Message: Scoring mesh causes Step's starting point to move Not Logged In (login)
 Next-in-Thread Next-in-Thread
 Next-in-Forum Next-in-Forum

None Scoring mesh causes Step's starting point to move 

Forum: Biasing and Scoring
Date: 08 Jul, 2016
From: Jayson Vavrek <Jayson Vavrek>

Hi all,

I'm trying to implement a scoring mesh to tally the passageCellCurrent in a simulation of bremsstrahlung and lead shielding. If I run/beamOn without setting up any scoring, I get the results I expect. However, once I add the scoring mesh, I get approximately one warning on every event:

-------- WWWW ------- G4Exception-START -------- WWWW -------
*** G4Exception : GeomNav1002
      issued by : G4Navigator::ComputeStep()
May lead to a crash or unreliable results.
        Position has shifted considerably without notifying the navigator !
        Tolerated safety: 1e-06
        Computed shift : 1.34858e+07
*** This is just a warning message. ***
-------- WWWW -------- G4Exception-END --------- WWWW -------


-------- WWWW ------- G4Exception-START -------- WWWW -------
*** G4Exception : GeomNav1002
      issued by : G4Navigator::ComputeStep()
Accuracy error or slightly inaccurate position shift.
     The Step's starting point has moved 3894.49 mm
     since the last call to a Locate method.
     This has resulted in moving 3671.88 mm from the last point at which the safety was calculated
     which is more than the computed safety= 0 mm at that point.
     This difference is 3671.88 mm.
     The tolerated accuracy is 1e-06 mm.

*** This is just a warning message. ***
-------- WWWW -------- G4Exception-END --------- WWWW -------


I know typically these warnings occur via geometry overlaps, but /geometry/test/run still gives the all-clear. I can, however, see a change in the trajectories when I run with the scoring; more trajectories escape from my source shielding, suggesting that their vertices are indeed being displaced. I have also noticed that when running beamOn with scoring, I don't just see the beamOn command, but instead the following:

/run/beamOn 1000000
/run/geometryModified

More bizarre still, the degree to which the behaviour changes seems to depend on the boxSize of the mesh. Even for similar boxSize parameters, I can get reasonable-looking (but still wrong) results or almost no tallies at all.

For reference, here are the relevant lines from the macro:
/score/create/boxMesh boxMesh_1
/score/mesh/boxSize 10. 10. 25. cm
/score/mesh/nBin 20 20 100
/score/mesh/translate/xyz 56.0876 -4.35 61.5173 cm
/score/mesh/rotate/rotateY 55 deg

/score/quantity/passageCellCurrent current
/score/close
/run/beamOn 1000000
/score/dumpQuantityToFile boxMesh_1 current score.txt
/score/drawProjection boxMesh_1 current

And lastly, here is some output of the recommended /tracking/verbose 1:

*********************************************************************************************************
* G4Track Information: Particle = gamma, Track ID = 1, Parent ID = 0
*********************************************************************************************************

