Fix computation of the radiation length and nuclear interaction lengt… by MarkusFrankATcernch · Pull Request #11060 · root-project/root

MarkusFrankATcernch · 2022-07-27T11:06:28Z

…h when ROOT uses G4 units

This Pull request:

The PR fixes the computation of the radiation length and nuclear interaction length
which are computed wrongly when ROOT uses in G4 units.

Changes or fixes:

All in TGeoMaterial.cpp
The fix is based on an initial pull request from Ivana:
#9401
which did miss the necessary changes for mixtures.

Output from constructing Iron from:

99.5 %   Element: FE      Z=26   N=56.000000   A=55.845000 [g/mole]
 0.5 %   Element: C      Z=6   N=12.000000   A=12.010700 [g/mole]

ROOT units:

    $> root.exe material_test.C\(\"ROOT\"\)
Material Iron    A=55.845 Z=26 rho=7.874 radlen=1.75666 intlen=0.0271712 index=0
TGeoMaterial   Iron
		 Density:  		7.874 [g/cm^3]
		 Radiation   Length: 	1.75666 [cm] 
		 Interaction Length: 	16.9589 [cm]

Geant4 units:

    $> root.exe material_test.C\(\"G4\"\)
Material Iron    A=55.845 Z=26 rho=7.874 radlen=17.5666 intlen=0.271712 index=3
TGeoMaterial   Iron
		 Density:  		7.874 [g/cm^3]
		 Radiation   Length: 	17.5666 [mm] 
		 Interaction Length: 	169.589 [mm]

PDG (https://pdg.lbl.gov/2020/AtomicNuclearProperties/HTML/iron_Fe.html):

Specific gravity 	        7.874 	g cm-3
Nuclear interaction length 	132.1 	g cm-2 	16.77 	cm
Radiation length 	        13.84 	g cm-2 	1.757 	cm

Checklist:

[X ] tested changes locally
updated the docs (if necessary)

This PR fixes #

…h when ROOT uses G4 units

phsft-bot · 2022-07-27T11:06:31Z

Can one of the admins verify this patch?

MarkusFrankATcernch · 2022-07-27T11:11:43Z

Small AClick test to verify that radiation length and nuclear interaction length are computed correctly when ROOT uses in G4 units. To invoke:

    $> root.exe material_test.C\(\"ROOT\"\)
or
    $> root.exe material_test.C\(\"G4\"\)

material_test.C.txt

…h when ROOT uses G4 units

agheata

@MarkusFrankATcernch it looks like this fixes several inconsistencies from the previous implementation. I trust that you have thoroughly checked the ROOT/G4 cases. However, we should prevent future bugs in material calculations and backward incompatibilities. Can you please provide a test macro validating the properties for some materials/mixtures when changing units?

MarkusFrankATcernch · 2022-07-28T09:59:07Z

Would something like this be sufficient ?
It checks three materials against the PDG values from https://pdg.lbl.gov/2020/AtomicNuclearProperties

$> root.exe test_material_units.C\(\"G4\"\)
$> root.exe test_material_units.C\(\"ROOT\"\)

test_material_units.C.txt

Whenever it would say "TEST FAILED" an error condition occurred.
On Success:

