lpsd correction

rowlesmr

If I use the lpsd correction (in this case through the GUI), I choose the size of the detector, the divergence angle, and I can also choose to correct for beam spill and intensity.

I'm assuming that the beam spill correction alters the peak shape such that it doesn't change whilst the beam spills over the sample, and that the intensity correction corrects for the fact that the illuminated area doesn't change whilst the beam is spilling over.

The resultant code is:

   lpsd_th2_angular_range_degrees  4
      lpsd_equitorial_divergence_degrees  1
      lpsd_equitorial_sample_length_mm  20
         lpsd_beam_spill_correct_intensity 1

The tech ref describes them as:

lpsd_th2_angular_range_degrees: corresponds to the angular range of the LPSD in 2Th degrees.
lpsd_equitorial_divergence_degrees: Equatorial divergence in degrees of the primary beam.
lpsd_equitorial_sample_length_mm: Lengths of the sample in the equatorial plane.
lpsd_beam_spill_correct_intensity: Corrects intensity deviations inherent to variable slits when !E is set to 1.

Is lpsd_beam_spill_correct_intensity purely a variable slit correction? I can't see how this would work, unless I gave lpsd_equitorial_divergence_degrees as a function of 2Th in order to get a constant illuminated length.

alancoelho

Hi Matthew

Its for fixed divergence slits.

I cant remember (I could check of course) but you my be able to model variable divergence slits for an LPSD by writing an equation for lpsd_equitorial_sample_length_mm that is a function of 2Th.

cheers
alan

rowlesmr

Ta for that Alan.

Taking guidance from the variable_divergence macros, wouldn't you need to have an equation for the divergence as well?

I don't know how the intensity correction differs for the lpsd vs point detector, but is it still just Sin(Th)?

alancoelho

Matthew

The lpsd convolution corrects the intensity for beam spill for a fixed divergence slit. You could on top of that (I think) write and equation for divergence to cater for variable divergence slits. It should work.

cheers
alan

rowlesmr

Done. Adding it to the wiki in a moment.

'macro for doing variable divergence (fixed illumination length) with the inbuilt lpsd convolution
' Inputs required: illumination length (mm), and detector size (degrees)
' Corrects for both peak shape and intensity.
macro Variable_Divergence_lpsd(lenv, detv) { Variable_Divergence_lpsd(, lenv, , detv) }
macro Variable_Divergence_lpsd(lenc, lenv, detc, detv) {
   #m_argu lenc 'illuminated length in mm
   #m_argu detc 'size of detector in degrees
   If_Prm_Eqn_Rpt(lenc, lenv, min 0.0001 max 60)
   If_Prm_Eqn_Rpt(detc, detv, min 0.0001 max 179.9999)

	lpsd_th2_angular_range_degrees =CeV(detc, detv);
      lpsd_equitorial_divergence_degrees  =(-2 ArcTan((Rp-Sqrt(Rp^2+CeV(lenc,lenv)^2 Cos(Th)^2)) Sin(Th) / (CeV(lenc,lenv) Cos(Th)^2))) Rad;
      lpsd_equitorial_sample_length_mm  =Sin(Th);
         lpsd_beam_spill_correct_intensity 1
}

alancoelho

I think Matthew you mean:

lpsd_equitorial_sample_length_mm = CeV(lenc, lenv);

I just tested and it looks good; the plot in the attached file is for two cases:

Variable_Divergence_lpsd(30,5)
Variable_Divergence_lpsd(3, 5)

cheers
alan

rowlesmr

If I use Sin(Th) as the length, and then switch on the intensity correction, then the intensity is corrected correctly.

I'll send you my input file, data, and a little more explanation.

jlipp20

Dear all,
How can I get under the hood to modify the beam spill code while using the LPSD correction? If I use the beam spill correction with the Point Detector model, then the .out file calls the 'Divergence_Sample_Length' macro in the topas.inc file, which I can edit (I scale the peaks point-by-point using the "scale_phase_X" keyword and use a more complex two-dimensional scaling formula which accounts for the circular nature of our sample holder). However, the LPSD model seems to be more of a black box. When I use the beam spill correction with the LPSD model, the .out file calls the 'lpsd_beam_spill_correct_intensity' macro which I cannot find in the topas.inc, interface.inc, startup.def, or any other TOPAS file which I have checked.
The TOPAS6 manual only says about the lpsd_beam_spill what Matthew posted in this thread back in 2015, which does not help me find the LPSD beam spill code. Here are my specific questions; I would appreciate any help with these:

1. Is the LPSD beam spill code accessible? If so, how?
2. Do I assume correctly that the LPSD beam spill code does not scale peaks point by point?
3. Do I assume correctly that the LPSD beam spill code is the simplified one-dimensional approximation used in the Point Detector Model's 'Divergence_Sample_Length' macro?

I can answer questions 2 and 3 myself if I learn how to view the LPSD beam spill code.
Thanks!
- Jeremiah

rowlesmr

Hi Jeramiah

The code behind the LPSD correction is in the Topas kernel and activated by the relevant keyword. I can't comment on the other two points, but I would assume they follow the same as the original Variable_Divergence macro.

Matthew