I have a model of the atmosphere consisting of layers of concentric spheres of different densities (so sphere radii are very large). I start >TeV electrons at the top of the atmosphere (400 km) and keep track of photons produced via synchrotron and brem. When sending an electron straight down, if the magnetic field is on, tracking occurs as expected. If off, the electron is sent to OutofWorld at the boundary between the first and second layers. Angled incidence is OK. I recall an issue with tracking on spheres in the user's guide with the suggestion that a small gap be placed between layers to ensure proper tracking. I already have 1 micron above and below each layer in place, and increasing that gap does not help. (You can see the first micron in the tracking output below. "atmosphere" is the mother volume.) The problem is still present for straight down electrons for larger gaps and any energy. ```Specifics: PhysicsList is from examples/electromagnetic/TestEm16 Geometry (last few layers of atmosphere): layer# height above sphere radius (km) density Earth (km) REarth+Height (g/cm3) 375 339.68927398 6717.826274 1e-22 376 389.68927398 6767.826274 1e-22 377 400 6778.137 1e-22``` Electrons start at 400 km (top of layer 377) going down into layer 376. Tracking output for a straightdown electron, no magnetic field on: ```********************************************************************************************************* * G4Track Information: Particle = e-, Track ID = 1, Parent ID = 0 ********************************************************************************************************* ``` ```Step# X(mm) Y(mm) Z(mm) KinE(MeV) dE(MeV) StepLeng TrackLeng NextVolume ProcName 0 0 0 6.78e+09 1e+06 0 0 0 atmosphere initStep 1 0 0 6.78e+09 1e+06 3.06e-29 0.001 0.001 atmlayer376 Transportation 2 0 0 6.77e+09 1e+06 3.15e-19 1.03e+07 1.03e+07 OutOfWorld Transportation 1 events processed``` A normal event at incidence angles other than vertical will continue to track through the variable thickness layers (first few steps shown below for off-vertical incident track): ```********************************************************************************************************* * G4Track Information: Particle = e-, Track ID = 1, Parent ID = 0 ********************************************************************************************************* ``` ```Step# X(mm) Y(mm) Z(mm) KinE(MeV) dE(MeV) StepLeng TrackLeng NextVolume ProcName 0 0 0 6.78e+09 1e+06 0 0 0 atmosphere initStep 1 0.000223 0.000166 6.78e+09 1e+06 3.17e-29 0.00104 0.00104 atmlayer376 Transportation 2 2.29e+06 1.71e+06 6.77e+09 1e+06 3.27e-19 1.07e+07 1.07e+07 atmosphere Transportation 3 2.29e+06 1.71e+06 6.77e+09 1e+06 6.34e-29 0.00208 1.07e+07 atmlayer375 Transportation 4 1.34e+07 1e+07 6.72e+09 1e+06 1.59e-18 5.19e+07 6.26e+07 atmosphere Transportation 5 1.34e+07 1e+07 6.72e+09 1e+06 6.35e-29 0.00208 6.26e+07 atmlayer374 Transportation 6 2.46e+07 1.83e+07 6.67e+09 1e+06 6.11e-17 5.19e+07 1.15e+08 atmosphere Transportation``` Any help in getting this to work would be appreciated. Straight down events are good checks of geometry and physics since depth (g/cm2) is known. Scott Nutter