MadGraph5_aMC@NLO

Madspin don't find the total width and return partial width (6.67) larger than total width (0.0) --from param_card--

Asked by Thomas Guérandel

Dear Madgraph experts,

I don't know how to use the Madspin extension, it can't calculate the branching ratio because it only calculate the partial width.
In the SM model, I do :
generate p p > t t~
launch
madspin=onshell
and edit the madspin_card to :
decay t > w+ b
decay t~ > w- b~
It gives me the same cross section 505.8 ± 0.8 before and after the decay as expected.
But in my BSM model, I do the same commands for a other processus and at the moment to calculate the branching ratio it returns me the error : WARNING: partial width (6.67) larger than total width (0.0) --from param_card--
I don't understand this because I don't find any parameter "width" in param_card (in MG5 or madspin) and it don't make an error for the SM processus.

Relative to this problem, I do on madspin :
import ../.../unweighted_events.lhe.gz
decay particule > all
launch
it gives me the total width with all the BR but return me the error : AttributeError : 'NoneType' object has no attribute 'is_fermion'

I appreciate any feedback and many thanks in advance,
Thomas

Question information

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

Hi,

You should have a block DECAY within your param_card that defines the total width of your particle.
This is what is currently set to zero and the information that madspin use to compute the cross-section after decay.
(this is production cross-section times partial width divided by total width).
So the fact that madspin computes a partial width larger than a total width means your cross-section will increase (not physical) and since your total width is zero the result will be nan.

For your second issue the syntax
decay particule > all
does not make sense you need to use
decay particule > all all
or
decay particule > all all all

all is a multiparticle label asking madgraph to test the decay with that label replace by one other particle (all means that we will test all), so the only valid diagram with your syntax is decay particule > particle which is obviously not a decay.

Cheers,

Olivier

Revision history for this message
Thomas Guérandel (thomasguerandel) said : 		#2

Hi,

Thanks, I didn't see this DECAY block..
And for the second issue I was wrong, I meant particule > all all.
But it returns the same error : AttributeError : 'NoneType' object has no attribute 'is_fermion'

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

For your second issue, this is likely related to a problem of your UFO model.

Do you have the debug file? With that file, I might be able to help (but not sure).
If they are no usefull information in that file, I would need you send the UFO to me by email? (<email address hidden> (indicating the number 705828 such that I can link to this question), and also a description of all the step that you do, such that I can reproduce your error.

Cheers,

Olivier

Revision history for this message
Thomas Guérandel (thomasguerandel) said : 		#4

I think that I found the problem : one of my particle have no anti-particle defined. So it returns a 'NoneType' object when Madspin searches the decays. Thanks for your help Olivier !

Cheers,
Thomas