I have a function to calculate the Hubble parameter at a given redshift:
$$H(z)=\sqrt{\Omega_R(1+z)^4+\Omega_m(1+z)^3+\Omega_k(1+z)^2+\Omega_{\Lambda}}$$
And I have another function to calculate the conformal time between two redshifts:
$$\eta(z1,z0)=\int_{z1}^{z0}\frac{1}{H(z]}dz$$
So now I want to calculate the particle horizon at the time of recombination.  I calculate
$$D_{PH}=c\space \eta(z_{CMB},\infty )$$
Have I just calculated the particle horizon at $t_{CMB}$ as it would be measured today after the expansion (comoving), or have I calculated the particle horizon as it was at $z_{CMB}$?  I want the actual (proper) particle horizon as measured by an observer 380,000 years after the big bang.  Do I divide the value returned by the $\eta$ function by $z_{CMB}$?