CAMB 'SwitchoffTime'

Use of Cobaya. camb, CLASS, cosmomc, compilers, etc.
Post Reply
Lachlan Lancaster
Posts: 2
Joined: July 03 2015
Affiliation: Carnegie Mellon University

CAMB ''SwitchoffTime''

Post by Lachlan Lancaster » July 03 2015

Hi,

I've been modifying the CAMB code to account for possible non-SM interactions. In doing this I have had trouble figuring out how to implement an appropriate 'TightSwitchoffTime' for the tight-coupling period of these interactions. I originally set it as:

TightSwitchoffTime = EV%k_buf/opacity

Where here the opacity refers to the opacity for the specific interaction. However this was just guess work as this is my first time doing meaningful modification to the CAMB code and doesn't seem to be working very well at all.

I tried comparing this to the equivalent for the Thompson Scattering process but was unable to derive a meaningful interpretation for the arguments involved, namely:

TightSwitchoffTime = min(tight_tau, Thermo_OpacityToTime(EV%k_buf/ep))

I realize this essentially comes down to translating the opacity of this interaction to a time in units of Mpc, but is there an efficient way to do this with minimal modification to the code (such as not creating an equivalent 'Thermo_OpacityToTime()' function)?

Thanks in advance for any advice!

Antony Lewis
Posts: 1941
Joined: September 23 2004
Affiliation: University of Sussex
Contact:

Re: CAMB ''''SwitchoffTime''''

Post by Antony Lewis » July 03 2015

I assume you're not actually reusing the Thomson switch name "TightSwitchoffTime" if you are doing something similar for a different interaction . Unless you have an analytic relation between conformal time and opacity, you may need to find the relation numerically (as done for Thomson).

(or, if you are just modifying the total photon interaction rate, you need to modify the total opacity, but not specifically the switch time?).

Lachlan Lancaster
Posts: 2
Joined: July 03 2015
Affiliation: Carnegie Mellon University

CAMB ''SwitchoffTime''

Post by Lachlan Lancaster » July 03 2015

Hi Antony,

Yes I am modeling a different interaction than Thompson scattering and am thus using a different variable name. Puzzling over it more yesterday made me realize I basically needed exactly what you suggest, an analytic or numerical relation between opacity and conformal time. Now I just have to figure that out! Thanks for confirming that I wasn't going mad and that the code was doing what I thought it was!

Just as a clarification, my impression was that the 'tight_tau' variable was referring to H/opacity and the 'Thermo_OpacityToTime()' part of the min() arguments was referring to k/opacity as is discussed in section 6.2 of (http://arxiv.org/pdf/astro-ph/9804301v2.pdf). Is this the correct impression?

Thanks!

Antony Lewis
Posts: 1941
Joined: September 23 2004
Affiliation: University of Sussex
Contact:

Re: CAMB ''''SwitchoffTime''''

Post by Antony Lewis » July 04 2015

Yes, I think so: the two timescales are where the interaction time is of order the Hubble time, and where the interaction length is order of the perturbation wavelength.

Post Reply