[Gate-users] "Dump for solid" problem

Niklas Rehfeld rehfeld at imnc.in2p3.fr
Fri Oct 2 10:36:52 CEST 2009


Hi,

So it is quite sure that it is not a rounding error bug in G4Tubs (quite  
unlikely anyway). I would guess that the particle already enters the  
volume with NaN direction. Perhaps NaN was'nt checked before. Look into  
your macro if somewhere there are division by 0 or values that could lead  
to a division by zero...

There is a small bug in GATE 4x and older GATE 5x that could potentially  
lead to undefined (even NaN) emission angles.

Replace in the file GateSource.cc

GateSource::GateSource(G4String name)
   :m_name(name),m_simpleDecay(0),m_particleVector(0)

by

GateSource::GateSource(G4String name)
   :m_name(name),m_simpleDecay(0),m_particleVector(0),m_accolinearityFlag(false),m_accoValue(0)

and recompile.

Otherwise I have no idea.

All the best


Niklas



On Fri, 02 Oct 2009 10:18:17 +0200, Benoît Collette <bcollett at ulb.ac.be>  
wrote:

> The Monte Carlo seed is changed in my simulation :
>
> #       C H A N G E    T H E     S E E D (1)  O R   N O T (0)
> /gate/output/root/setSaveRndmFlag 1
>
>
> Benoît
>
>
> Niklas Rehfeld a écrit :
>> Dear Benoît,
>>
>> Do you obtain the same error if you change the Monte Carlo seed?
>>
>> All the best
>>
>> Niklas
>>
>>
>>
>> On Fri, 02 Oct 2009 09:36:27 +0200, Benoît Collette  
>> <bcollett at ulb.ac.be> wrote:
>>
>>> Hello,
>>>
>>> I've splited the simulation to don't have an overlapping but my  
>>> problem still exists. The error message is :
>>>
>>> -----------------------------------------------------------
>>>     *** Dump for solid - Phantom_S ***
>>>     ===================================================
>>>  Solid type: G4Tubs
>>>  Parameters:
>>>     inner radius : 0 mm
>>>     outer radius : 108 mm
>>>     half length Z: 93 mm
>>>     starting phi : 0 degrees
>>>     delta phi    : 360 degrees
>>> -----------------------------------------------------------
>>> Position:
>>>
>>> p.x() = -10.91954980241942 mm
>>> p.y() = -5.601603180160216 mm
>>> p.z() = -42.20623432221317 mm
>>>
>>> Direction:
>>>
>>> v.x() = nan
>>> v.y() = nan
>>> v.z() = nan
>>>
>>> Proposed distance :
>>>
>>> snxt = 9e+99 mm
>>>
>>>
>>> *** G4Exception : Notification
>>>       issued by : G4Tubs::DistanceToOut(p,v,..)
>>> Undefined side for valid surface normal to solid.
>>> *** This is just a warning message.
>>>
>>>  *** Break *** segmentation violation
>>> Using host libthread_db library "/lib64/tls/libthread_db.so.1".
>>> Attaching to program: /proc/32511/exe, process 32511
>>>
>>> ...
>>>
>>> -----------------------------------------------------------
>>>
>>> Could anyone help me ?
>>>
>>> Thanks a lot.
>>>
>>>
>>> Benoît COLLETTE, M.Sc.
>>> Biomedical engineer
>>>
>>> Department of Nuclear Medicine
>>> Erasmus Hospital - University of Brussels (ULB)
>>> 808 Route de Lennik
>>> 1070 Brussels - BELGIUM
>>> Phone : +32 2 555 7068
>>> Fax : +32 2 555 6800
>>> Benoit.Collette at ulb.ac.be
>>>
>>>
>>> Benoît Collette a écrit :
>>>> Hello,
>>>>
>>>> I'm using GATE v. 4.0.0 and I'm trying to modelise a Jaszczak SPECT  
>>>> Phantom Deluxe (  
>>>> http://www.biodex.com/radio/phantoms/phantoms_750.htm ). The geometry  
>>>> I've created is very simple : a cylinder containing activity and,  
>>>> inside the cylinder, 6 spheres also containing activity. But, the  
>>>> execution is stopped and I receive error messages (see  
>>>> "testVG.output"). I think it's linked with the overlaping of the  
>>>> source in the cylinder and the sources in the spheres. But I don't  
>>>> know how to do this otherwise ...
>>>>
>>>> Could somebody explain me how to define a source with a geometry like  
>>>> a cylinder extruded for the spheres ? Or how to do this otherwise  
>>>> with the version 4.0.0 ?
>>>>
>>>> Thanks a lot.
>>>>
>>>>
>>>>
>>>> Benoît COLLETTE, M.Sc.
>>>> Biomedical engineer
>>>>
>>>> Department of Nuclear Medicine
>>>> Erasmus Hospital - University of Brussels (ULB)
>>>> 808 Route de Lennik
>>>> 1070 Brussels - BELGIUM
>>>> Phone : +32 2 555 7068
>>>> Fax : +32 2 555 6800
>>>> Benoit.Collette at ulb.ac.be
>>> _______________________________________________
>>> Gate-users mailing list
>>> Gate-users at lists.healthgrid.org
>>> http://lists.healthgrid.org/mailman/listinfo/gate-users
>>
>>
>




More information about the Gate-users mailing list