Processing ../test_material_units.C("G4")...
Info in <TGeoManager>: Changing system of units to Geant4 units (mm, ns, MeV).
 Using Geant4 system of units. 
Element: SI      Z=14   N=28.000000   A=28.085500 [g/mole]
Material Si    A=28.0855 Z=14 rho=2.329 radlen=93.5361 intlen=457.729 index=0
TEST SUCCEEDED Si       TGeoMaterial Units: Geant4 Deviation to PDG density: 0  %  RadLen:  0.17   %  IntLen:  1.6  % 
TEST SUCCEEDED Mix_1_Si TGeoMixture  Units: Geant4 Deviation to PDG density: 0  %  RadLen:  0.17   %  IntLen:  1.6  % 
TEST SUCCEEDED Mix_1_Si TGeoMixture  Units: Geant4 Deviation to PDG density: 0  %  RadLen:  0.0015 %  IntLen:  1.6  % 
TEST SUCCEEDED Mix_2_Si TGeoMixture  Units: Geant4 Deviation to PDG density: 0  %  RadLen:  0.17   %  IntLen:  1.6  % 
TEST SUCCEEDED Mix_3_Si TGeoMixture  Units: Geant4 Deviation to PDG density: 0  %  RadLen:  0.17   %  IntLen:  1.6  % 
Element: FE      Z=26   N=56.000000   A=55.845000 [g/mole]
Material Fe    A=55.845 Z=26 rho=7.874 radlen=17.5666 intlen=169.589 index=4
TEST SUCCEEDED Fe       TGeoMaterial Units: Geant4 Deviation to PDG density: 0  %  RadLen:  0.019  %  IntLen:  1.1  % 
TEST SUCCEEDED Mix_1_Fe TGeoMixture  Units: Geant4 Deviation to PDG density: 0  %  RadLen:  0.019  %  IntLen:  1.1  % 
TEST SUCCEEDED Mix_1_Fe TGeoMixture  Units: Geant4 Deviation to PDG density: 0  %  RadLen:  0.028  %  IntLen:  1.1  % 
TEST SUCCEEDED Mix_2_Fe TGeoMixture  Units: Geant4 Deviation to PDG density: 0  %  RadLen:  0.019  %  IntLen:  1.1  % 
TEST SUCCEEDED Mix_3_Fe TGeoMixture  Units: Geant4 Deviation to PDG density: 0  %  RadLen:  0.019  %  IntLen:  1.1  % 
Element: U      Z=92   N=238.000000   A=238.028900 [g/mole]
Material U    A=238.029 Z=92 rho=18.95 radlen=3.16948 intlen=114.473 index=8
TEST SUCCEEDED U        TGeoMaterial Units: Geant4 Deviation to PDG density: 0  %  RadLen:  0.11   %  IntLen:  3.8  % 
TEST SUCCEEDED Mix_1_U  TGeoMixture  Units: Geant4 Deviation to PDG density: 0  %  RadLen:  0.11   %  IntLen:  3.8  % 
TEST SUCCEEDED Mix_1_U  TGeoMixture  Units: Geant4 Deviation to PDG density: 0  %  RadLen:  0.0093 %  IntLen:  3.8  % 
TEST SUCCEEDED Mix_2_U  TGeoMixture  Units: Geant4 Deviation to PDG density: 0  %  RadLen:  0.11   %  IntLen:  3.8  % 
TEST SUCCEEDED Mix_3_U  TGeoMixture  Units: Geant4 Deviation to PDG density: 0  %  RadLen:  0.11   %  IntLen:  3.8  %

agheata · 2022-07-28T11:07:46Z

@MarkusFrankATcernch this looks quite good, but ideally, we need the macro to call both the "G4" and "ROOT" in the same session versions and compare them between each other, not only versus the PDG reference. Is it easy to do that? When this works properly, we merge this PR and add the macro as a separate PR, to be integrated with the CI/nightlies. We can do this next Monday when I'm back at work.

MarkusFrankATcernch · 2022-07-28T11:55:03Z

@agheata
I am a bit hesitant to have one TGeoManager with materials created with different systems of units....

Concerning the comparison:
if PDG == TROOTMaterial and PDG == TGeant4Material then also: TROOTMaterial == TGeant4Material

Clearly both tests must succeed.

agheata · 2022-07-28T12:03:38Z

@agheata I am a bit hesitant to have one TGeoManager with materials created with different systems of units....

We can use 2 different TGeoManager, deleting in between, and caching the results in a struct.

Concerning the comparison: if PDG == TROOTMaterial and PDG == TGeant4Material then also: TROOTMaterial == TGeant4Material

Clearly both tests must succeed.

AFAICT you only compare to PDG with some tolerance, but it is essential that the equality above is strict between TROOTMaterial == TGeant4Material

MarkusFrankATcernch · 2022-07-28T16:10:39Z

Here you go!

