MadGraph5_aMC@NLO

Issues with MSSM decays

Asked by Prashant Saraswat on 2011-09-19

Hi all,

I've been encountering some strange results when I calculate some processes using the built-in MSSM model in MadGraph, specifically those involving decays of neutralinos. I've tried using param_cards from different sources such as SPheno, SuSpect+SDECAY, etc. which all result in similar answers.

Bizarre results for cross-section times branching ratio of n1:

When I generate the process p p > n1 n1, n1 > ta+ ta1- and generate events, I find that it takes an extremely long time compared to similar processes, and ends up generating only 3-10 events and giving a cross-section several orders of magnitude too large-- much larger than the result for p p > n1 n1, which generates 1000 events in a minute and gives a sensible cross-section. The process p p > n2 n2, n2 > ta+ ta1- however is fast and gives a sensible answer (smaller than the p p > n2 n2 cross-section). I get a similar problem replacing the stau with a selectron/smuon if those decays are allowed.

Inconsistent results for decay widths to staus:

When I calculate the decay widths for the processes n1 > ta+ ta1- and n2 > ta+ ta1-, I find them to be significantly off (factor of a few) from the results in the param_card that I produced using various SUSY calculators (SPheno, SDECAY, etc.). The decay widths to e's or mu's are very similar between MadGraph and the calculators though.

Does anyone have some idea what could be happening here to give these results? I can post the banners if they would help.

Question information

Language:
English Edit question
Status:
Solved
For:
MadGraph5_aMC@NLO Edit question
Assignee:
No assignee Edit question
Solved by:
Johan Alwall
Solved:
2011-09-20
Last query:
2011-09-20
Last reply:
2011-09-20
Best Johan Alwall (johan-alwall) said : #1

Hi Prashant,

The problem is that you are using an SLHA1 parameter card. The MG5 MSSM model uses SLHA2-type cards, not SLHA1. So most of the parameters in your card are ignored and default parameters are used instead. In a future version, we will include an SLHA1<->2 converter (hopefully in a few weeks), but right now I would recommend to use the v4 model:
import model_v4 mssm
which takes SLHA1 cards by default.

All the best,
Johan

Prashant Saraswat (prashant88) said : #2

Hi Johann,

Thanks, using model_v4 mssm worked, I now get sensible results for everything. But I don't think it can be a SLHA1/2 issue, as I'm pretty sure the parameter cards I was using were in SLHA2 form-- one of them explicitly stated this in the card, and all of them were from the latest versions of spectrum generators (including the online calculator on the MadGraph homepage). So I'm still not sure why the version 5 model wasn't working.

Using model_v4 is somewhat annoying right now because I run into the problem discussed in https://answers.launchpad.net/madgraph5/+question/169739 where I have to run "change_compiler.py gfortran" before generating events. Right now it seems that I have to do this every time I generate a new process, is there some way to make MadEvent remember this as a default?

Anyway, thanks for helping me get this working!

-- Prashant

Prashant Saraswat (prashant88) said : #3

Thanks Johan Alwall, that solved my question.

Arian Abrahantes (arian-abrahantes) said : #4

Hi Prashant, I believe that running the python code right in the Template folder will solve the issue once and for all I didnt test this since I ran in different compilers for cross check but once I think I tried it.

RAFIQUL RAHAMAN (rahaman-rafiqul786) said : #5

Hi Prashant, I am unable to use the SPheno mass spectrum in madgraph, can you tell me how to use? Thankx in advance.
Rafiqul