WATER VAPOR', DIAMETER=750. /
&PROP ID='K-11', QUANTITY='SPRINKLER LINK TEMPERATURE', OFFSET=0.10, PART_ID='water drops', FLOW_RATE=180., PARTICLE_VELOCITY=10., SPRAY_ANGLE=30.,80., SMOKEVIEW_ID='sprinkler_upright' /
&DEVC XYZ=0.0,0.0,4.9, PROP_ID='K-11', ID='Spr_1', QUANTITY='TIME', SETPOINT=0.  /

Hi Kevin,
thanks for your answer.
 I've done the same case with the water vapor droplets and i see a similar behavior about the smoke propagation through water curtain because a fraction of the smoke can pass the barrier.
In this case i saw that the smoke has cooled more than the previous case and then it moves slowly.
However there is a prolem about the smoke de-stratification that could be worse the evacuation process of the people.


 

[dr_jfloyd]

Water curtains do not block 100 % of the smoke. I recommend you find some test data for water curtains and validate your modeling approach against that data before attempting to make predictions for your tunnel.

[LT]

Hi dr_jfloyd, you're right.
I don't find experimental works about the smoke spread through water barrier.
All the works done in the past are carried out analyzing only the thermal radiaction reduction and i didn't find measures about the smoke spread reduction.
I saw that  were carried out some experiments about the water curtain  in road tunnels in the european project UPTUN but i don't find the technical reports about the results.

My work is just a first approach to evaluate this possibility using fds  in order to evaluate if this kind of system can improve (or not) the life safey aspect.
In particular, i will expect that the hot smoke can produce an increase the humidty level  due to evaporation of the droplets in the region close to the barrier and therefore i'm not sure that it is a system that can stop the smoke propagation like a solid curtain.
Obviously all  the results that  i can obtain (positive or negative) should be compared with some experimental cases in order to improve the model reliability for further simulations.

Another question that i would ask to you is about the field of velocity in x direction (see attached image).
When nozzles are activated i see the same condition that could be created with an air barrier that points in the -z direction.
Is it due to air entrainment?

[dr_jfloyd]

There is work. Here is on paper that references a number of others:

Fong, Chow, and Chan PRELIMINARY STUDIES ON SMOKE SPREADING PREVENTION AND THERMAL RADIATION BLOCKAGE BY A WATER CURTAIN International Journal on Engineering Performance-Based Fire Codes, Volume 3, Number 3, p.98-103, 2001  
http://www.bse.polyu.edu.hk/researchCentre/Fire_Engineering/summary_of_output/journal/IJEPBFC/V10/P89-95.pdf

Sprinkler sprays do entrain air as a result of drag from the water droplets. This is one of the mechanisms by which a water curtain can reduce smoke spread.

Zhou, Dong, Gao, Wu Experimental Study on Blocking Fire and Smoke by Water Curtain Journal of Applied Fire Science 18(1):1-36 · January 2008