Message: Re: G4Exception : mat205 Not Logged In (login)
 Next-in-Thread Next-in-Thread
 Next-in-Forum Next-in-Forum

Idea Re: G4Exception : mat205 

Forum: Processes Involving Optical Photons
Re: Question G4Exception : mat205 (George Dedes)
Date: 16 Jun, 2014
From: Michael H. Kelsey <Michael H. Kelsey>

On Sat, 14 Jun 2014 16:56:47 GMT, George Dedes wrote:
> after having implemented the material properties in my code based on the
> LXe example, I get the following runtime error:
> -------- EEEE ------- G4Exception-START -------- EEEE -------
> *** G4Exception : mat205
>       issued by : G4MaterialPropertiesTable::SetGROUPVEL()
> Optical Photon Energy <= 0
> *** Fatal Exception *** core dump ***
> -------- EEEE -------- G4Exception-END --------- EEEE -------
> The is the class that seems to throw this
> exception, although I don't understand why/how a 0 or negative energy
> optical photon is created. Any directions/suggestion where to look for
> the problem are really welcomed!

I ran into this problem myself, but without more details about your own app, I don't know if my situation will match. The _immediate_ cause of your problem is that your RINDEX table has zeroes for one or more of the energy bin entries. You should see this by adding the statement "mpt->DumpTable();" into your code right after you've created and filled the table (here, "mpt" is your pointer).

In my case, that problem was caused because I had all of my materials being created in a static object during the kernel initialization (i.e., before the first line of main()). The order in which static and const objects are filled up is strongly compiler-dependent, and my compiler decided to create my materials manager _before_ the G4SystemOfUnits parameters had been filled! Thus, among other things, eV was zero :-(

By changing my materials manager to be a regular singleton, which got created the first time it's "Instance()" function was called, everything worked normally.

     -- Michael Kelsey

 Add Message Add Message
to: "Re: G4Exception : mat205"

 Subscribe Subscribe

This site runs SLAC HyperNews version 1.11-slac-98, derived from the original HyperNews