CalcHEP

Interference between two heavy mass-eigenstates in lljj channel

Asked by Matt

Dear CalcHEP team,

I have a query about using CalcHEP in processes at LHC with a four particles final state (eejj) which receives contributions from two heavy mass eigenstates say n1 and n2 that are known in terms of the model parameters.

I am interested in computing the cross section for the process pp->e+e+jj in some regions of the parameters space where the interference between the two mass eigenstates is crucial.

I first tried (working in the batch mode) using the syntax
...
Process: p,p->e+,N
Decay: N->e+,j,j
Alias: N=n1, n2
...
but the results don't seem to consider interference at all.

Unfortunately the full simulation (pp->e+e+jj) becomes too lengthy and therefore unpractical for a phenomenological study.

Is there any other way to take into account of interferences in this process?
Or otherwise, are there any tricks and/or particular settings to use in order to speed up the CalcHEP simulation of the full process?

Thank you in advance.

Question information

Language:
English Edit question
Status:
Solved
For:
CalcHEP Edit question
Assignee:
No assignee Edit question
Solved by:
Alexander Belyaev
Solved:
Last query:
Last reply:
Revision history for this message
Alexander Belyaev (alexander.belyaev) said : 		#1

Hello Mateo,

 pp->e+e+jj

process should be quite easy with CalcHEP at modern laptops/desktops.
Could you send me (<email address hidden>) the model and the batch file, so we can test it?

You can also try HEPMDB -- https://hepmdb.soton.ac.uk/
where you can run CalcHEP in the batch mode at the powerful HPC cluster

2->4 process should not be a problem at all for the CalcHEP

Revision history for this message
Best Alexander Belyaev (alexander.belyaev) said : 		#2

Hello Matteo,

did you try to use HEPMDB?
I have run you batch file at HEPMDB myself and will send you results in the separate e-mail.
Here are the summary/remarks::

1. To compile and generate 10K events it took about 4 hours at HEPMDB,
so next time it could be 3 hours, since ~ 1 hour it took for the compilation
which will not be eventually repeated

2. To generate bigger number of events, say 100k it will take about the same time i.e. 3 hours

3. To get sensible results I have modified your original batch file to include "regularisation"
since you have two narrow resonances, I have also increased the number of points for the integration.
Details about this will be in my separate e-mail.
The problem is that without regularisation , the accuracy of the integration
was bad and event generation was taking forever.
As soon as the accuracy of the integration became around 1% or better after my modification,
event generation became very fast.

4. The original model you have sent me
had an extlib.mdl file with the line

#/usr/local/lib/libLHAPDF.so

which is not correct, since in calchep the comment should start with "%"
, so I have commented this out.
The line
-L$HOME/packages/lhapdf/lib/ -lLHAPDF
is sufficient to link LHAPDF

So, after using "regularisation" CalcHEP produces results quickly.
We are planning to implement automatic regularisation soon
to avoid its introduction manually

Revision history for this message
Matt (elvin.j) said : 		#3

Thanks Alexander Belyaev, that solved my question.