Decay Width difference between MG5-SPheno

[jackaraz]

Hi Florian

I realised a strange inconsistency between MG5 and SPheno. I tried to calculate chargino production crossection with a decay chain and what I realised was without a decay chain I got 0.001624 pb but with specific decay chain (without using madspin) I got something around 10^11 pb. Thus I checked the decay widths of some couple of particles and the result was as follows

(they are generated by generate p1 > all all)
cha1 = MG5 5.235 GeV SLHA = 1.98734909E-03 GeV, neutralino1 MG5 = 0.488 GeV SLHA = 8.87194754E-09 GeV, higgs MG5 = 0.001615 GeV SLHA = 4.13290311E-03 GeV also MG5 uses decay channels that are not listed in the param card

which explains the reason that why my crossection becomes quite big since it multiplied by width(MG5)/width(param_card) and due to extremly low width from spheno it becomes huge. However I couldn't understand the reason of such difference. Spheno seems calculating them extremely low with respect to MG5. Do you know why this might happening or how can I avoid one or other?

Thank you very much

Best regards
Jack

[FStaub]

Hi,

could you please attach the SPheno.spc file all parameters and decays as well as the results for MG for the decays? Maybe, I can see something obvious that goes wrong. If not, I would need to run things myself but because of workshops and travels that won't be possible in the next 1.5 weeks.

Cheers,
Florian

PS: Did you set

Code: Select all

 14 0               # Run couplings to scale of decaying particle 
 

for the SPheno calculation?

[jackaraz]

Dear Florian

Yes, Flag 14 was set to zero. just to check the difference I changed it to 1 and ran it again but there was no significant difference to my eyes. I'm attaching the lheFile which also includes slha file with decay tables. Thank you very much for your time.

Best regards
Jack

[FStaub]

Sorry, I found a problem with the model file: SNUMIX is used twice. MG uses therefore the wrong matrices.

You need to change parameters.m as

Code: Select all

{ZVL,	    {Description -> "Neutrino-Mixing-Matrix",
             LaTeX->"Z^{V_L}", 
             LesHouches ->  SNULMIX,
             OutputName-> ZVL             
             }},

I hope that solves the problem. If not, I have to take next week a closer look.

Cheers,
Florian

[FStaub]

Hi again,

looking at the spectrum, the width of MG makes actually more sense. And then a saw

Code: Select all

Block MSL2 Q=  1.39738515E+03  # (SUSY Scale)
  1  1     0.00000000E+00   # Real(ml2(1,1),dp)
  1  2     0.00000000E+00   # Real(ml2(1,2),dp)
  1  3     0.00000000E+00   # Real(ml2(1,3),dp)
  2  1     0.00000000E+00   # Real(ml2(2,1),dp)
  2  2     2.50000000E+03   # Real(ml2(2,2),dp)
  2  3     0.00000000E+00   # Real(ml2(2,3),dp)
  3  1     0.00000000E+00   # Real(ml2(3,1),dp)
  3  2     0.00000000E+00   # Real(ml2(3,2),dp)
  3  3     1.00000000E+04   # Real(ml2(3,3),dp)
 

What looks like that you put these numbers by hand?

Cheers,
Florian

[jackaraz]

Hi Florian;

Code: Select all

Block MSL2 Q=  1.39738515E+03  # (SUSY Scale)
  1  1     0.00000000E+00   # Real(ml2(1,1),dp)
  1  2     0.00000000E+00   # Real(ml2(1,2),dp)
  1  3     0.00000000E+00   # Real(ml2(1,3),dp)
  2  1     0.00000000E+00   # Real(ml2(2,1),dp)
  2  2     2.50000000E+03   # Real(ml2(2,2),dp)
  2  3     0.00000000E+00   # Real(ml2(2,3),dp)
  3  1     0.00000000E+00   # Real(ml2(3,1),dp)
  3  2     0.00000000E+00   # Real(ml2(3,2),dp)
  3  3     1.00000000E+04   # Real(ml2(3,3),dp)
 

What looks like that you put these numbers by hand?

Yes I'm putting soft masses by hand for scalar leptons and scalar neutrinos.

Sorry, I found a problem with the model file: SNUMIX is used twice. MG uses therefore the wrong matrices.

You need to change parameters.m as

Code: Select all

{ZVL,	    {Description -> "Neutrino-Mixing-Matrix",
             LaTeX->"Z^{V_L}", 
             LesHouches ->  SNULMIX,
             OutputName-> ZVL             
             }},

I actually wasn't using originally your parameter.m so in my case my ZVL was containing SNUMIX instead of SNULMIX and it stated only once in my case. Do you think it may cause a problem? Also isn't it just neutrino mixing matrix, how can it make that much of a difference because my neutralino basically doesnt decay at all the width is around 10^-9 GeV according to SLHA file, and I dont understand how it is that much related with neutrino mixing. I'm attaching my model files, should I change LesHouches of my ZVL to SNULMIX?

Thank you very much for every thing.
Best regards

[FStaub]

Hi,

something strange goes on. How did you set the slepton soft-terms? By hacking the code or modifying SPheno.m in SARAH? Maybe, it's the easiest if you can also attach your SPheno Fortran files to check why you get such a small width.

Cheers,
Florian

[jackaraz]

Hi all

Due to a problem in the website (or my connection), I couldn't ask my question from here. Thus I'm adding our conversation with Florian here.

Thank you very much by the way, as you said the problem was version of SARAH.

------------------------------------------------------

My email:

Yes I set the soft masses as free variables from spheno (spheno.m is attached). Because my research required such an arrangement. I just realised that for that partucular parameter space selection in the event file that I sent you, neutralino1 is NLSP thus that might be the reason of such low decay width (which still feels extremely low, for other cases it is still ~10^-6 GeV) but it still doesn't explain to me the difference between mg5 and spheno and I couldn't see any reason between slepton softmasses and neutralino decay width, which is not just neutralino but it effects charginos as well (width ~10^-3 with respect to SPheno). I checked the UFO file as well to see if there is any mistake but couldn't find anything that might explain this difference.

----------------------------------------------------

Florian's response:


Hi,

actually, I just remember that there was a problem with the UMSSM model files:
- before 4.6.0 no right-handed neutrinos were included
- These were added in 4.6.0 BUT the hypercharge was wrong (1 instead 0)
- This was fixed in the next update

The hypercharge in the model files you sent me is correct. But, might
it be that you used the buggy file to generate the UFO files? Because,
it seem that I get only the large widths from MG if I use that
particular version.

Cheers,

-----------------------------------------

Thank you very much
Best regards