Step# X(mm) Y(mm) Z(mm) KinE(MeV) dE(MeV) StepLeng TrackLeng NextVolume ProcName
    0 2.53 2.55 0.051 1.92 0 0 0 CuPlate initStep
    1 13.3 1.41 10.1 1.92 0 14.8 14.8 TheHall Transportation
    2 546 -54.8 503 1.92 0 728 743 TheHall boxMesh_1
    3 554 -55.6 510 1.92 0 10.3 753 TheHall boxMesh_1
    4 555 -55.7 512 1.92 0 2.1 756 TheHall boxMesh_1
    5 561 -56.4 517 1.92 0 8.16 764 TheHall boxMesh_1
    6 569 -57.2 524 1.92 0 10.3 774 TheHall boxMesh_1
    7 573 -57.6 528 1.92 0 5.25 779 TheHall boxMesh_1
    8 574 -57.7 529 1.92 0 1.42 781 StainlessSteelHousing1 Transportation
    9 576 -57.9 530 1.92 0 2.57 783 TheHall Transportation
   10 576 -57.9 530 1.92 0 0 783 TheHall Transportation
   11 576 -58 531 1.92 0 1.03 784 TheHall boxMesh_1
   12 584 -58.7 538 1.92 0 10.3 794 TheHall boxMesh_1
   13 590 -59.4 544 1.92 0 8.41 803 TheHall boxMesh_1
   14 591 -59.5 545 1.92 0 1.86 805 TheHall boxMesh_1
   15 599 -60.3 552 1.92 0 10.3 815 TheHall boxMesh_1
   16 607 -61.1 559 1.92 0 10.3 825 TheHall boxMesh_1
   17 607 -61.2 560 1.92 0 1.3 827 TheHall boxMesh_1
   18 614 -61.9 566 1.92 0 8.97 836 TheHall boxMesh_1
   19 622 -62.7 573 1.92 0 10.3 846 TheHall boxMesh_1
   20 625 -63.1 576 1.92 0 4.45 850 TheHall boxMesh_1
   21 629 -63.5 580 1.92 0 5.8 856 TheHall boxMesh_1
   22 629 -63.5 580 1.92 0 0.0126 856 TheHall boxMesh_1
   23 637 -64.3 587 1.92 0 10.3 866 TheHall boxMesh_1
   24 642 -64.9 592 1.92 0 7.6 874 TheHall boxMesh_1
   25 644 -65.1 594 1.92 0 2.66 877 TheHall boxMesh_1
   26 652 -65.9 600 1.92 0 10.3 887 TheHall boxMesh_1
   27 659 -66.7 607 1.92 0 10.3 897 TheHall boxMesh_1
   28 659 -66.7 608 1.92 0 0.487 898 TheHall boxMesh_1
   29 667 -67.5 614 1.92 0 9.78 907 TheHall boxMesh_1
   30 667 -67.5 615 1.92 0 1.03 908 StainlessSteelHousing1 Transportation
   31 669 -67.7 617 1.92 0 2.57 911 TheHall Transportation
   32 674 -68.3 621 1.92 0 6.67 918 TheHall boxMesh_1
   33 677 -68.5 624 1.92 0 3.64 921 TheHall boxMesh_1
   34 682 -69 628 1.92 0 6.62 928 TheHall boxMesh_1
   35 689 -69.8 635 1.92 0 10.3 938 TheHall boxMesh_1
   36 694 -70.4 640 1.92 0 6.79 945 TheHall boxMesh_1
   37 697 -70.6 642 1.92 0 3.47 948 TheHall boxMesh_1
   38 815 -83.1 751 1.92 0 161 1.11e+03 ParallelLayerDet2Extra Transportation
   39 817 -83.3 753 1.92 0 2.8 1.11e+03 ParallelLayerDet2Extra compt
   40 827 -84.5 763 1.45 0 14.3 1.13e+03 ParallelLayerDet2Extra compt
   41 834 -85.8 765 1.22 0 6.69 1.13e+03 ParallelLayerDet2Extra compt
   42 851 -90.2 764 0.217 0 18.1 1.15e+03 ParallelLayerDet2Extra compt
   43 850 -89.7 764 0 0.088 1.11 1.15e+03 ParallelLayerDet2Extra phot

Has anyone observed this behaviour before?

Thanks,
Jayson V.

Inline Depth:
 1 1
 All All
Outline Depth:
 1 1
 2 2
 All All
Add message: (add)

1 More: Re: Scoring mesh causes Step's starting point to move   (Jayson Vavrek - 11 Jul, 2016)
2 None: Re: Scoring mesh causes Step's starting point to move   (Tsukasa Aso - 13 Jul, 2016)
1 None: Re: Scoring mesh causes Step's starting point to move   (Jayson Vavrek - 13 Jul, 2016)
 Add Message Add Message
to: "Scoring mesh causes Step's starting point to move"

 Subscribe Subscribe

This site runs SLAC HyperNews version 1.11-slac-98, derived from the original HyperNews