RE: Unknown error in geometry

From: Christopher Knievel (Christopher.Knievel@lightsource.ca)
Date: Thu Apr 03 2003 - 23:32:31 CEST


Dear Alberto,

Thank you very much for your reply. I understand what my problem was after you explained it to me and I went and made the changes to the input file. However it then gave me an error about tracking particles in Region 2. I also realized, after you pointed it out, about the problem of the undefined region left of plane 3 and right of plane 4 inside cylinder 1. I just went and redid my geometry now and made it a little different. Instead of enclosing the cylindrical beam stop in a cylindrical vacuum shell, I tried to enclose it in a vacuum sphere of radius 100 cm. Once again I received almost the exact same error message when I tried this approach. The error reads:

Geofar: Particle in region 3 (cell # 0) in position -5.338737686E-01 3.151553314E+00 4.610356519E+01
  is now causing trouble, requesting a step of 1.808186627E+01 cm
  to direction -5.339776322E-01 5.616004021E-01 6.320386671E-01, error count: 0
  R2: 3.196452642E+00 R3: 4.621424058E+01 cm
  X*U (2D): 2.054990259E+00 X*U (3D): 3.119422615E+01 cm
  X*UOLD(2D): 2.108308487E+00 X*UOLD(3D): -2.700856933E-01 cm
  Kloop: 39, Irsave: 3, Irsav2: 3, error code: -33 Nfrom: 5000
  old direction 6.291678888E-01 7.755555675E-01 -5.158807504E-02, lagain, lstnew, lsense, lsnsct F F F F
  Particle index 7 total energy 7.251255011E-04 GeV Nsurf 0
 Try again to establish the current region moving the particle of a 1.000000000E-06 long step
 We succeeded in saving the particle: current region is n. 3 (cell # 0)
  *** Emfgeo: Ustep, Dnear, Nreg, Newreg, Iq(np), X, Y, Z, Ddnear 1.E-06
  2.73727372 3 3 0 -0.533873769 3.15155331 46.1035652 0.

Once again I will attach the input file which I currently using. Now the error says that the problem occurs in region 3. But I am pretty sure that I have described that region correctly.
* Region 3 (beam stop)
    3 5 +1 -2 +3
This would correspond to points inside cylinder 1, to the right of plane 2 and to the left of plane 3. Therefore a point can be simulateneously >0 and <61. Once again I am at a loss and any help would be appreciated.

Chris Knievel

From: Alberto Fasso' [mailto:Alberto.Fasso@cern.ch]
Sent: Tuesday, April 01, 2003 5:15 PM
To: FLUKA Discuss (E-mail)
Subject: Re: Unknown error in geometry

Indeed, your geometry description is incorrect.
I will try to list a few things which are obviously wrong or inconsistent, but the main message is that a geometry such as the one you are trying to describe cannot be done without the use of the OR operator.

1) Region 1 (blackhole):
         7 -2 4 -6
   that means "the space inside sphere 7, which is outside cylinder 2, on the
   left of plane 4 and on the right of plane 6"
   No point can be at the same time on the left of plane 4 and on the right of
   plane 6: it is the same as saying that its Z coordinate must be both < -5 and
> 66. What you probably wanted was:
   OR 7 -2OR 7 4OR 7 -6

2) Region 2 (vacuum at the beginning):
         2 -1 3 -4 6 -5
   This would mean: "the space inside the wall of a cylindrical shell contained
   between cylinder 2 and cylinder 1, left of plane 3, right of plane 4, left
   of plane 6 and right of plane 5"
   Again, a point cannot be at the same time be left of plane 3 and right of
   plane 5. And it is not clear why you want vacuum only between 2 and 1 and not
   also inside 1. In any case, the space inside cyl. 1, left of plane 3 and
   right of plane 4 remains undefined. What you probably wanted was:
   OR 2 3 -4OR 2 6 -5OR 2 -1 6 -4

  Alberto

On Tue, 1 Apr 2003, Christopher Knievel wrote:

> Dear FLUKA users and authors:
>
> I am having a problem when I run my input with fluka. I am trying to
> simulate an electron beam interacting with a cylindrical aluminum beam
> stop. The error which I receive in the .err file is as follows:
>
> Geofar: Particle in region 3 (cell # 0) in position 5.730835694E+00 9.686117683E+00 3.982049539E+01
> is now causing trouble, requesting a step of 9.549625271E+00 cm
> to direction 1.929431495E-01 7.823632011E-01 5.921830482E-01, error count: 0
> R2: 1.125448149E+01 R3: 4.138037224E+01 cm
> X*U (2D): 8.683787525E+00 X*U (3D): 3.226480986E+01 cm
> X*UOLD(2D): -2.805983023E+00 X*UOLD(3D): 2.332732094E+01 cm
> Kloop: 49, Irsave: 3, Irsav2: 3, error code: -33 Nfrom: 5000
> old direction 4.859413563E-01 -5.772006160E-01 6.562777210E-01, lagain, lstnew, lsense, lsnsct F F F F
> Particle index 7 total energy 7.510855076E-04 GeV Nsurf 0
> Try again to establish the current region moving the particle of a 1.000000000E-06 long step
> We succeeded in saving the particle: current region is n. 3 (cell # 0)
>
> This error continues for many more cycles. In the .out file the error
> reads as:
>
> 0EXIT BEING CALLED FROM G1, NEXT REGION NOT FOUND.
> IR= 3 XB= 0.57308359E+01 0.96861185E+01 0.39820496E+02 WB= 0.19294315E+00 0.78236320E+00 0.59218305E+00 DIST= 0.48005062E+01
> IRLTGG= 0 SB= 0.57308359E+01 0.96861185E+01 0.39820496E+02 UB= 0.19294315E+00 0.78236320E+00 0.59218305E+00
> PR 10 0.57308359E+01 0.96861185E+01 0.39820496E+02 0.19294315E+00 0.78236320E+00 0.59218305E+00 3
> 3 3 1 -1 1 0 1 54 54 10
> 93 101 1 93 95 4 0 105 10 10
> -0.31547868E+02 0.48005062E+01 0.48005062E+01 0.48005062E+01
>
> I will attach my input file which I used. I understand that the error
> means that the program is having problems tracking the particle from
> point (x,y,z) to point (x',y',z') defined by the directional vectors
> given above, I just do not know what to do to correct the problem. As
> one can see from my input, the beam dump should be 61 cm long and it
> is enclosed in a cylindrical vacuum shell that is 5 cm thick. At least this is what I tried to simulate.
> Transport should also start at the front of the aluminum cylinder, which is
> the coordinate (0,0,0). If anyone could assist me with my problem, it would
> be greatly appreciated.
>
> Sincerely,
>
> Christopher Knievel
> Health, Safety and Environment Assistant
> Canadian Light Source
>
> <<dice.inp.txt>>
>

-- 
Alberto FassÚ
CERN-EP/AIP, CH-1211 Geneve 23 (Switzerland)
Phone: (41 22) 767 2398    Fax: (41 22) 767 9480   Alberto.Fasso@cern.ch




This archive was generated by hypermail 2.1.6 : Tue Jun 10 2003 - 14:47:42 CEST