Message: Re: dose depth curve  Not Logged In (login) 
On Fri, 12 Sep 2008 16:46:49 GMT, Daniel Braun wrote:
> Thank you for your answer. > > I had a look at the example you recommended but as I get it right, my > Situation is a bit different. It is shown in the geometry file I > attached in Case 3. It must be remembered that the shieldose spherical shell method is derived from an analytical expression from the slab geometry. (See the various Shieldose references from the SPENVIS website). Reproducing the spherical shell results is prone to difficulties. Another difference in the simulation to consider: shieldose gives results for infinitesimally small targets (Si, SiO2, etc.). The target must be "insignificantly" small, yet you have to be able to hit it enough to get good statistics. Additionally, the dose for the slab geometries is calculated from the shielded flux spectrum exiting the slab thickness folded with the dE/dX curve of the target  the target itself is not included in the simulation.
> > I'm shoot a fluence of particles ,I calculated with Spenvis, from the > outside on the sphere. If it is a simple result that you wish, then I'd recommend using Mulassis (also available to run on SPENVIS, and source code/local installation is available from the QinetiQ site). I've made comparable simulations to SHIELDOSE using Mulassis (particularly for protons, electrons requires some playing with the energy biasing to get a good statistical sampling of the effects of the various energy ranges).
> > And I want to get a curve like the one I calculated with Spenvis > (Ionizing and nonionizing dose models for simple geometries). I also > attached the curve (sd2_dose). You do realise that for the slab or spherical geometry that you have to run the complete simulation for each of the shielding thicknesses in the curve? This can be a bit tedious (and slow) :(.
> > I modeled the same geometry in Geant4 and shoot the same particles on > it. But the result isn't the same. But I guess it should be. > > Is there a way to create a model with G4 that delivers the same results > as Spenvis? > Yes. but the simulation isn't an exact match to the SHIELDOSE geometry. Regards, Hugh

Inline Depth:  Outline Depth:  Add message: 
to: 