Cut Decays on HepMC

Asked by Joel Jones

Dear MadGraph, I have a question regarding cut_decays. I am trying to generate events that will automatically pass a 27 GeV lepton trigger. For this, I run:

generate p p > Z H, (Z > l+ l-)

On the run card, I set:

27.0 = xptl ! minimum pt for at least one charged lepton

Of course, I set cut_decays to True. If I inspect the generated lhe file, everything is ok. However, after running Pythia, the leptons on the HepMC output do not necessarily pass the cuts. For instance, one of the events on the lhe has an electron that passes the cut:

11 1 3 3 0 0 -2.8638742914e+01 -1.6432462349e+00 +1.2197184318e+02 1.2529967431e+02 0.0000000000e+00 0.0000e+00 -1.0000e+00

However, on the HepMC this electron will end up with lower pT (it appears in two "vertices", one of which has only one particle):

P 901 11 -2.5005472801606885e+01 -3.0725136642052724e+00 1.0882166276154923e+02 1.1169990284999402e+02 5.1099999999999995e-04 23 0 0 -686 0
P 904 11 -2.5005470553154794e+01 -3.0725133879297672e+00 1.0882165297647948e+02 1.1169989280611753e+02 5.1099999999999995e-04 1 0 0 0 0

As you can see, the pT is reduced from 28.6 GeV to 25.2 GeV. This makes me lose about 2% of the events, which I guess I can live with. A similar thing happens if I run t t~ H production. Still, first request, could you shed light on why this happens?

The second question is much more worrying. I now run:

generate p p > W+ H, (W+ > l+ nu)
add process p p > W- H, (W- > l- nu)

Here, MadGraph does not give me the option for setting xptl, so I set:

 27.0 = ptl ! minimum pt for the charged leptons

In this case, I end up losing about 25% of the generated events! Here I have again traced an example. On the lhe, everything is good, look at this electron:

11 1 3 3 0 0 -2.3672448034e+01 -4.7028479844e+01 -8.5434356876e+00 5.3339038290e+01 0.0000000000e+00 0.0000e+00 -1.0000e+00

On the HepMC, one finds that a lot of energy is lost by photon emission. Let me copy the two relevant vertices:

V -306 0 -4.4112901754140028e-14 -1.7186855386499357e-14 1.5056863069913799e-14 9.8482619688719761e-14 0 2 0
P 399 11 -2.3145609862544049e+01 -4.5384393666338731e+01 -8.1201750435059932e+00 5.1588755435352397e+01 5.1099999999999995e-04 23 0 0 -307 0
P 400 12 -1.8361489881036054e+01 2.9212786439463372e+01 2.2287611153811561e+01 4.1076377799808377e+01 0 23 0 0 -308 0

V -307 0 -4.4112901754140028e-14 -1.7186855386499357e-14 1.5056863069913799e-14 9.8482619688719761e-14 0 2 0
P 401 11 -1.1751173031602711e+01 -1.3360162148954583e+01 -7.5024521132602118e+00 1.9309862460357117e+01 5.1099999999999995e-04 51 0 0 -309 0
P 402 22 -1.1900316011696987e+01 -3.1219387217311933e+01 -3.6748031852731344e-03 3.3410592226222057e+01 0 51 0 0 -310 0

Here, one sees that the photon takes 33.4 GeV as pT, leaving the original electron with 17.9 GeV (so it does not pass the cut). I understand that these radiative processes are to be expected, but what I do not understand is why they do not happen (or do not happen so frequently) for the other processes I generate?

Many thanks!

Question information

Language:
English Edit question
Status:
Answered
For:
MadGraph5_aMC@NLO Edit question
Assignee:
No assignee Edit question
Last query:
Last reply:
Revision history for this message
Olivier Mattelaer (olivier-mattelaer) said :
#1

Hi,

I do not have a good answer on this point. Pythia's authors might have look at that in details and might be able to give you the reason.

Cheers,

Olivier

Revision history for this message
Joel Jones (n-jones-j) said :
#2

Thanks for the quick reply! Should I contact them directly?

Revision history for this message
Olivier Mattelaer (olivier-mattelaer) said :
#3

I can not transfer the question to their tracking website: https://gitlab.com/Pythia8/releases/-/issues/
and this is unlikely that they will read this thread randomly

Cheers,

Olivier

Revision history for this message
Joel Jones (n-jones-j) said :
#4

Great, many thanks for the link. I'll report back here once I get an answer, in case anyone else has the same problem.

Can you help with this problem?

Provide an answer of your own, or ask Joel Jones for more information if necessary.

To post a message you must log in.