MadGraph5_aMC@NLO

Problems with behaviour of leptons at high eta

Asked by Alessio Mastroddi

Good morning,
I'm simulating events for a lepton collider (e+ e-) in a personalized model (a simple axion-like particle model, coupling only to the photons). I am trying to compute the "inclusive" cross sections of the process 'e+ e- > e+ e- alp' , so with the least amount of cuts possible.
 However, for small values of m_alp (below 0.1 GeV) there is a problem when drawing the histograms of the pseudorapidity of the final leptons. In particular when the cut on eta is high (etal = 15), there are still a number of leptons produced with an eta greater than 15, which was the limit set in the run card. Why does this happen?

Also, there is another feature that is happening: in the pseudorapidity histogram for the particle corresponding to the "forward" direction in the initial scattering (electron if I generate e- e+ > ..., positron if I generate e+ e- > ....) there is a ''jump'' at around eta=9. In practice the histogram follows a smooth curve, and then around eta=9 it suddenly drops and then restarts its course. This is happening only for one of the final leptons, and since they are basically the same I can't understand why this is happening.

I will write below all the relevant parameters of the run card that I'm using for the simulations.
Thanks in advance,
Alessio

''''''''''''''''''
Process: e- e+ > e- e+ alp
m_alp=0.01 GeV
m_e=5.11 e-4 GeV (from SM model withoud massless restrictions)
symmetric collider: ebeam1=ebeam2=5.29 GeV
set no_parton_cut
set ptl 1e-12
set etal 15

Question information

Language:
English Edit question
Status:
Expired
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 would need the model in order to comment.

Cheers,

Olivier

Revision history for this message
Alessio Mastroddi (alessio-m) said : 		#2

Hi,
I will attach here the folder containing the model that I import in Madgraph. Also I attach one of the histograms of the pseudo rapidity so that it’s clearer, I’m afraid it was kind of difficult to explain without any visual aid.
Thanks,
Alessio

P.S. If You want I can send you also the Mathematica notebook of the UFO model. Let me know if it would be more useful.

> On 5 Jun 2020, at 11:45, Olivier Mattelaer <email address hidden> wrote:
>
> Your question #691146 on MadGraph5_aMC@NLO changed:
> https://answers.launchpad.net/mg5amcnlo/+question/691146
>
> Status: Open => Needs information
>
> Olivier Mattelaer requested more information:
> Hi,
>
> I would need the model in order to comment.
>
> Cheers,
>
> Olivier
>
> --
> To answer this request for more information, you can either reply to
> this email or enter your reply at the following page:
> https://answers.launchpad.net/mg5amcnlo/+question/691146
>
> You received this question notification because you asked the question.

Revision history for this message
Alessio Mastroddi (alessio-m) said : 		#3

P.P.S. My attachments were in the mail I sent in response to your comment, but I can't see them here on the website. Do you need me to send them in another way? Thanks again and sorry for the inconvenience

Revision history for this message
Olivier Mattelaer (olivier-mattelaer) said : 		#4

Yes i did not receive them,
You can send them to <email address hidden> <mailto:<email address hidden>>
(please include the question number: 691146)

Cheers,

Olivier

> On 5 Jun 2020, at 12:01, Alessio Mastroddi <email address hidden> wrote:
>
> Question #691146 on MadGraph5_aMC@NLO changed:
> https://answers.launchpad.net/mg5amcnlo/+question/691146
>
> Alessio Mastroddi posted a new comment:
> P.P.S. My attachments were in the mail I sent in response to your
> comment, but I can't see them here on the website. Do you need me to
> send them in another way? Thanks again and sorry for the inconvenience
>
> --
> You received this question notification because you are an answer
> contact for MadGraph5_aMC@NLO.

Revision history for this message
Olivier Mattelaer (olivier-mattelaer) said : 		#5

How many events did you get? Did the code succesfully generated the number of events that you requested?

When I run with 2.7.2 the phase-space integrator indeed struggle in that case and therefore I would not trust the result.
However on a development version of the code, it seems to behave correctly but that branch is not ready for release.

Cheers,

Olivier

Revision history for this message
Alessio Mastroddi (alessio-m) said : 		#6

I am working on the version 2.6.7 actually, and the code actually generated the correct number of events (10k), even if the integration took some time; does this mean that if the program runs for more than some time, say n minutes, I should not totally trust the results?

