[Gate-users] GeomTest Error: Overlapping daughter volumes and overshooting daughter volume

Ida Häggström ida.haggstrom at radfys.umu.se
Wed Sep 30 15:33:26 CEST 2009


Dear all,
I'm still working with the problem, but I have found some small "solutions"
that get rid of most errors. If you change some dimensions, e.g

/gate/module/geometry/setYLength 7.999999 cm
/gate/module/geometry/setZLength 9.999999 cm

instead of

/gate/module/geometry/setYLength 8 cm
/gate/module/geometry/setZLength 10 cm

Den 24 september 2009 15.53 skrev Ida Häggström <ida.haggstrom at radfys.umu.se
>:

> Dear Simon and Niklas,
> thank you for your quick replies! Ok, so it seems that the geometry
> overlaps might be the root to my segmentation violations then... The reason
> I checked the benchmark camera is just cause I ran another one as well, and
> thought that maybe there was some fault in that setup. When I found errors
> in the benchmark as well I got really confused... anyway, I will try and dig
> deeper into this and see how I can change the camera setup to avoid this.
> Hopefully this will enable me to run with no problems! =) Thank you for your
> valuable insight, and I'll let you know how it works out!
> Sincerely,
> Ida
>
> Den 24 september 2009 10.58 skrev Simon Stute <gate.stute at gmail.com>:
>
> Hi Ida,
>>
>> I have never checked if there is any overlap in the benchmarkPET camera,
>> but I know the Geant4 policy for overlapping volumes.
>> They strickly say that the behaviour of the code in such cases is unknown,
>> and that one must avoid making overlaps in volume geometry.
>>
>> I have also already experienced some overlaps (between phantom and
>> shielding) and it always resulted in Seg Viol, until I found this overlap.
>> So it could be that !
>>
>> Good luck,
>> Simon
>>
>> 2009/9/24 Ida Häggström <ida.haggstrom at radfys.umu.se>
>>
>>>  Hello!
>>> I run a main macro in Gate where I have a camera, phantom, source etc
>>> described. I added the line "/geometry/test/recursive_test" after I
>>> initialize the simulation in my main macro file, to check that the geometry
>>> setup is ok. I get some errors like:
>>>
>>> GeomTest Error: Overlapping daughter volumes
>>>     The volumes module_P[10] and module_P[9],
>>>     both daughters of volume head_P[0],
>>>     appear to overlap at the following points in global coordinates:
>>> (list truncated)...
>>> ....
>>> GeomTest Error: Overshooting daughter volume
>>>     The volume LSO_P[0] appears to extend outside the mother volume
>>> crystal_P[0]
>>>     at the following points in global coordinates: (list truncated)...
>>> ....
>>>
>>> for different parts of my camera and phantom. Even when I comment out the
>>> phantom and just run the camera I get errors. I find this a bit weird since
>>> I've just copied the camera from the benchmarkPET straight off, no changes.
>>> Maybe these errors aren't important and can be ignored, I don't know. Can
>>> these cause segmentation violations and abort entire runs as I described in
>>> a previous post ("segmentation violation aborts my run")?
>>> Thanks!
>>> Ida
>>>
>>> _______________________________________________
>>> Gate-users mailing list
>>> Gate-users at lists.healthgrid.org
>>> http://lists.healthgrid.org/mailman/listinfo/gate-users
>>>
>>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.opengatecollaboration.org/mailman/private/gate-users/attachments/20090930/5bf2ce1d/attachment.htm>


More information about the Gate-users mailing list