particle width set to -ve

Asked by arindam chatterjee on 2018-10-13

I am generating NMSSM processes, chargino neutralino_2,3 pair production in MadGraph 2.4.3, the chargino and neutralinos are to decay into the leptonic modes. The param_card is adopted from NMSSM_Tools slha output, and some blocks are edited to make it
compatible with MG5 (e.g. using 6*6 mixing matrices for squarks and sleptons, and the QNUMBERS info added). However, during event generation I am always getting this message :
WARNING: For consistency, the width of particle 1000025 (n3) is changed to *a -ve value depending on the input*.
Also, the pythia log file has this :
 * Warning: will use DECAY table (fixed-width, flat PS) for Z0
 * (PYSLHA:) Reading DECAY table for KF = 1000024, ~chi_1+
At line 757 of file pyslha.f
Fortran runtime error: End of file

Error termination. Backtrace:
#0 0x7f36ed030f3a
#1 0x7f36ed031a95
#2 0x7f36ed03224c
#3 0x7f36ed0fe353
#4 0x7f36ed0f8ee9
#5 0x561bce
#6 0x4f43c0
#7 0x461bdd
#8 0x405b8e
#9 0x406f16
#10 0x7f36ec74a82f
#11 0x402cb8
#12 0xffffffffffffffff
 * (PYSLHA:) Reading DECAY table for KF = 1000037, ~chi_2+

and

 * (PYSLHA:) Reading DECAY table for KF = 1000023, ~chi_20
 * (PYSLHA:) Reading SLHA stable particle KF = 1000025, ~chi_30

     Error type 9 has occured after 0 PYEXEC calls:
     (PYSLHA:) Negative width forced to zero !
 * (PYSLHA:) Reading DECAY table for KF = 1000035, ~chi_40

Kindly suggest how to address this issue.

regards,
arindam

Question information

Language:
English Edit question
Status:
Answered
For:
MadGraph5_aMC@NLO Edit question
Assignee:
No assignee Edit question
Last query:
2018-10-14
Last reply:
2018-10-17

Hi,

Looks like PY6 does not support negative width. You should check if this is the case for PY8.

Cheers,

Olivier

> On 13 Oct 2018, at 09:17, arindam chatterjee <email address hidden> wrote:
>
> New question #675126 on MadGraph5_aMC@NLO:
> https://answers.launchpad.net/mg5amcnlo/+question/675126
>
> I am generating NMSSM processes, chargino neutralino_2,3 pair production in MadGraph 2.4.3, the chargino and neutralinos are to decay into the leptonic modes. The param_card is adopted from NMSSM_Tools slha output, and some blocks are edited to make it
> compatible with MG5 (e.g. using 6*6 mixing matrices for squarks and sleptons, and the QNUMBERS info added). However, during event generation I am always getting this message :
> WARNING: For consistency, the width of particle 1000025 (n3) is changed to *a -ve value depending on the input*.
> Also, the pythia log file has this :
> * Warning: will use DECAY table (fixed-width, flat PS) for Z0
> * (PYSLHA:) Reading DECAY table for KF = 1000024, ~chi_1+
> At line 757 of file pyslha.f
> Fortran runtime error: End of file
>
> Error termination. Backtrace:
> #0 0x7f36ed030f3a
> #1 0x7f36ed031a95
> #2 0x7f36ed03224c
> #3 0x7f36ed0fe353
> #4 0x7f36ed0f8ee9
> #5 0x561bce
> #6 0x4f43c0
> #7 0x461bdd
> #8 0x405b8e
> #9 0x406f16
> #10 0x7f36ec74a82f
> #11 0x402cb8
> #12 0xffffffffffffffff
> * (PYSLHA:) Reading DECAY table for KF = 1000037, ~chi_2+
>
> and
>
> * (PYSLHA:) Reading DECAY table for KF = 1000023, ~chi_20
> * (PYSLHA:) Reading SLHA stable particle KF = 1000025, ~chi_30
>
> Error type 9 has occured after 0 PYEXEC calls:
> (PYSLHA:) Negative width forced to zero !
> * (PYSLHA:) Reading DECAY table for KF = 1000035, ~chi_40
>
> Kindly suggest how to address this issue.
>
> regards,
> arindam
>
> --
> You received this question notification because you are an answer
> contact for MadGraph5_aMC@NLO.

PY8 seems to have other issues with the param_card, I am looking into it.
But for PY6, why is MG5 trying to assign -ve width overwriting the param_card supplied?
Any suggestions would be very helpful.

regards,
arindam

Hi,

> But for PY6, why is MG5 trying to assign -ve width overwriting the param_card supplied?
> Any suggestions would be very helpful.

This is because your UFO model states that the width of such particle is not a free parameter of the model
but is fixed to the other parameter of the model.
Having that parameter to any other value than the one given by the model will lead to inconsistency in the computation.

If this is not the behaviour that you want (or the relation that you want), you will need to change your model (or use another one)

Cheers,

Olivier

> On 13 Oct 2018, at 23:02, arindam chatterjee <email address hidden> wrote:
>
> Question #675126 on MadGraph5_aMC@NLO changed:
> https://answers.launchpad.net/mg5amcnlo/+question/675126
>
> Status: Answered => Open
>
> arindam chatterjee is still having a problem:
> PY8 seems to have other issues with the param_card, I am looking into it.
> But for PY6, why is MG5 trying to assign -ve width overwriting the param_card supplied?
> Any suggestions would be very helpful.
>
> regards,
> arindam
>
> --
> You received this question notification because you are an answer
> contact for MadGraph5_aMC@NLO.