Thanks,
Alessio

> On 5 Jun 2020, at 16:11, Olivier Mattelaer <email address hidden> wrote:
>
> Your question #691146 on MadGraph5_aMC@NLO changed:
> https://answers.launchpad.net/mg5amcnlo/+question/691146
>
> Olivier Mattelaer proposed the following answer:
> How many events did you get? Did the code succesfully generated the
> number of events that you requested?
>
> When I run with 2.7.2 the phase-space integrator indeed struggle in that case and therefore I would not trust the result.
> However on a development version of the code, it seems to behave correctly but that branch is not ready for release.
>
> Cheers,
>
> Olivier
>
> --
> If this answers your question, please go to the following page to let us
> know that it is solved:
> https://answers.launchpad.net/mg5amcnlo/+question/691146/+confirm?answer_id=4
>
> If you still need help, you can reply to this email or go to the
> following page to enter your feedback:
> https://answers.launchpad.net/mg5amcnlo/+question/691146
>
> You received this question notification because you asked the question.

Revision history for this message
Olivier Mattelaer (olivier-mattelaer) said : 		#7

Ok no sorry, I aslo have correctly the 10k events in 2.7.2
I just did not run the correct script (too high energy)

Revision history for this message
Olivier Mattelaer (olivier-mattelaer) said : 		#8

So the issue seems a numerical accuracy issue due to the small mass of the leptons.

Funny enough I change that part of the code to avoid such type of inacuraccy (in a fail attempt to see if those inacurracy was leading to some slow down of the code in VBF like processes). I kind of break causality here...

You can find the associated patches (and a link to a version of the code) at this page.
https://bazaar.launchpad.net/~maddevelopers/mg5amcnlo/2.7.2_tchannel/revision/298

Thanks,

Olivier

Revision history for this message
Alessio Mastroddi (alessio-m) said : 		#9

Dear Olivier,
I downloaded the 2.7.2 version and the patched code (substituting the old files in the folder). However, while the jump in eta seems to have disappeared, there is still a problem with the cut: I set in the run card etal<15, but when drawing the histograms there are still a number of events that go over said cut. Is this fixable, or we should avoid these configurations with very low masses?
Thanks again,
Alessio

> On 5 Jun 2020, at 21:50, Olivier Mattelaer <email address hidden> wrote:
>
> Your question #691146 on MadGraph5_aMC@NLO changed:
> https://answers.launchpad.net/mg5amcnlo/+question/691146
>
> Olivier Mattelaer proposed the following answer:
> So the issue seems a numerical accuracy issue due to the small mass of
> the leptons.
>
> Funny enough I change that part of the code to avoid such type of
> inacuraccy (in a fail attempt to see if those inacurracy was leading to
> some slow down of the code in VBF like processes). I kind of break
> causality here...
>
> You can find the associated patches (and a link to a version of the code) at this page.
> https://bazaar.launchpad.net/~maddevelopers/mg5amcnlo/2.7.2_tchannel/revision/298
>
> Thanks,
>
> Olivier
>
> --
> If this answers your question, please go to the following page to let us
> know that it is solved:
> https://answers.launchpad.net/mg5amcnlo/+question/691146/+confirm?answer_id=7
>
> If you still need help, you can reply to this email or go to the
> following page to enter your feedback:
> https://answers.launchpad.net/mg5amcnlo/+question/691146
>
> You received this question notification because you asked the question.

Revision history for this message
Olivier Mattelaer (olivier-mattelaer) said : 		#10

Is this not due to difference between rapidity and pseudo-rapidity?

Olivier

> On 6 Jun 2020, at 10:01, Alessio Mastroddi <email address hidden> wrote:
>
> Question #691146 on MadGraph5_aMC@NLO changed:
> https://answers.launchpad.net/mg5amcnlo/+question/691146
>
> Status: Answered => Open
>
> Alessio Mastroddi is still having a problem:
> Dear Olivier,
> I downloaded the 2.7.2 version and the patched code (substituting the old files in the folder). However, while the jump in eta seems to have disappeared, there is still a problem with the cut: I set in the run card etal<15, but when drawing the histograms there are still a number of events that go over said cut. Is this fixable, or we should avoid these configurations with very low masses?
> Thanks again,
> Alessio
>
>> On 5 Jun 2020, at 21:50, Olivier Mattelaer <email address hidden> wrote:
>>
>> Your question #691146 on MadGraph5_aMC@NLO changed:
>> https://answers.launchpad.net/mg5amcnlo/+question/691146
>>
>> Olivier Mattelaer proposed the following answer:
>> So the issue seems a numerical accuracy issue due to the small mass of
>> the leptons.
>>
>> Funny enough I change that part of the code to avoid such type of
>> inacuraccy (in a fail attempt to see if those inacurracy was leading to
>> some slow down of the code in VBF like processes). I kind of break
>> causality here...
>>
>> You can find the associated patches (and a link to a version of the code) at this page.
>> https://bazaar.launchpad.net/~maddevelopers/mg5amcnlo/2.7.2_tchannel/revision/298
>>
>> Thanks,
>>
>> Olivier
>>
>> --
>> If this answers your question, please go to the following page to let us
>> know that it is solved:
>> https://answers.launchpad.net/mg5amcnlo/+question/691146/+confirm?answer_id=7
>>
>> If you still need help, you can reply to this email or go to the
>> following page to enter your feedback:
>> https://answers.launchpad.net/mg5amcnlo/+question/691146
>>
>> You received this question notification because you asked the question.
>
> --
> You received this question notification because you are an answer
> contact for MadGraph5_aMC@NLO.

