set to 0 in the param_card.dat error for some aTGC parameter values

Asked by senka duric

Dear experts,

I am using an BSM aTGC NLO model in MG5 v2_4_3. I can generate events for some aTGC parameter values while for some I get the error:

aMCatNLOError : Some tests failed, run cannot continue.
 Please check that widths of final state particles (e.g. top) have been set to 0 in the param_card.dat.

My process card is VBF higgs 5f:

import model EWHdim6NLO-restrict
define p = p b b~
define j = j b b~
generate p p > h j j $$ w+ w- z HIW=0 HIG=0 NP=2 QED=3 QCD=0 [QCD]

in the restrict card I set b-mass and yukawa to 0, and aTGC values to:

Block dim6
    1 1.100000e+00 # CWWWL2
    2 1.200000e+00 # CWL2
    3 1.300000e+00 # CBL2
    4 1.400000e+00 # CPWWWL2
    5 1.500000e+00 # CPWL2
    6 1.600000e+00 # CphidL2
    7 1.700000e+00 # CphiWL2
    8 1.800000e+00 # CphiBL2
    9 1.900000e+00 # CphiGL2
   10 1.010000e+00 # Cphi6L2
   11 1.020000e+00 # CtphiL2
   12 1.030000e+00 # CbphiL2

This is higgs VBF process so I modify FKS_params.dat and set "IRPoleCheckThreshold" and "PrecisionVirtualAtRunTime" to -1.
I can generate events with the default param_card.dat, and for some modified values of some parameters. Below are some parameter values that are running fine and one where it fails. Does this indicate some problem in the model or am I doing something wrong?
Thank you in advance!

working param_card:

Block dim6
    1 1.100000e+00 # CWWWL2
    2 1.200000e+00 # CWL2
    3 1.300000e+00 # CBL2
    4 1.400000e+00 # CPWWWL2
    5 1.500000e+00 # CPWL2
    6 1.600000e+00 # CphidL2
    7 1.700000e+00 # CphiWL2
    8 0.010000e+00 # CphiBL2
    9 1.900000e+00 # CphiGL2
   10 1.010000e+00 # Cphi6L2
   11 1.020000e+00 # CtphiL2
   12 1.030000e+00 # CbphiL2

Block dim6
     1 0.000000e+00 # CWWWL2
     2 4.500000e+00 # CWL2
     3 2.000000e+01 # CBL2
     4 0.000000e+00 # CPWWWL2
     5 0.000000e+00 # CPWL2
     6 6.500000e-03 # CphidL2
     7 6.500000e-04 # CphiWL2
     8 1.800000e+00 # CphiBL2
     9 6.500000e-05 # CphiGL2
    10 0.000000e+00 # Cphi6L2
    11 6.500000e-03 # CtphiL2
    12 6.500000e-03 # CbphiL2

failing param_cards:

Block dim6
     1 0.000000e+00 # CWWWL2
     2 4.500000e+00 # CWL2
     3 2.000000e+01 # CBL2
     4 0.000000e+00 # CPWWWL2
     5 0.000000e+00 # CPWL2
     6 6.500000e-03 # CphidL2
     7 6.500000e-04 # CphiWL2
     8 6.500000e-04 # CphiBL2
     9 6.500000e-05 # CphiGL2
    10 0.000000e+00 # Cphi6L2
    11 6.500000e-03 # CtphiL2
    12 6.500000e-03 # CbphiL2

Block dim6
     1 0.000000e+00 # CWWWL2
     2 4.500000e+00 # CWL2
     3 2.000000e+01 # CBL2
     4 0.000000e+00 # CPWWWL2
     5 0.000000e+00 # CPWL2
     6 6.500000e-03 # CphidL2
     7 6.500000e-04 # CphiWL2
     8 0.100000e+00 # CphiBL2
     9 6.500000e-05 # CphiGL2
    10 0.000000e+00 # Cphi6L2
    11 6.500000e-03 # CtphiL2
    12 6.500000e-03 # CbphiL2

Greetings,
Senka

Question information

Language:
English Edit question
Status:
Answered
For:
MadGraph5_aMC@NLO Edit question
Assignee:
marco zaro Edit question
Last query:
Last reply:
Revision history for this message
marco zaro (marco-zaro) said :
#1

Hi Senka,
which tests fail? test_ME?
what is the largest fraction of failure printed on screen?
Thanks!
cheers,

Marco

Revision history for this message
senka duric (senka-duric) said :
#2

Hi Marco,
the largest fraction that I have seen was 0.78:

 Collinear tests done for (Born) config 1
 Failures: 78
Collinear test 6 FAILED. Fraction of failures: 0.78

Senka

Revision history for this message
marco zaro (marco-zaro) said :
#3

Hi Senka,
these tests control the convergence of the real-emission ME to the known limit in the case of soft/collinear partons.
the check is done numerically and the effects of adding anomalous vertices, and therefore higher powers of the momentum, can make the test fail.
A (rather dangerous) solution would be either to skip these tests, or to increase the threshold for considering the test a failure.
For the latter, just open
SubProcesses/symmetry_fks_test_ME.f and change
      parameter (max_fail=0.3d0)

to
      parameter (max_fail=0.85d0)

Remember to double check the results

cheers,

Marco

Revision history for this message
senka duric (senka-duric) said :
#4

Hi Marco,
thank you for the info, I will increase the threshold. What are the checks I should perform on the sample to make sure it is ok?

Greetings,
Senka

Can you help with this problem?

Provide an answer of your own, or ask senka duric for more information if necessary.

To post a message you must log in.