It checks pdg against TGeoROOT with tolerance
It checks pdg against TGeoG4 with tolerance
It checks TGeoROOT against TGeoG4 with numerical tolerance 1e-15 (some math simply is different)

$> root.exe test_material_units.C
  ------------------------------------------------------------------
  | Welcome to ROOT 6.27/01                        https://root.cern |
  | (c) 1995-2022, The ROOT Team; conception: R. Brun, F. Rademakers |
  | Built for linuxx8664gcc on Jul 27 2022, 14:22:00                 |
  | From heads/fix_radlen_g4units@v6-25-01-4723-g1344269685          |
  | With                                                             |
  | Try '.help'/'.?', '.demo', '.license', '.credits', '.quit'/'.q'  |
   ------------------------------------------------------------------

root [0] 
Processing ../test_material_units.C...
 Using ROOT system of units. 
Element: SI      Z=14   N=28.000000   A=28.085500 [g/mole]
Material Mat_Si_TGeo    A=28.0855 Z=14 rho=2.329 radlen=9.35361 intlen=45.7729 index=0
Element: FE      Z=26   N=56.000000   A=55.845000 [g/mole]
Material Mat_Fe_TGeo    A=55.845 Z=26 rho=7.874 radlen=1.75666 intlen=16.9589 index=4
Element: U      Z=92   N=238.000000   A=238.028900 [g/mole]
Material Mat_U_TGeo    A=238.029 Z=92 rho=18.95 radlen=0.316948 intlen=11.4473 index=8
Info in <TGeoManager>: Changing system of units to Geant4 units (mm, ns, MeV).
 Using Geant4 system of units. 
Element: SI      Z=14   N=28.000000   A=28.085500 [g/mole]
Material Mat_Si_G4    A=28.0855 Z=14 rho=2.329 radlen=93.5361 intlen=457.729 index=0
Element: FE      Z=26   N=56.000000   A=55.845000 [g/mole]
Material Mat_Fe_G4    A=55.845 Z=26 rho=7.874 radlen=17.5666 intlen=169.589 index=4
Element: U      Z=92   N=238.000000   A=238.028900 [g/mole]
Material Mat_U_G4    A=238.029 Z=92 rho=18.95 radlen=3.16948 intlen=114.473 index=8
TEST PASSED Si             vs. Mat_Si_TGeo    Units: TGeo Deviation  density: 0  %  RadLen:  0.17     %  IntLen:  1.6  % 
TEST PASSED Si             vs. Mix_1_Si_TGeo  Units: TGeo Deviation  density: 0  %  RadLen:  0.0015   %  IntLen:  1.6  % 
TEST PASSED Si             vs. Mix_2_Si_TGeo  Units: TGeo Deviation  density: 0  %  RadLen:  0.17     %  IntLen:  1.6  % 
TEST PASSED Si             vs. Mix_3_Si_TGeo  Units: TGeo Deviation  density: 0  %  RadLen:  0.17     %  IntLen:  1.6  % 
TEST PASSED Si             vs. Mat_Si_G4      Units: G4   Deviation  density: 0  %  RadLen:  0.17     %  IntLen:  1.6  % 
TEST PASSED Si             vs. Mix_1_Si_G4    Units: G4   Deviation  density: 0  %  RadLen:  0.0015   %  IntLen:  1.6  % 
TEST PASSED Si             vs. Mix_2_Si_G4    Units: G4   Deviation  density: 0  %  RadLen:  0.17     %  IntLen:  1.6  % 
TEST PASSED Si             vs. Mix_3_Si_G4    Units: G4   Deviation  density: 0  %  RadLen:  0.17     %  IntLen:  1.6  % 
TEST PASSED Mat_Si_TGeo    vs. Mat_Si_G4      Units: G4   Deviation  density: 0  %  RadLen:  0        %  IntLen:  0    % 
TEST PASSED Mix_1_Si_TGeo  vs. Mix_1_Si_G4    Units: G4   Deviation  density: 0  %  RadLen:  0        %  IntLen:  0    % 
TEST PASSED Mix_2_Si_TGeo  vs. Mix_2_Si_G4    Units: G4   Deviation  density: 0  %  RadLen:  0        %  IntLen:  0    % 
