Re: [fluka-discuss]: Building an ideal pixelated X-ray detector

From: Primidis, Thomas <Thomas.Primidis_at_liverpool.ac.uk>
Date: Sun, 17 Mar 2019 22:58:25 +0000

Hello Mario,


thank you for the reply. So the idea is sound, I am happy.


I agree with you on the MGDRAW so I was wondering if I could do this with the default scorers.


I am wondering how can I get each photon to be detected by only one pixel regardless of the pixel "thickness" because close to the edges of the pixel the track can enter neighbouring pixels and add a blur. It could be some kind of flag on the photon(or any particle) or some particle killing method. I guess just having very thin pixels would be the simplest way to do that but is this adding any small number precision issue?


And yes, the detector material is vacuum(in practice I define a USRBIN in VOID). Is this of any interest?


Finally, TXX,TYY,TZZ return zero and XX,YY,ZZ return the direction cosines times 1E4. Quite peculiar but at least it looks like it does the job. Using cztrck would be a cleaner method which avoids this peculiarity.


Cheers,

Thomas

________________________________
From: Santana, Mario <msantana_at_slac.stanford.edu>
Sent: Sunday, March 17, 2019 9:56:42 PM
To: Primidis, Thomas
Cc: fluka-discuss
Subject: Re: [fluka-discuss]: Building an ideal pixelated X-ray detector

Hi,


If the pixel is deep (z axis), then particles hitting the near edge of a voxel and with the ‘right’ angle will also enter neighboring voxels(s), their track will be apportioned to several voxels so your ‘detector’ will have some resolution blur, so make them very thin, unless you are trying to reproduce a realistic detector. Anyway, what is the material of your detector? Is it ‘vacuum’?

You could probably implement the same scheme (without the DeltaZ uncertainty) via mgdraw.f routine, but I guess here the advantage is that you get a file that is easier to plot with standard tools, e.g. Flair.

Finally, are you sure ZZ is providing you with what you want? Not TZZ? Nor ‘cztrck' via \(TRACKR\) COMMON?

-M

On Mar 17, 2019, at 10:05 AM, Primidis, Thomas <Thomas.Primidis_at_liverpool.ac.uk<mailto:Thomas.Primidis_at_liverpool.ac.uk>> wrote:


Dear all,



I wish to simulate a pixelated detector that scores the photon energy/cm2/primary that “reaches” each of its pixels. To do so I have used a USRBIN as bellow. It is parallel to the XY plane.


USRBIN 10. PHOTON -21. 1.8 1.3 16.00051detector
USRBIN -1.8 -1.3 16. 360. 260. 1. &

along with fluscw.f as such:

      INCLUDE '(SCOHLP)'
      IF ( ISCRNG.EQ.2 ) THEN
      IF ( PLA.GE.ZERZER ) THEN
      FLUSCW = PLA*(ZZ*1D-4)
      ELSE
      FLUSCW = -PLA*(ZZ*1D-4)
      END IF
      ENDIF
      LSCZER = .FALSE.
      RETURN
*=== End of function Fluscw ===========================================*
      END


In short, when the scorer is a USRBIN it multiplies with the energy to get the energy fluence and then it multiplies with the Z-direction cosine to get rid of the 1/cosz of the fluence.



Could you please confirm this indeed simulates a perfect X-ray detector which scores the energy/cm2/primary that reaches its pixels? Does the pixel width on the Z axis effect my results in any way? I used a small width to avoid measuring twice any X-rays that could pass through two pixels in case they come almost parallel to the detector.



P.S.
I am using fluka2011.2x
In fluscw.f, ZZ returns the Z direction cosine multiplied by 1E4 so I have to divide to get the actual value.



Best regards,
Thomas



__________________________________________________________________________
You can manage unsubscription from this mailing list at https://www.fluka.org/fluka.php?id=acc_info
Received on Mon Mar 18 2019 - 01:08:20 CET

This archive was generated by hypermail 2.3.0 : Mon Mar 18 2019 - 01:08:26 CET