Encountered too many geometry errors

[YapingQi]

Hi everyone, I build a chamber in egs_chamber package, and run with an error：

******** Encountered 1geometry errors (maximum allowed is 0).You can change this limit with the 'geometry error limit' run control input key. ****  then I changed the geometry error limit = 100000, the codes can run, but have so many info like this : 
************** Last 2 geometry steps:
old region=37, position=(0.931625,0.344161,-0.0112392), direction=(0.00965597,0.00256443,0.99995) intended step=0.115387, new region=1, step=0.0582207
old region=1, position=(0.932188,0.34431,0.0469787), direction=(0.00965597,0.00256443,0.99995) intended step=0.066012, new region=1, step=0.066012
EGS_ConeStack::howfar: called from the outside but I find x=(0.932188,0.34431,0.0469787) to be inside
layer=4 distance to planes=14.0133
distance to outer cone=7.62594e-10

******************************

can any experts tell me what's wrong with my codes? Thanks in advance!

[rtownson]

First, use egs_view to check for any "disappearing" or "speckled" surfaces. Does anything look like you have two surfaces overlapping, so particles might get confused? Are there any unexpected "holes"? What you see in egs_view is the same as what particles see.

Next, in your conestack set all of the outer radii to be equal. This might mean that you want to "pad" the outside of your shape with whatever the surrounding material is. This can greatly reduce the chance for floating point errors.

If you're still having the problem, do a thorough search through your geometries for any surfaces that are touching, and envelopes where the contents touch any surface of the envelope. Check that your source is not defined exactly on a physical surface. Sometimes you just need to offset things by a micrometer to solve the issue.

[rtownson]

Yes @mxxo is right. enhance regions = -1 is no longer supported (I see it's still in the example). To set a range of the regions to be enhancement regions, provide a start and end region as negative numbers, e.g. enhance regions = -1 -1000. Otherwise, just provide the whole list of region numbers.

[YapingQi]

@mxxo @rtownson Thank you, guys. These suggestions are helpful!

[YapingQi]

Yes @mxxo is right. enhance regions = -1 is no longer supported (I see it's still in the example). To set a range of the regions to be enhancement regions, provide a start and end region as negative numbers, e.g. enhance regions = -1 -1000. Otherwise, just provide the whole list of region numbers.

Is there any relation between enhancement and adding magnetic field? I found the calculated doses are the same when adding magnetic field. Btw, what areas the magnetic field works for? Attached are two output files with the similar above input file. Thank you in advance.
zptw30013_0T.txt
zptw30013_1.5T.txt

[YapingQi]

I also used the example input file and got the same results. The only difference between the two inputs is adding a magnetic field parameter in MC transport parameter. Any ideas about this problem? Thank you.
The input *.egsinp
try0T.txt
try1T.txt

The outputs:
try1T.txt
try0T.txt

[rtownson]

In egs_home/egs_chamber/Makefile change the line:

EGSPP_USER_MACROS = egs_chamber.macros

to instead be

EGSPP_USER_MACROS = egs_chamber.macros $(EGS_SOURCEDIR)EEMF_macros.mortran

and then recompile egs_chamber with make clean; make (on linux).