Видео

Hi! You can look in the description at the buymeacoffee link if you are interested for any support :)

I can upload the case if you want but it will be a bit useless without the models and modified code as well.

I am not sure if you have checked my tutorial at ruclips.net/video/MYakoxQ3Jz4/видео.html. It should help you set up a basic OpenFOAM pool fire scenario at least. For a flame spread simulation additional steps are required. I have in mind to upload such a tutorial at some point. I am not sure about other online sources, I haven't looked to be honest. Normally you should be able to run the fire tutorials that come with each OpenFOAM version though (which include both pool fires and flame spread scenarios). I hope it helps!

Do you mean to simulate flame extinction just by dilution or with a water spray? Or something else?

A bit too complex to analyze in a sentence but, in general terms, yes. You specify the mass flow rate of the fire and then various parameters for the solid (including the heat of pyrolysis, heat of combustion of the pyrolysis gases, solid material properties, etc).

@@pyr-fire will you make a tutorial video of it since I watch a lot of videos and no one talk about it and normally only talk about localised fire and I think you are competent to make it

That's the next video I'll make but there's no timeline yet, I'm too busy I'm afraid. I think you can find some firefoam flame spread tutorials on the openfoam releases though

Do tell me how I can do it 👍🙂

@@pyr-fire One could record the HRRPUV integrated over the simulation domain for every time step. Then create a plot with the HRR development from the experiment static as background. Now, with each simulation time step, one creates a new plot where the HRR from the simulation is drawn up to this time step. I sometimes also draw a larger marker at the end point, then it looks like there is some pen is drawing and it is easier for the viewer to focus on. With all these plot images, one can create an animation/video. I think ffmepg can do this for example, or Blender. But I guess any video editor can do it. I'm not familiar with OpenFOAM/FireFOAM, so I asked a friend how to determine the HRRPUV integrated over the domain per time step. This is what he said, or at least the gist of it: In the file "system/controlDict" there needs to be a function set up. The type would be "volFieldValue", write control "timeStep", operation "volIntegrate" with "Qdot". I hope this answers the question somewhat... :D (also: sorry for the late reply)

You can search for qFuel in the openfoam code. I don't think you can manually change it in the simulation though.

Thanks doctor I'm a big fan of your work on GitHub. I searched in the source code itself about qfuel as well but couldn't find a pointer to the value in any file. But it appears in the log during the run time. Could you please help me to find the file which contains the pointer to qfuel value ? ​@@pyr-fire

No, not by me I'm afraid

Thank you, I will try to upload some more soon :)

I don't know the UNICORN sorry. What's that?

Hi! Take a look at github.com/pyrTriangle/firePlume_McCaffrey_57.5kW You can also find some papers where I have used this test case at github.com/pyrTriangle/CFD_papers I hope it helps :)

My knowledge on Amon Amarth is very limited :)

I don't even remember anymore to tell you the truth but no I don't think so

Will try soon :)

You can of course but you will need to use a solver with an extinction model (not provided with the tutorial itself). Also, flame extinction will occur if there's deficiency of oxygen. The way the compartment is set up is probably more of a well-ventilated case.

There are too many custom models and libraries involved that makes it too complicated to share the code I'm afraid.

What a pity! Thank you!@@pyr-fire

Thanks, I'll try to post another tutorial soon 🙂

Thank you

👍

How was the fire visualised after simulation?

Using paraview

How to get hrrpuv function in paraview

It's dQ in the old openfoam versions or Qdot in the new ones.

Thanks!

Hobby 🔥

Nice! I certainly can't 🙂

why is your channel fire themed? @@pyr-fire

Because I work with fire everyday (in a more virtual way though :))

This is just an approximation: you could calculate the initial amount of oxygen in the room. Then considering a % of oxygen that will lead to extinction (maybe 10% or so) and an average temperature inside the room (say maybe 400-500 C), estimate the amount of oxygen consumed. From that, you can calculate the total heat that would be released, and assuming an average hrr for the fire during this period, estimate how long it would take for extinction to occur.