TEST PASSED Mix_3_Si_TGeo  vs. Mix_3_Si_G4    Units: G4   Deviation  density: 0  %  RadLen:  0        %  IntLen:  0    % 
TEST PASSED Fe             vs. Mat_Fe_TGeo    Units: TGeo Deviation  density: 0  %  RadLen:  0.019    %  IntLen:  1.1  % 
TEST PASSED Fe             vs. Mix_1_Fe_TGeo  Units: TGeo Deviation  density: 0  %  RadLen:  0.028    %  IntLen:  1.1  % 
TEST PASSED Fe             vs. Mix_2_Fe_TGeo  Units: TGeo Deviation  density: 0  %  RadLen:  0.019    %  IntLen:  1.1  % 
TEST PASSED Fe             vs. Mix_3_Fe_TGeo  Units: TGeo Deviation  density: 0  %  RadLen:  0.019    %  IntLen:  1.1  % 
TEST PASSED Fe             vs. Mat_Fe_G4      Units: G4   Deviation  density: 0  %  RadLen:  0.019    %  IntLen:  1.1  % 
TEST PASSED Fe             vs. Mix_1_Fe_G4    Units: G4   Deviation  density: 0  %  RadLen:  0.028    %  IntLen:  1.1  % 
TEST PASSED Fe             vs. Mix_2_Fe_G4    Units: G4   Deviation  density: 0  %  RadLen:  0.019    %  IntLen:  1.1  % 
TEST PASSED Fe             vs. Mix_3_Fe_G4    Units: G4   Deviation  density: 0  %  RadLen:  0.019    %  IntLen:  1.1  % 
TEST PASSED Mat_Fe_TGeo    vs. Mat_Fe_G4      Units: G4   Deviation  density: 0  %  RadLen:  2.2e-14  %  IntLen:  0    % 
TEST PASSED Mix_1_Fe_TGeo  vs. Mix_1_Fe_G4    Units: G4   Deviation  density: 0  %  RadLen:  0        %  IntLen:  0    % 
TEST PASSED Mix_2_Fe_TGeo  vs. Mix_2_Fe_G4    Units: G4   Deviation  density: 0  %  RadLen:  2.2e-14  %  IntLen:  0    % 
TEST PASSED Mix_3_Fe_TGeo  vs. Mix_3_Fe_G4    Units: G4   Deviation  density: 0  %  RadLen:  2.2e-14  %  IntLen:  0    % 
TEST PASSED U              vs. Mat_U_TGeo     Units: TGeo Deviation  density: 0  %  RadLen:  0.11     %  IntLen:  3.8  % 
TEST PASSED U              vs. Mix_1_U_TGeo   Units: TGeo Deviation  density: 0  %  RadLen:  0.0093   %  IntLen:  3.8  % 
TEST PASSED U              vs. Mix_2_U_TGeo   Units: TGeo Deviation  density: 0  %  RadLen:  0.11     %  IntLen:  3.8  % 
TEST PASSED U              vs. Mix_3_U_TGeo   Units: TGeo Deviation  density: 0  %  RadLen:  0.11     %  IntLen:  3.8  % 
TEST PASSED U              vs. Mat_U_G4       Units: G4   Deviation  density: 0  %  RadLen:  0.11     %  IntLen:  3.8  % 
TEST PASSED U              vs. Mix_1_U_G4     Units: G4   Deviation  density: 0  %  RadLen:  0.0093   %  IntLen:  3.8  % 
TEST PASSED U              vs. Mix_2_U_G4     Units: G4   Deviation  density: 0  %  RadLen:  0.11     %  IntLen:  3.8  % 
TEST PASSED U              vs. Mix_3_U_G4     Units: G4   Deviation  density: 0  %  RadLen:  0.11     %  IntLen:  3.8  % 
TEST PASSED Mat_U_TGeo     vs. Mat_U_G4       Units: G4   Deviation  density: 0  %  RadLen:  0        %  IntLen:  0    % 
TEST PASSED Mix_1_U_TGeo   vs. Mix_1_U_G4     Units: G4   Deviation  density: 0  %  RadLen:  0        %  IntLen:  0    % 
TEST PASSED Mix_2_U_TGeo   vs. Mix_2_U_G4     Units: G4   Deviation  density: 0  %  RadLen:  0        %  IntLen:  0    % 
TEST PASSED Mix_3_U_TGeo   vs. Mix_3_U_G4     Units: G4   Deviation  density: 0  %  RadLen:  0        %  IntLen:  0    % 

TEST PASSED  Hurray!!!! 0 failures detected.