Revision history for this message
Alessio Mastroddi (alessio-m) said : 		#11

You are correct, I’m sorry. I did not consider the fact that the cut is in the absolute rapidity, not the pseudo rapidity; now everything seems to be working.

Just for concluding: in order to avoid the weird jump in the pseudo rapidity distribution I should use the 2.7.2 patched version, until the new stable release comes, am I correct?

Thanks for everything,
Alessio

> On 6 Jun 2020, at 10:20, Olivier Mattelaer <email address hidden> wrote:
>
> Your question #691146 on MadGraph5_aMC@NLO changed:
> https://answers.launchpad.net/mg5amcnlo/+question/691146
>
> Status: Open => Answered
>
> Olivier Mattelaer proposed the following answer:
> Is this not due to difference between rapidity and pseudo-rapidity?
>
> Olivier
>
>> On 6 Jun 2020, at 10:01, Alessio Mastroddi <email address hidden> wrote:
>>
>> Question #691146 on MadGraph5_aMC@NLO changed:
>> https://answers.launchpad.net/mg5amcnlo/+question/691146
>>
>> Status: Answered => Open
>>
>> Alessio Mastroddi is still having a problem:
>> Dear Olivier,
>> I downloaded the 2.7.2 version and the patched code (substituting the old files in the folder). However, while the jump in eta seems to have disappeared, there is still a problem with the cut: I set in the run card etal<15, but when drawing the histograms there are still a number of events that go over said cut. Is this fixable, or we should avoid these configurations with very low masses?
>> Thanks again,
>> Alessio
>>
>>> On 5 Jun 2020, at 21:50, Olivier Mattelaer <email address hidden> wrote:
>>>
>>> Your question #691146 on MadGraph5_aMC@NLO changed:
>>> https://answers.launchpad.net/mg5amcnlo/+question/691146
>>>
>>> Olivier Mattelaer proposed the following answer:
>>> So the issue seems a numerical accuracy issue due to the small mass of
>>> the leptons.
>>>
>>> Funny enough I change that part of the code to avoid such type of
>>> inacuraccy (in a fail attempt to see if those inacurracy was leading to
>>> some slow down of the code in VBF like processes). I kind of break
>>> causality here...
>>>
>>> You can find the associated patches (and a link to a version of the code) at this page.
>>> https://bazaar.launchpad.net/~maddevelopers/mg5amcnlo/2.7.2_tchannel/revision/298
>>>
>>> Thanks,
>>>
>>> Olivier
>>>
>>> --
>>> If this answers your question, please go to the following page to let us
>>> know that it is solved:
>>> https://answers.launchpad.net/mg5amcnlo/+question/691146/+confirm?answer_id=7
>>>
>>> If you still need help, you can reply to this email or go to the
>>> following page to enter your feedback:
>>> https://answers.launchpad.net/mg5amcnlo/+question/691146
>>>
>>> You received this question notification because you asked the question.
>>
>> --
>> You received this question notification because you are an answer
>> contact for MadGraph5_aMC@NLO.
>
> --
> If this answers your question, please go to the following page to let us
> know that it is solved:
> https://answers.launchpad.net/mg5amcnlo/+question/691146/+confirm?answer_id=9
>
> If you still need help, you can reply to this email or go to the
> following page to enter your feedback:
> https://answers.launchpad.net/mg5amcnlo/+question/691146
>
> You received this question notification because you asked the question.

Revision history for this message
Roberto Franceschini (franceschini-roberto) said : 		#12

So the run_card.dat variable is called `etal` but is in reality `y` ? So, why the descriptive text is "max rap for ZYX " ?

Revision history for this message
Roberto Franceschini (franceschini-roberto) said : 		#13

Ehehe, I was confused, the real question is " So, why the variable name is _eta_ if it is coded as a rapidity y? "
In any case, a probably more substantial issue I see is that a detector has a maximum angle of coverage, which is literally eta, why one should make a cut on y?

PS
I hope I am not mixing up things again!

Revision history for this message
Launchpad Janitor (janitor) said : 		#14

This question was expired because it remained in the 'Needs information' state without activity for the last 15 days.