Fano Problems:
Hi, just a short one (propably asked before).
What quantity scores cavity.cpp in Dose calculations exactly?
Per incident particle? Something flux-related? What are the units?
Greetings
Fano Problems:
Published
Note that the post authors are incorrect. Find our current active community here: https://www.reddit.com/r/EGSnrc
EGSnrc Google+ Archive 
The reason I’m asking is, that I’m failing in simulating a very basic fano test using cavity and g. I will post my Input files here:
g:
https://pastebin.com/fXzMu3Wc
cavity:
https://pastebin.com/0AifupbE
This gives me a dose of:
6.47576e-12 +/- 0.37 %,
and:
= 7.42565e-12 Gy cm^2 +/- 0.28%.
LikeLike
In cavity the dose is normalized per fluence. The fluence is calculated differently for different sources, so check the documentation for the source you are using. For the fano source it is the number of histories.
In g the result is per initial particle.
LikeLike
Thank you! I’m confused about the Fano mode in cavity. I thought it would guarantee CPE but I can see that the dose to my target depends on the source position for parallel beam sources. Also, in the tracks I can see that the probability for electron production is high near the source and low far away. Primary photons seem to disappear. I would have expected the primary photons to be reseted after each interaction.
LikeLike
Hi Stefan,
If memory serves me well, the Fano mode in cavity uses the regeneration technique to undo the effect of attenuation and scatter on the photons. This means one only has primary photons. And the expected result should be that the dose to the cavity equals kerma. To create the conditions under which the Fano theorem applies, one has to make sure that the different materials only differ by the density, but have the same cross sections and stopping powers. As Reid said, the normalization in cavity will depend on the source you use. If you use the same source in cavity and g then there shouldn’t be an issue with normalization. For a proper implementation of a Fano test I would recommend you read section II of Kawrakow’s 2000 paper onlinelibrary.wiley.com – onlinelibrary.wiley.com/doi/10.1118/1.598918/epdf
LikeLike
A good read, thank you! 🙂 I’ve decided to use another approach for the moment but I am still interested in this. Maybe my problem is how to define the source in g. Is there any documentation out there? I think it isn’t mentioned in PIRS702, but it’s hard to search for a code called “g”.
LikeLike
The g code is not documented. Here is a summary of the source inputs (or read the code directly in `subroutine source`):
:start source input:
incident spectrum =MONO-ENERGY, SPECTRUM,
MONO-ENERGY-LIN-RANGE or MONO-ENERGY-LOG-RANGE
incident charge=-1, 0 or 1
incident kinetic energy = E, for MONO-ENERGY
Emin, Emax, BinWidth, for MONO-ENERGY-LIN-RANGE
Emin, Emax, NPoints, for MONO-ENERGY-LOG-RANGE
spectrum file=path to file, for SPECTRUM.
Must be the standard EGSnrc spectrum file format
:stop source input:
LikeLike