Problem 2390 - Wrong formula for variance calculation
Summary: Wrong formula for variance calculation
Status: RESOLVED FIXED
Alias: None
Product: Geant4
Classification: Unclassified
Component: processes/electromagnetic/standard (show other problems)
Version: 10.7
Hardware: All All
: P4 normal
Assignee: Vladimir.Ivantchenko
URL:
Depends on:
Blocks:
 
Reported: 2021-06-19 20:14 CEST by Konstantin
Modified: 2021-08-04 14:03 CEST (History)
0 users

See Also:

Attachments
derivation of the variance formula (119.26 KB, application/pdf)
2021-06-19 20:14 CEST, Konstantin 		Details
Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this problem.
Description Konstantin 2021-06-19 20:14:01 CEST 
Created attachment 703 [details]
derivation of the variance formula

Wrong formula for variance calculation in file /geant4.10.07.p01/source/processes/electromagnetic/standard/src/G4UniversalFluctuation.cc (line 149). Here we have for variance:
  Var = siga^2 = K*l*Tcut*(1-beta^2/2),
  where
  K = 2 * pi * m_e * r_cl^2 * c^2 * n_el * z^2.
Instead we should use in this case (see attached file for the derivation of the formula):
  Var = K*l*Tcut*(1-beta^2/2*Tcut/Tmax).
Comment 1 Vladimir.Ivantchenko 2021-08-04 14:03:17 CEST 
Hello,

thank you for this bug report. It is fixed and the fix will be available with the next Geant4 public release. In tests, which are used for EM physics validation the effect of the fix is very limited, however in other conditions this fix may be useful.

VI