According to Eq. 2 in Thepsuriya & Lewis JCAP 01 34 (2015)
the baryon drag optical depth is defined as
\tau_d \int_{\eta_0}^{\eta} d\eta' dtau/d\eta'/R
where dtau/d\eta= sigma_T n_p x_e a and R=\rho_b/rho_{\gamma}
where sigma_T is Thompson scattering cross section, x_e the fraction of free electros, n_p the number density of protons today and a the scale factor, rho_b is the baryon density and \rho_{gamma} the total radiation density
This quantity is caculated in the latest version of the fortran version of camb in the function ddragoptdepth_dz as
ddragoptdepth_dz = noreion_doptdepth_dz(this,z)/this%ThermoData%r_drag0/a
where
noreion_doptdepth_dz = this%CP%Recomb%x_e(a)*this%akthom*dtauda(this,a)
Here ThermoData%r_drag0*a =R akthom=sigma_T*n_e x_e=xe and dtauda= d\eta/da
and my question is if there is not a factor a missing from dtau/d\eta
redshift at the drag epoch

 Posts: 20
 Joined: December 10 2004
 Affiliation: Buenos Aires University