MadSpin interrupted with error during susy particle decay

Asked by Ewan McCulloch on 2017-08-24

In MG5 I import the model MSSM_SLHA2 and then generate the process p p > n2 n1 and want MadSpin to handle the decay of the n2. I only want to considet the decay n2 > n1 z, z > l+ l- so I add this to the madspin card. I leave the other cards as default except the run card where I reduce the number of events from 10000 to 100.

I then run ./bin/generate_events and ask Madspin to handle the decays.

MadSpin gets as far as this:

INFO: generating the production square matrix element
INFO: geberate p p > n2 n1 --no_warning=duplicate;
INFO: Done 12.6
INFO: generating the full matrix element squared (with decay)
INFO: generate p p > n2 n1, (n2 > n1 z, z > l+ l- QCD=99) --no_warning=duplicate;
Command "generate_events run_03" interrupted with error:
InvalidCmd : No matrix elements generated, check overall coupling orders
quit

Experts, any idea what causes this to fail?

Cheers,
Ewan

Question information

Language:
English Edit question
Status:
Answered
For:
MadGraph5_aMC@NLO Edit question
Assignee:
No assignee Edit question
Last query:
2017-08-24
Last reply:
2017-08-24

Hi Ewan,

My guess is that the branching ratio n2 > n1 z is computed at zero since
If you use the default card you have Mn2 < Mn1+MZ :
  1000022 9.668807e+01 # Mneu1
  1000023 1.810882e+02 # Mneu2
   23 9.118760e+01 # MZ

Cheers,

Olivier

> On 24 Aug 2017, at 15:27, Ewan McCulloch <email address hidden> wrote:
>
> Question #656856 on MadGraph5_aMC@NLO changed:
> https://answers.launchpad.net/mg5amcnlo/+question/656856
>
> Description changed to:
> In MG5 I import the model MSSM_SLHA2 and then generate the process p p >
> n2 n1 and want MadSpin to handle the decay of the n2. I only want to
> considet the decay n2 > n1 z, z > l+ l- so I add this to the madspin
> card. I leave the other cards as default except the run card where I
> reduce the number of events from 10000 to 100.
>
> I then run ./bin/generate_events and ask Madspin to handle the decays.
>
> MadSpin gets as far as this:
>
> INFO: generating the production square matrix element
> INFO: geberate p p > n2 n1 --no_warning=duplicate;
> INFO: Done 12.6
> INFO: generating the full matrix element squared (with decay)
> INFO: generate p p > n2 n1, (n2 > n1 z, z > l+ l- QCD=99) --no_warning=duplicate;
> Command "generate_events run_03" interrupted with error:
> InvalidCmd : No matrix elements generated, check overall coupling orders
> quit
>
> Experts, any idea what causes this to fail?
>
> Cheers,
> Ewan
>
> --
> You received this question notification because you are an answer
> contact for MadGraph5_aMC@NLO.

Ewan McCulloch (ewanmcc) said : #2

Dear Olivier,

If I use the syntax "decay n2 > n1 l+ l-" will I be allowing Z to be off-shell? When I try this I get the same error message and the process fails:

INFO: generating the production square matrix element
INFO: generate p p > n2 n1 --no_warning=duplicate;
INFO: Done 16.82
INFO: generating the full matrix element squared (with decay)
INFO: generate p p > n2 n1, n2 > n1 e+ e- QCD=99 --no_warning=duplicate;
Command "generate_events " interrupted with error:
InvalidCmd : No matrix elements generated, check overall coupling orders

Should I put bwcutoff parameter to use instead to allow the Z mass to be far from the BW peak centre? (I hope I've understood that parameter properly).

Best,
Ewan

Ewan McCulloch (ewanmcc) said : #3

Looking at the param card I see that a "x1+ > n1 w+, w+ > e+ e-" decay should be allowed with the W being on shell. I tried this and it succeeded. So you are right about why my previous decay failed. Can you advise how I would handle the above n2 > n1 e+ e- decay via a virtual Z in MadSpin?

Best,
Ewan

Ewan McCulloch (ewanmcc) said : #4

Trying the original "n2 > n1 z, z > e+ e-" decay with bwcutoff at 100 GeV the process still fails with the same error. So it seems like I misinterpreted the use of this parameter.

Ewan McCulloch (ewanmcc) said : #5

Correction: I changed BW_cut to 100, is this a scaling factor or in fact where the cut is, i.e at 100GeV?

Dear Ewan,

The reason of " No matrix elements generated, check overall coupling orders"
is directly related to the fact that branching ratio of the "n2 > n1 z, z > l+ l-"
is computed as BR_n2>n1z * BR_Z>ll
The problem is that BR_n2>n1z is estimated to zero and therefore such contribution is dropped independently of the BW_cut parameter.

One thing that you can try is to specify in advance a fake branching ratio such that MadSpin does not have to compute it.
If MadSpin does not re-check the spectrum this should be enough to reach the second phase of the computation.
Now since NWA does not really apply in this case, the MadSpin method is not ensure to work and you might face some trouble.

(Note that MadSpin can not handle three body decay with full off-shell effect and narrow width approximation)

My advise, would be to not use MadSpin but instead use the decay chain syntax at the madgraph level:
generate p p > n2 n1, n2 > n1 e+ e- QCD=99
such direct computation does not use the Branching ratio and should be more accurate and stable.

Cheers,

Olivier

Can you help with this problem?

Provide an answer of your own, or ask Ewan McCulloch for more information if necessary.

To post a message you must log in.