RGE running stops and NaN in Bmu

[Amin]

Hi Florian,

In the MSSM Stueckelberg model the mixing between the hidden sector and the MSSM one is very small. So the RGE running should not be affected and must proceed as in the MSSM (with gauge coupling unification achieved at GUT scale). I have tried many different things to make it work but couldn't and kept crashing with:

Calculating mass spectrum
1 .-iteration
RGE running not possible. Errorcode: -4
Note: The following floating-point exceptions are signalling: IEEE_DIVIDE_BY_ZERO IEEE_UNDERFLOW_FLAG IEEE_DENORMAL
STOP Subroutine TerminateProgram

Then I set the GUT scale by hand which made it work. However, increasing MDBpS above a certain value causes this:

Calculating one loop decays
NaN appearing in solution of tadpole equations for Bmu
Note: The following floating-point exceptions are signalling: IEEE_INVALID_FLAG IEEE_DIVIDE_BY_ZERO IEEE_UNDERFLOW_FLAG IEEE_DENORMAL
STOP Subroutine TerminateProgram

Using Mathematica, I checked the expression of Bmu by solving the tadpole equations and I couldn't understand how a NaN can be obtained.
I will attach the model files again with the error output from the NaN in Bmu. I hope you can give me a suggestion on how to fix this issue.

Thanks a lot,
Amin

[FStaub]

Hi,

since the second problem appears just for the one-loop decays, that might be a subtle problem with the renormalisation procedure in that calculation. I guess, for your purposes, you can turn stick to tree-level decays, or?

When you fix the GUT scale by hand: what are the running values of g1 and g2 at the GUT scale?

Cheers,
Florian

[Amin]

Hi Florian,

Yes, I guess I can stick to tree level decays.
The values of g_1 and g_2 at the GUT scale are very different. Setting the scale by hand to around 10^16 or 10^17 gives g_1 around 0.68 and g_2 around 0.83. Only when I go up to 10^21 or 10^22 is when the couplings approach each other. All of this happened after turning on gauge kinetic mixing. I checked the RGEs and they all make sense. Turning off the Stueckelberg sector decouples all equations from the MSSM. So if I switch it on with a small mixing then the running of the couplings should not be drastically affected and unification must be achieved. This is puzzling ... unless I am missing something in the SARAH implementation.

Best,
Amin

[FStaub]

Hi,

the 0.68 for g1 sounds kind of okay (should be 0.7-0.72 for TeV SUSY), but 0.83 for g2 is much too big.

I just tried your SPheno.m and it seems to work for me using the default values in the Les Houches file. Note, I also uploaded SARAH 4.14.0 this morning. Maybe, you can give a try with that version.

Cheers,
Florian

[Amin]

Hi Florian,

I tried the new SARAH version; nothing has changed.
Yes, the default values in the LesHouches file work, but try setting, for example, M2Input to zero and M1Input to 200 or more then you can see SPheno breaks. What I can't understand is that this is almost like the MSSM since the hidden sector couples weakly to the MSSM and so the gauge coupling RGEs should not be dramatically affected!!

Thank you,
Amin

[Amin]

Hi again,

In my model, the mixed gauge couplings, g_YB and g_BY, are related to the U(1) gauge coupling g_1 such that g_YB = -g_1*sin(delta), where delta is the coefficient of kinetic mixing. Is there a way to tell SARAH to write RGEs only for g_1, g_2, g_3 and g_p (new gauge coupling of U(1)') since g_YB is given in terms of g_1? I tried the "Dependence" option in parameters.m but it didn't work.

Thank you,
Amin

[FStaub]

Hi,

no, because this rotation is not scale invariant, i.e. you would need the RGE for your angle delta. Thus, it's easier to let gYB and gBY run.

Cheers,
Florian

[FStaub]

Hi Florian,

I tried the new SARAH version; nothing has changed.
Yes, the default values in the LesHouches file work, but try setting, for example, M2Input to zero and M1Input to 200 or more then you can see SPheno breaks. What I can't understand is that this is almost like the MSSM since the hidden sector couples weakly to the MSSM and so the gauge coupling RGEs should not be dramatically affected!!

Thank you,
Amin

Hi,

okay, that's something different then. It's not a general problem, but one with specific parameter combinations. Well, one would need to check what you parameters mean for the masses of the particles...

Cheers,
Florian

[Amin]

Hi Florian,

Thank you. I have one final question: the error causing the RGEs to stop running is due to SPheno not being able to satisfy g1==g2 at the GUT scale. Looking at the RGEs, g2 is not affected by gauge kinetic mixing; only g1 is. So how come g2 is turning out to be that large? Also, modifying the input M1Input which is MBDpS causes this problem, but the gauge couplings g1 and g2 do not depend on this parameter.
The SARAH output makes sense but what's happening in SPheno is the strange one.
Sorry for making this too long. I appreciate your help.

Best,
Amin

[FStaub]

Hi,

it seems that the one-loop matching conditions for g2 are highly influenced by the values of M1, M2 (more precisely, the W transversal self-energy is quite sensitive on these parameters). The reason might be that these parameters also enter the Higgs masses and rotation matrices. However, I must admit that I don't understand this behaviour so far, but I won't have time the next days to take a deeper look to see if this is a bug or a (physical) feature.

Cheers,
Florian