Overlaps in Component-SBendFringefields test

[Jochem Snuverink]

The Component-SBendFringefields test has overlaps since Jan 31, e.g. http://abp-cdash.web.cern.ch/abp-cdash/testDetails.php?test=156348&build=32163

Checking overlaps for volume cl_143_0_cl_pv ... 
-------- WWWW ------- G4Exception-START -------- WWWW -------
*** G4Exception : GeomVol1002
      issued by : G4PVPlacement::CheckOverlaps()
Overlap with volume already placed !
          Overlap is detected for volume cl_143_0_cl_pv
          apparently fully encapsulating volume cl_141_0_cl_pv
          at the same level !
*** This is just a warning message. ***
-------- WWWW -------- G4Exception-END --------- WWWW -------

 G4PhysicsListHelper::AddTransportation()--- G4CoupledTransportation is used 

This is the last curvilinear volume.

[Jochem Snuverink]

The overlap is no longer present. Possibly due to the changed curvilinear beamline construction or the overlap fixes in the dipole poled geometry?

[Laurie Nevay]

Yes, this was due to the curvilinear world building algorithm and how it dealt with thin elements. We're back to 1:1 now so this has resolved itself. Back to 1:1 is safe as we've got padding in the beam line and also no physics process for using the boundaries of the curvilinear world that would affect tracking.