No BSM particles generated from MadGraph

[GabrielG]

Hi,

Please correct me if I am missing something.

I am learning how to use SARAH and MadGraph to generate events for some new physics model. I thought it would be a good idea to try out the workflow with a model built-in to SARAH.

So I would…

• Start["MSSM"] or Start["Vectorlike/BottomR"]
  MakeUFO[]
  Copy the UFO files into MadGraph/models/newmodel
  In MadGraph, import model newmodel
  generate [some process]

Now, when I do this for a Standard Model process, e.g. p p > t t~, it works fine,
but if I try for a BSM process, say p p > x1+ n2, MadGraph returns me this error message:

Survey return zero cross section.
Typical reasons are the following:
1) A massive s-channel particle has a width set to zero.
2) The pdf are zero for at least one of the initial state particles
or you are using maxjetflavor=4 for initial state b:s.
3) The cuts are too strong.
Please check/correct your param_card and/or your run_card.

I checked these and they seem okay to me, but maybe I am missing something.

Any help on this would be much appreciated.

[FStaub]

Hi,

what param card do you use? The template written by MG has many zeros, especially for all BSM parameters. I usually recommend to use a SPheno spectrum file. See for instance 1503.04200, sec. 7.5. (Note, some time after this tutorial, there has a new flag been added (78->1) in the Les Houches input to generate spc files compatible with MG).


Cheers,
Florian

[GabrielG]

Hi Florian,

Thank you for your quick reply. I'm trying to generate some processes now by using SPheno file as you suggested. I'd just like to document some of the things I'm encountering as I attempt this:

(1) No -U option for ar command
When running make Model=SARAHMODEL in SPheno, I got an error saying there was not "-U" option for the "ar" command. I found the corresponding line in the Makefile for SARAHMODEL (which I presume was generated by SARAH?) and removed the "-U" option manually and it seems to work.

(2) Invalid param_card
I have tried to MakeUFO[] and MakeSPheno[] for the SM model in SARAH.

When I try to generate a process in MadGraph, say p p > t t~, and use the SPheno.spc.SM_low file as the param_card.dat, I get this error message:

ERROR: problem detected: higgsboundsinputhiggscouplingsbosons (3, 25, 23) is already define to 23 23 impossible to assign 23 22

I don't know how to solve this. Could you help?

(3) Complex Argument to ASIN() function
If I try to launch some process in MadGraph with the default param_card, say

• import model SARAH_SM
  generate p p > t t~
  output SARAH_SM_TTBAR
  launch

I would get this error:

Code: Select all

intparam_definition.inc:322.22:
    Included at couplings.f:16:

        MDL_TW = ASIN(SQRT(DCMPLX(1.000000D+00-MDL_MWP__EXP__2                 
                      1
Error: 'x' argument of 'asin' intrinsic at (1) must be REAL

I have traced this to the file SARAH_SM_TTBAR/Source/MODEL/intparam_definition.inc file. Manually changing this line:

Code: Select all

MDL_TW = ASIN(SQRT(DCMPLX(1.000000D+00-MDL_MWP__EXP__2/MDL_MZ__EXP__2)))

into this line (i.e. casting the result of the square root to real):

Code: Select all

MDL_TW = ASIN(REAL(SQRT(DCMPLX(1.000000D+00-MDL_MWP__EXP__2/MDL_MZ__EXP__2))))

fixes the problem.

As I understand it, this intparam_definition.inc file has its roots in the parameters.m file of the SARAH model. If I try to change the corresponding line from

Code: Select all

{ThetaW,    { Description -> "Weinberg-Angle",
              DependenceNum -> ArcSin[Sqrt[1 - Mass[VWp]^2/Mass[VZ]^2]]  }},

to

Code: Select all

{ThetaW,    { Description -> "Weinberg-Angle",
              DependenceNum -> ArcSin[Re[Sqrt[1 - Mass[VWp]^2/Mass[VZ]^2]]]  }},

I will get this error in MadGraph instead:

Error detected in "output SARAH_SM_TTBAR"
write debug file MG5_debug
If you need help with this issue please contact us on https://answers.launchpad.net/mg5amcnlo
MadGraph5Error : Unable to evaluate mdl_TW = cmath.asin(Re(cmath.sqrt(1 - mdl_MWp__exp__2/mdl_MZ__exp__2))): raise error: name 'Re' is not defined

Is there any way to work around this other than to manually change the intparam_definition.inc file?