@@pyr-fire thanks for your response, it gives me some idea. In case of atrium or open space, there will always be unlimited oxygen. Will the fire remain burning ?

Talking about combustion modeling with fast chemistry then as long as there's enough oxygen (and fuel of course) then the fire will keep burning. Only if you have vitiated conditions you will see a reduction in the burning of the fire and potentially also going to extinction. With finite-rate chemistry things would also depend on the flame temperature (apart from oxygen) since you might have lowe values which will slow down the chemistry /reactions. Hope it helps!

What do you mean by complete video? This was as long as the simulation was run.

What exactly do you need Saketh?

You are welcome 🙂

Hi Qi Guan, The bc will change the k (sub-grid scale kinetic energy) values based on the local velocity field and the prescribed turbulence intensity. Overall, I don't think it matters a lot, you could get away with using a fixedValue bc as well and avoid any other potential instabilities like the one you mentioned with janaf. It can be a little important if you use a turbulence model based on k (higher k values will dissipate the flow more for example because of added sgs viscosity). The change in the pressure file is indeed also needed because the buoyantPressure bc stopped being available after a certain OF version. I'd also use the same bc that you applied.

Hi Ekaterina, You are probably trying to run the case with another OF version than which was initially setup. The error tells you you need to have the keyword PIMPLE defined in fvSolution file in the system folder. The tutorial currently uses PISO.

@@pyr-fire Thank you! Also I get this error. Can you suggest how to fix it? --> FOAM FATAL ERROR: Cannot find file "points" in directory "polyMesh" in times "0" down to constant From function virtual Foam::IOobject Foam::fileOperation::findInstance(const Foam::IOobject&, Foam::scalar, const Foam::word&) const in file global/fileOperations/fileOperation/fileOperation.C at line 871. FOAM exiting

This would suggest to me that the mesh was probably not properly created (if at all). Maybe one of the steps in the pre script failed to run. It's a little difficult to understand exactly what the problem is only from this.

Hi Qi Guan, Typically, a one step infinitely fast reaction is good enough for most fire scenarios (unless you are interested in carbon monoxide, minor species, etc). You can also include multi-step reactions with FireFOAM, I have done it in the past, but I don't remember from the top of my hear if you need to do any modifications in the solver. With fast chemistry you don't need to have a chemistryProperties file, this is only needed only if you consider finite rate chemistry (like it is done with reactingFoam). Hope it helps!

Thank you :)

You are welcome :)

Thanks! Nothing fancy actually. It is simply ParaView with a contour similar in a way to what is reported in www.paraview.org/Wiki/Volume_rendering_-_Fire

@@pyr-fire Thanks for the hint. Honestly I thought you were using mass fraction contour like C_H2O = 0.5 maybe. I will see if I can run this on OFv2012.

Thank you. The tutorial could be used with OpenFOAM 9 but only after doing modifications in the input files (i.e., in the boundary conditions in 0 folder, in the models o nthe constant folder, settings in the system folder) and then run it with the buoyantReactingFoam solver. To create the mesh and to visualize the geometry in paraView you will have to follow all the steps included in the pre script included in the tutorial (there are more steps than simply typing blockMesh). It will require a little effort to adapt it but it is possible if you have some experience with OpenFOAM. I hope this helps!

@@pyr-fire In your rackStorage github, there is no blockMeshDict. One more thing I cannot run ./pre nor with bash pre command. I am still a newbie in OpenFOAM. Just learning to simulate fire now.

@@freyathegoddess3972 The structure in OpenFOAM 2.2.x that I used for the tutorial is slightly different than in OpenFOAM 9. You can find the blockMeshDict file in constant/polyMesh and not in system as you would expect in OpenFOAM 9. Maybe you need to give permission to the script before you run it. You could try to type chmod +x ./pre in the terminal before you run ./pre

@@pyr-fire Alright I will try it, thank you for your guidance, if it is working I will report it to you

@@pyr-fire I successfully run ./pre after move the blockMeshDict to ../system folder, and then when I run the solver "buoyantReactingFoam" it prompts an error: 'FOAM FATAL IO ERROR: Keyword Pimple is undefined in dictionary /system/fvSolution'