Thunderhead Engineering Forum

Please login or register.

Login with username, password and session length
Advanced search  


Forum moved to

Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.

Messages - Dave McGill

Pages: [1] 2 3 ... 12
PyroSim / Re: Stop - Resume
« on: August 06, 2020, 02:09:28 pm »
It depends.

Restarting a simulation requires a .RESTART file for each mesh in the simulation. PyroSim has a default setting to have FDS write these files every 300 s (Analysis -> Simulation Parameter -> Output -> Restart). So, with the default setting, the simulation has to run for > 300s for a restart to be possible.  If the simulation runs for 599 s and dies due to a power failure, then it will restart at 300 s. If it runs to 601 s, then it will restart at 600 s.

If you run FDS through the PyroSim interface, then there are two options; you can Kill the job or Stop the job. If you choose the latter, then the .RESTART files are written as the job shuts down.

See section 7.3 of the FDS Users Guide.

The developers describe the restart feature as being fragile; it has become more reliable over the years, but the longer the simulation has run, the less likely you are to be able to successfully restart it.


PyroSim / Re: Layered Surface Fire Not Extinguishing With Sprinkler
« on: July 28, 2020, 02:54:09 pm »

If you increase the value of the extinguishing coeficient then the fire will diminish more, and may go out. Just be aware that the fire is going out because you told it to go out with an arbitrarily large value of E_COEFFICIENT.

If you haven't done so, then read about E_COEFFICIENT in the FDS Users Guide.

I started it, and it ran.

PyroSim / Re: Problems in the results
« on: July 23, 2020, 05:27:18 pm »
Hi Felipe,

There are a couple of obvious issues.

1. There is one mesh with over 3 million cells. There is no way that can run. Sub-divide it to match the number of processors you have.

2.  The model and mesh are located, literally, hundreds of km from the origin, in all three dimensions. (See the attached image.) You've probably run up against a limit of either FDS or PyroSim. Use the copy/move feature to put the lower left corner of the mesh on the origin (0,0,0).


That is a very ambitious project for a beginner. I suggest you read Dan Swenson's notes on modelling jet fans.

PyroSim / Re: Sprinkler
« on: July 03, 2020, 10:04:01 am »
The short answer is no.

By "one big fire" do you mean a burner? A sprinkler will have no effect upon a default burner, or surface, for that matter. If you put in a value for the extinguishing coefficient (E_COEFFICIENT), then the sprinkler will influence the burner, and may in fact put it out.

The problem is that nobody knows the values for extinguishing coefficient. (Read about it in the FDS Users Guide)

PyroSim / Re: Modeling fuel flowing into a drain
« on: July 01, 2020, 01:22:14 pm »
At first glance, there are a couple of problems. A sloped surface can't be modelled. (It would look sloped in PyroSim, but would be stair-stepped in FDS. It's not possible to have anything move in FDS. You could create a series of burner vents to progressively appear on the descending steps to simulate the fuel flow, but you would be prescribing the events, FDS would not be modelling them.

That shouldn't be happening. Excel has recognized .cvs files and automatically parsed them out for at least the last 10 years. When you look at the folder in Windows Explorer, you should see the Excel symbol beside all of the .cvs files.

If you don't, then change the file association.

right-click -> Properties -> General -> Opens with -> Change -> select Excel

After that, just double click the file to open it in Excel, with the columns properly parsed.

PyroSim / Re: Assigning two controls to an obstruction, hole etc...
« on: June 13, 2020, 08:37:51 am »
There is probably an elegant way to do it with custom controls.

This is the inelegant way.
I'm assuming there is enough room in the roof to make duplicates of all of the vents, so that there are pairs. (I'll refer to them as set A and set B.

Have all of the set A vents activated by individual heat detectors.

Have all of the B vents activated at the same prescribed time.

At the prescribed time, when all of the B vents are activated, also create obstructions that block all of the A vents (whether they had been created or not).

That should be equivalent to what you want.

PyroSim / Re: How to see when fire spread to new objects?
« on: June 12, 2020, 02:00:57 pm »
I suggest a boundary file of burning rate. To make it as clear as possible, in the &MISC line set BNDF_DEFAULT=.FALSE.. (This turns off the display for everything that is not specifically tagged.) Then for each of your car obstructions, use the Advanced feature to set BNDF_OBST=.TRUE. (This tags specific OBSTs to display the burning rate.)

If the OBSTs ignite, then it will be displayed on a cell by cell basis.

PyroSim / Re: Different types of modeling fire and fire propagation
« on: May 23, 2020, 10:23:15 am »
If you want the doors to "burn away", then you will have to have the combustion governed manually by the surface properties.

Until that happens, the room is still a sealed box. There might not be enough air in the room to support enough combustion to burn away a portion of the door.

If there is, then you need to expand the domain by a couple of meters on the side with the door on fire, and you need to put an open vent on that boundary of the domain.

I suggest you read about all of these features in the FDS Users Guide.

PyroSim recognizes that a surface has been labelled in Revit, but the surfaces created from the Revit file have no relevant properties.

PyroSim / Re: Different types of modeling fire and fire propagation
« on: May 22, 2020, 05:22:25 am »

The most obvious issue is that your domain is a sealed box.

PyroSim / Re: Different types of modeling fire
« on: May 21, 2020, 01:11:12 pm »
A burner can be a vent, or the face of an obstruction. It ejects fuel. With the default simple chemistry mixing model, if the fuel mixes with oxygen in the correct proportions, then combustion will occur. This is modelled on a cell-by-cell basis. No ignition source is required.

It is possible to specify an ignition temperature for a fuel or an initialization region. In those cases, a source of heat to ignite the fuel/oxygen mixture is required. That can be supplied by an ignitor, which is essentially a hot particle, placed above the source of fuel.

PyroSim / Re: Isothermal surface
« on: May 11, 2020, 12:04:50 pm »
Check the FDS Users Guide.

Pages: [1] 2 3 ... 12