WARNING: Release 6.7.8: SPEC xxx is not in the table of pre-defined species.

[Michael]

Hi! 

I have the problem that the following warning always comes up with version 6.7.8: 

WARNING: SPEC REAC_FUEL is not in the table of pre-defined species. Any unassigned SPEC variables in the input were assigned the properties of nitrogen.

The simulation results then show that the smoke density is far too low. With version 6.7.7 there are no problems with the exact same file. What exactly has changed here in version 6.7.8? I didn't find any changes in this context in the release notes.

&REAC ID='POLYURETHANE',

      FYI='NFPA Babrauskas',
      FUEL='REAC_FUEL',
      C=6.3,
      H=7.1,
      O=2.1,
      N=1.0,
      AUTO_IGNITION_TEMPERATURE=0.0,
      SOOT_YIELD=0.1/

Even if I define a custom SPEC ID, this error message comes up. What do I have to do so that a polyurethane combustion can run according to the Simple Pyrolysis Model in connection with HRRPUA? 

Thank you for your help

Michael

[Kevin McGrattan]

The warning is just telling you that REAC_FUEL is not a listed species, and all FDS knows about it is its formula and molecular weight. All other properties are assumed to be for N2.

As for the smoke, we need a simple example.

[dr_jfloyd]

To expand a little on Kevin's response. For undefined conductivity, diffusivity, and viscosity, FDS uses the Lennard-Jones properties for N2. In the standard VLES mode these are not used for gas transport but are used in various Re, Sh, and Nu number correlations in the code. For undefined specific heat, FDS assumes gamma=1.4 (the N2 gamma). As discussed in the User's Guide this can cause issues for large molecular weights.

We allow the undefined properties as in most cases this is because of the fuel on a REAC line. For many engineering applications of FDS, the fuel mass fraction is small except perhaps right at the burner surface so the impact of errors in the assumed values is not likely to be significant. We added the warning message to try and call better attention to the discussion in the Guide on how FDS treats undefined properties. 

[Michael]

Thank you for your fast response.

I have now made a small test scenario. With complex stoichiometry I also got the warning although the SPEC ID is defined (see attachment):

WARNING: SPEC POLYURETHANE is not in the table of pre-defined species. Any unassigned SPEC variables in the input were assigned the properties of nitrogen.
WARNING: SPEC WOOD is not in the table of pre-defined species. Any unassigned SPEC variables in the input were assigned the properties of nitrogen.

During testing, however, I found out that the almost invisible smoke is a problem with Pyrosim's SMV Viewer. 

The slice view of the optical density is the same for 6.7.7 and 6.7.8. 

However, in the Pyrosim SMV Viewer, the 6.7.8 results have the smoke almost transparent, which made me think this was related to the warning. 

When viewed in the NIST SMV Viewer, the smoke is also displayed correctly with 6.7.8:

Pyrosim Viewer with 6.7.7 Results:

Pyrosim Viewer with 6.7.8 Results (Smoke nearly invisible):

Original SMV Viewer with 6.7.8 Results:

[dr_jfloyd]

The warning is written any time a non-predefined species is used, which is the case in your input file. Additionally, the warning definitely applies as you did not define any properties for the species.

[Michael]

Hello,

for information for all Pyrosim users - the answer from the Thunderhead Engineering support:

We have looked in to this issue and believe that we have identified the root cause. It appears that FDS 6.7.8 now outputs an additional Temperature quantity in the 3D Smoke data, which is overriding the smoke opacity in the visualization. Thankfully, this is easily worked around prior to our official support of 6.7.8. To fix this in the current version, you can simply disable the 3D Smoke > Temperature visualization in the PyroSim Results Viewer, as shown below.

[Michael]