Backing='insulated' in input file conducting to 'exposed in the output file

[GoogleCodeExporter]

Please complete the following lines...

Application Version: FDS  5.3.1 Serial
SVN Revision Number: 3729
Compile Date: Wed, 08 Apr 2009
Operating System: Windows XP

Describe details of the issue below:

When the argument BACKING='INSULATED' is written in the SURF line, the
diagnostic file outputs 'Exposed backing'.

eg:
Input file
-----------------------
&SURF ID = 'FIRE BRICK'
MATL_ID = 'FIRE BRICK'
THICKNESS = 0.113
BACKING=’INSULATED’ /

&MATL ID = 'FIRE BRICK'
CONDUCTIVITY = 0.40
SPECIFIC_HEAT = 1.04
DENSITY = 750.
EMISSIVITY = 0.80 /

Output file
-----------------------
   1 FIRE BRICK                    
     Material List
          1  FIRE BRICK                    
     Layer  1
        Thickness   (m):  0.11300
        Density (kg/m3):   750.00
        FIRE BRICK, Mass fraction:    1.00
     Total surface density    84.750 kg/m2
     Reaction products considered from the first  0.50 layers.
     Solid Phase Nodes (m):  ..........
     Exposed Backing

Miguel

Original issue reported on code.google.com by Migou...@gmail.com on 24 Jul 2009 at 12:15

Attachments:

• backing_condition.zip

[GoogleCodeExporter]

This is what I see when I look at the file with a simple vi editor on a linux 
machine:

&SURF ID = 'FIRE BRICK'
MATL_ID = 'FIRE BRICK'
THICKNESS = 0.113
BACKING=<92>INSULATED<92> /

This is what Wordpad looks like on a Windows PC

&SURF ID = 'FIRE BRICK'
MATL_ID = 'FIRE BRICK'
THICKNESS = 0.113
BACKING=’INSULATED’ /

The quotes surrounding the word INSULATED are not included in the ASCII 
character 
set (or let's just say that I cannot make that mark using my American-style 
keyboard). It looks like these were cut and pasted from some word processing 
package. You must use pure text in the input file or else FDS (which is just 
Fortran) will not understand it. I am surprised that the case ran on my linux 
machine. Usually it flags non-standard characters with an error.

How did you write that character?

Original comment by mcgra...@gmail.com on 24 Jul 2009 at 12:32

• Changed state: MoreInfo

[GoogleCodeExporter]

I've copied it from the PDF User Guide, and that's why it didn't work.
And before posting the "issue", to be sure not to commit any error, I re-copied 
from
the PDF !

I've re-run the case with the appropriate characters and there isn't actually 
any error.
I'm very sorry to waste your time. 

Original comment by Migou...@gmail.com on 24 Jul 2009 at 12:46

[GoogleCodeExporter]

No problem. There is a warning about cutting and pasting, but I understand why 
people still do it. It is easier than having to type. I am concerned that my 
linux 
version of FDS allowed it. Usually, the calcs fail with a read error when the 
characters are non-standard.

Original comment by mcgra...@gmail.com on 24 Jul 2009 at 1:15

[GoogleCodeExporter]

I'm not sure that the NAMELIST standard requires '' for a CHARACTER input (it 
does
help prevent parsing errors though).  If that is a true statement, then from the
point of view of FORTRAN the non-standard ` may be OK.  If this is the case, 
then the
solution is to add a CASE DEFAULT to the SELECT CASE the gives an error 
message. 

Original comment by drjfloyd on 24 Jul 2009 at 2:22

[GoogleCodeExporter]

FDS did not produce an error, but it also did not register the string 
INSULATED. Do 
it interpreted the string differently based on the presence of the non-standard 
character.

Original comment by mcgra...@gmail.com on 24 Jul 2009 at 4:25

• Changed state: WontFix

[GoogleCodeExporter]

[deleted comment]

[GoogleCodeExporter]

But if it didn't register the string INSULATED, why did it apply an EXPOSED 
backing
instead of the default VOID backing ?

Original comment by Migou...@gmail.com on 24 Jul 2009 at 4:41

[GoogleCodeExporter]

Good question. When these character strings are read in, we assign an integer 
code 
number for it so that the logic in the code is more efficient -- it is better 
to do 
comparisons on integers rather than character strings. EXPOSED was code number 
0. I 
added an error statement such that if BACKING is not recognized for whatever 
reason, 
FDS stops in ERROR. This new error statement stops your case.

Original comment by mcgra...@gmail.com on 24 Jul 2009 at 5:11

• Changed state: Fixed

[GoogleCodeExporter]

Original comment by mcgra...@gmail.com on 23 Nov 2009 at 9:46

• Changed state: Verified