Thanks. However, that would be true for chargino1 or neutralino2, the other particles I have considered for production as well.
For these particles there is no issue, and the decay width and br information in the param_card is read. Only for neutralino3 (which also happens to have a -ve eigenvalue) this error shows up. In the models folder, the models/nmssm/parameters.py file has the following,
Wneu3 = Parameter(name = 'Wneu3' (providing a sample value, like the other particles),
                  nature = 'external',
                  type = 'real',
                  value = 16.449054,
                  texname = '\\text{Wneu3}',
                  lhablock = 'DECAY',
                  lhacode = [ 1000025 ])
I am still confused. Does it imply that only the Br and decay width provided for neutralino3 is inconsistent and not for neutralino2 and chargino1? All of these have been computed using NMSSMTools

If so, would 1000025 auto (i.e. asking MG to compute the decay) help, kindly advise.

regards,
arindam

Hi,

Looks like this is another type of consistency check that the one, I was describing before.
In this case, the consistency check is the fact that the sign of the mass and the sign of the width of a given particle are identical (Which is important to have at the matrix-element level).

Now, this might not be the good behaviour to write it as negative in the param_card.
I will ask some SUSY expert if they know what the correct behaviour is.
Otherwise, I will re-read the various convention agreement.

Cheers,

Olivier

> On 14 Oct 2018, at 05:52, arindam chatterjee <email address hidden> wrote:
>
> Question #675126 on MadGraph5_aMC@NLO changed:
> https://answers.launchpad.net/mg5amcnlo/+question/675126
>
> arindam chatterjee gave more information on the question:
> Thanks. However, that would be true for chargino1 or neutralino2, the other particles I have considered for production as well.
> For these particles there is no issue, and the decay width and br information in the param_card is read. Only for neutralino3 (which also happens to have a -ve eigenvalue) this error shows up. In the models folder, the models/nmssm/parameters.py file has the following,
> Wneu3 = Parameter(name = 'Wneu3' (providing a sample value, like the other particles),
> nature = 'external',
> type = 'real',
> value = 16.449054,
> texname = '\\text{Wneu3}',
> lhablock = 'DECAY',
> lhacode = [ 1000025 ])
> I am still confused. Does it imply that only the Br and decay width provided for neutralino3 is inconsistent and not for neutralino2 and chargino1? All of these have been computed using NMSSMTools
>
> If so, would 1000025 auto (i.e. asking MG to compute the decay) help,
> kindly advise.
>
> regards,
> arindam
>
> --
> You received this question notification because you are an answer
> contact for MadGraph5_aMC@NLO.

Hi,

Thanks Benj,

Arindam, you will find a patch in order to keep the width positive here:
https://bazaar.launchpad.net/~mg5core1/mg5amcnlo/2.6.4/diff/303?context=3

Cheers,

Olivier

On 15 Oct 2018, at 10:34, Benjamin Fuks <<email address hidden><mailto:<email address hidden>>> wrote:

Hi Olivier, Arindam,

In the SLHA conventions, the width has always to be made positive, even if the mass negative. This is (IMO) the reason why Pythia crashes.

Cheers,

Benjamin

On 15 Oct 2018, at 05:48 , Olivier Mattelaer <<email address hidden><mailto:<email address hidden>>> wrote:

Hi,

Looks like this is another type of consistency check that the one, I was describing before.
In this case, the consistency check is the fact that the sign of the mass and the sign of the width of a given particle are identical (Which is important to have at the matrix-element level).

Now, this might not be the good behaviour to write it as negative in the param_card.
I will ask some SUSY expert if they know what the correct behaviour is.
Otherwise, I will re-read the various convention agreement.

Cheers,

Olivier

On 14 Oct 2018, at 05:52, arindam chatterjee <<email address hidden><mailto:<email address hidden>>> wrote:

Question #675126 on MadGraph5_aMC@NLO changed:
https://answers.launchpad.net/mg5amcnlo/+question/675126

arindam chatterjee gave more information on the question:
Thanks. However, that would be true for chargino1 or neutralino2, the other particles I have considered for production as well.
For these particles there is no issue, and the decay width and br information in the param_card is read. Only for neutralino3 (which also happens to have a -ve eigenvalue) this error shows up. In the models folder, the models/nmssm/parameters.py file has the following,
Wneu3 = Parameter(name = 'Wneu3' (providing a sample value, like the other particles),
               nature = 'external',
               type = 'real',
               value = 16.449054,
               texname = '\\text{Wneu3}',
               lhablock = 'DECAY',
               lhacode = [ 1000025 ])
I am still confused. Does it imply that only the Br and decay width provided for neutralino3 is inconsistent and not for neutralino2 and chargino1? All of these have been computed using NMSSMTools

If so, would 1000025 auto (i.e. asking MG to compute the decay) help,
kindly advise.

regards,
arindam

--
You received this question notification because you are an answer
contact for MadGraph5_aMC@NLO.

Thank you. I am looking into it.

Also, I realized that I opted for an version update in my MG5_v2.4.3 folder, which led to
v 2.6.2.

regards,
arindam

Can you help with this problem?

Provide an answer of your own, or ask arindam chatterjee for more information if necessary.

To post a message you must log in.