CAMB_Source:Angular power spectra for low z source number counts

[Song Chen]

Hello, I try to make the Angular power spectra for number counts.
Could you tell me which output file correspond to the number count CL?
Another question regards to the count_obs_window_z(Win, z, winamp).
according to your code:

dz = z-Win%Redshift
winamp = exp(-(dz/Win%sigma)**2/2)
count_obs_window_z =winamp/Win%sigma/root2pi

could you plz explain more detail about count_obs_window_z and winamp?

[Antony Lewis]

Output files are documented in the Output section of the readme:

http://camb.info/sources/

winamp is the window function normalized to peak at 1. count_obs_window_z is the normalized distribution.

[Song Chen]

Output files are documented in the Output section of the readme:

http://camb.info/sources/

winamp is the window function normalized to peak at 1. count_obs_window_z is the normalized distribution.

Thanks for your answer.
Sorry I am still confuse about the first question. If I understand readme correctly, the number count Cl should be in the scalar_covariance_output_file where ij is the T, E, Phi, win_1,win2.
Is CL^(win_1 win_1) correspond to the number count CL of win_1 param in the .ini file?
I modified the count_obs_window_z(Win, z, winamp) without use the win_1 params.

but some times when I vary Win_1 params I get
"thermo: shouldn't happen"