Problem with RG's in the model munuSSM3G
Posted: 26. Feb 2019, 17:03
Hi Florian
I've found what I believe is a bug in the SPheno files generated for the model munuSSM3G with SARAH.
I've generated the files using SARAH-v4.14.0 and Mathematica-v11.3. Then, I've compile with SPheno-v4.0.3 using gfortran.
Running the code with an input file with a spectrum above 1TeV, except for the right-stop of mass around 600GeV. When calculating the decays (with the flag to run the couplings down to the scale of the decaying particle activated) the program starts complaining that some of the masses are very negative (~-1E+24), and finally crashes.
I've checked and the problem seems to be in the running of vR down to the mass of the stop. With an input value of 1600 GeV for the three of them, they blow up to ~1E+24 after the running. If I use the flag to set the RG's to one loop instead of two loops no problems seems to appear.
Taking a look inside the file with the definitions of the beta functions, I've found that the two-loop beta function for vR seems to be ill defined. since vR is a vector of dimension three, the beta function is also a vector as well, and each term of the expression should be a dim3 vector. The expression however includes some terms that are vectors, but the expression is accessing to each value of them through an unmanaged index, probably resulting on the beta function taking a random value:
[line 18237 on file RGEs_munuSSM3G.f90] betavR2= ... +8*adjkap3vR(1)*kap1adjkap1Tpkap1(3,i1)
Changing 'i1' for ':' seems to solve the problem, but something should be not working fine with SARAH I guess.
Best,
Iñaki Lara
I've found what I believe is a bug in the SPheno files generated for the model munuSSM3G with SARAH.
I've generated the files using SARAH-v4.14.0 and Mathematica-v11.3. Then, I've compile with SPheno-v4.0.3 using gfortran.
Running the code with an input file with a spectrum above 1TeV, except for the right-stop of mass around 600GeV. When calculating the decays (with the flag to run the couplings down to the scale of the decaying particle activated) the program starts complaining that some of the masses are very negative (~-1E+24), and finally crashes.
I've checked and the problem seems to be in the running of vR down to the mass of the stop. With an input value of 1600 GeV for the three of them, they blow up to ~1E+24 after the running. If I use the flag to set the RG's to one loop instead of two loops no problems seems to appear.
Taking a look inside the file with the definitions of the beta functions, I've found that the two-loop beta function for vR seems to be ill defined. since vR is a vector of dimension three, the beta function is also a vector as well, and each term of the expression should be a dim3 vector. The expression however includes some terms that are vectors, but the expression is accessing to each value of them through an unmanaged index, probably resulting on the beta function taking a random value:
[line 18237 on file RGEs_munuSSM3G.f90] betavR2= ... +8*adjkap3vR(1)*kap1adjkap1Tpkap1(3,i1)
Changing 'i1' for ':' seems to solve the problem, but something should be not working fine with SARAH I guess.
Best,
Iñaki Lara