[FStaub]

Hi,

1) I guess you have an older Fortran compiler which doesn't know that flag. So yes, you need to modify the Makefile by hand.

2) You need to turn off the HiggsBounds output because the MG SLHA parser doesn't come along with these blocks

3) Also that is a compiler dependent issue. I think, instead of Re you should use Abs in parameters.m. That solved this problem for other users if I remember correctly.

Cheers,
Florian

[GabrielG]

Hi Florian,

2) How do I disable the HiggsBound output?
I've been settings "520 0" and "78 1" in the LesHouches files, but the HiggsBound parameters are still output to the spc file anyway.

3) I tried Abs[], it is not automatically converted to the correct C++/FORTRAN code. I've elected to automate adding the REAL() FORTRAN function by the sed command.

[FStaub]

I've been settings "520 0" and "78 1" in the LesHouches files, but the HiggsBound parameters are still output to the spc file anyway.

The other way round: 520->1, 78->0.

Cheers,
Florian

[GabrielG]

Hi Florian,

I am confused. The comments on 520 and 78 are:

520 1. # Write effective Higgs couplings (HiggsBounds blocks): put 0 to use file with MadGraph!
78 0 # Output for MadGraph (writes also vanishing blocks)

and setting the 520->1 and 78->0 as you suggested still gives me the same error.

ERROR: problem detected: higgsboundsinputhiggscouplingsbosons (3, 25, 23) is already define to 23 23 impossible to assign 23 22

Gabriel

[FStaub]

Yes, sorry. I mixed that up. Your settings have been correct, i.e. the Higgsbounds blocks shouldn't show up. Did I understand you correctly that they do nevertheless?!

[GabrielG]

I have looked into this further. Apparently this error was due to my own oversight (I forgot to run SPheno/bin/SPhenoVLBottomR). Setting 520->0 and 78->1 does in fact give a file without "higgsboundsinputhiggscouplingsbosons". However, now I have a new error message:

ERROR: problem detected: decay (6,) is already define to 1.55526376 impossible to assign 6 1.40435722e+00

I have tracked this to these lines in the param_card:

Code: Select all

DECAY1L         6     1.40435722E+00   # Fu_3
#    BR                NDA      ID1      ID2
     1.00000000E+00    2            7         24   # BR(Fu_3 -> Fd_4 VWp )
DECAY1L        25     6.53833083E-03   # hh
#    BR                NDA      ID1      ID2
     4.45723843E-04    2           -5          5   # BR(hh -> Fd_3^* Fd_3 )
     8.48669892E-01    2           -7          7   # BR(hh -> Fd_4^* Fd_4 )
     1.48107080E-04    2          -13         13   # BR(hh -> Fe_2^* Fe_2 )
     4.18315779E-02    2          -15         15   # BR(hh -> Fe_3^* Fe_3 )
     7.98999169E-02    2           -4          4   # BR(hh -> Fu_2^* Fu_2 )
     3.15432212E-02    2           21         21   # BR(hh -> VG VG )
     2.91505607E-03    2           22         22   # BR(hh -> VP VP )

What is DECAY1L? Is it meant to override or complement these lines?

Code: Select all

DECAY         6     1.55526376E+00   # Fu_3
#    BR                NDA      ID1      ID2
     1.00000000E+00    2            7         24   # BR(Fu_3 -> Fd_4 VWp )
DECAY        25     4.31335609E-03   # hh
#    BR                NDA      ID1      ID2
     2.65844278E-03    2           22         22   # BR(hh -> VP VP )
     8.41520007E-02    2           21         21   # BR(hh -> VG VG )
     3.12770844E-02    2           23         23   # BR(hh -> VZ VZ )
     2.69866022E-01    2          -24         24   # BR(hh -> VWp^* VWp_virt )
     1.95754539E-04    2           -5          5   # BR(hh -> Fd_3^* Fd_3 )
     5.25625175E-01    2           -7          7   # BR(hh -> Fd_4^* Fd_4 )
     2.12094833E-04    2          -13         13   # BR(hh -> Fe_2^* Fe_2 )
     6.12247492E-02    2          -15         15   # BR(hh -> Fe_3^* Fe_3 )
     2.47880221E-02    2           -4          4   # BR(hh -> Fu_2^* Fu_2 )

If I comment the DECAY1L lines out, events can be generated in MadGraph.

[FStaub]

Hi,

DECAY1L are the decays including one-loop corrections. If you don't expect large loop effects, you can turn them off via flag 16->0 and stay with tree-level decays as input for MG.

Cheers,
Florian