test_material_units.C.txt

agheata · 2022-07-29T12:39:39Z

OK @MarkusFrankATcernch looks good, LGTM and we can commit separately your macro in the test suite, but let's do this Monday to check how to integrate it properly.

root-project#11060) * Fix computation of the radiation length and nuclear interaction length when ROOT uses G4 units * Fix computation of the radiation length and nuclear interaction length when ROOT uses G4 units (cherry picked from commit 103bb7e)

#11060) (#11103) * Fix computation of the radiation length and nuclear interaction length when ROOT uses G4 units * Fix computation of the radiation length and nuclear interaction length when ROOT uses G4 units (cherry picked from commit 103bb7e) Co-authored-by: MarkusFrankATcernch <[email protected]>

#11060) (#11104) * Fix computation of the radiation length and nuclear interaction length when ROOT uses G4 units * Fix computation of the radiation length and nuclear interaction length when ROOT uses G4 units (cherry picked from commit 103bb7e) Co-authored-by: MarkusFrankATcernch <[email protected]>

Fix computation of the radiation length and nuclear interaction lengt…

7bb3617

…h when ROOT uses G4 units

MarkusFrankATcernch requested a review from agheata as a code owner July 27, 2022 11:06

Fix computation of the radiation length and nuclear interaction lengt…

df6fd2d

…h when ROOT uses G4 units

agheata requested changes Jul 27, 2022

View reviewed changes

couet assigned agheata Jul 29, 2022

agheata approved these changes Jul 29, 2022

View reviewed changes

agheata merged commit 103bb7e into root-project:master Jul 29, 2022

MarkusFrankATcernch deleted the fix_radlen_g4units branch July 29, 2022 13:13

bcouturi mentioned this pull request Aug 1, 2022

Backport the fix on computation of the radiation and nuclear interaction lengths to v6.24 #11080

Closed

agheata mentioned this pull request Aug 3, 2022

Fix computation of the radiation length and nuclear interaction lengt… #11103

Merged

2 tasks

agheata mentioned this pull request Aug 3, 2022

Fix computation of the radiation length and nuclear interaction lengt… #11104

Merged

2 tasks

MarkusFrankATcernch mentioned this pull request Aug 4, 2022

Fix to compute radiation length on call TGeoMixture::ComputeDerivedQu… #11115

Merged

2 tasks

agheata mentioned this pull request Aug 15, 2022

Fix computation of radiation lengh in TGeoMaterial #9401

Closed

2 tasks

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Fix computation of the radiation length and nuclear interaction lengt…#11060

Fix computation of the radiation length and nuclear interaction lengt…#11060
agheata merged 2 commits intoroot-project:masterfrom
MarkusFrankATcernch:fix_radlen_g4units

MarkusFrankATcernch commented Jul 27, 2022

Uh oh!

phsft-bot commented Jul 27, 2022

Uh oh!

MarkusFrankATcernch commented Jul 27, 2022 •

edited

Loading

Uh oh!

agheata left a comment

Uh oh!

MarkusFrankATcernch commented Jul 28, 2022

Uh oh!

agheata commented Jul 28, 2022

Uh oh!

MarkusFrankATcernch commented Jul 28, 2022

Uh oh!

agheata commented Jul 28, 2022

Uh oh!

MarkusFrankATcernch commented Jul 28, 2022

Uh oh!

agheata commented Jul 29, 2022

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

Conversation

MarkusFrankATcernch commented Jul 27, 2022

This Pull request:

Changes or fixes:

Checklist:

Uh oh!

phsft-bot commented Jul 27, 2022

Uh oh!

MarkusFrankATcernch commented Jul 27, 2022 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

agheata left a comment

Choose a reason for hiding this comment

Uh oh!

MarkusFrankATcernch commented Jul 28, 2022

Uh oh!

agheata commented Jul 28, 2022

Uh oh!

MarkusFrankATcernch commented Jul 28, 2022

Uh oh!

agheata commented Jul 28, 2022

Uh oh!

MarkusFrankATcernch commented Jul 28, 2022

Uh oh!

agheata commented Jul 29, 2022

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

MarkusFrankATcernch commented Jul 27, 2022 •

edited

Loading