Webcam imaging with Philips ToUcam 740 Pro
An UV/IR-cut filter helps removing the effects of atmospheric refraction / atmospheric dispersion in CCD-imaging. The direction of light as it passes through the atmosphere is changed because of refraction, the direction of refraction is that a sky object apparently moves towards the zenith. The amount of this apparent movement is wavelength dependent, the shorter the wavelength of light the greater the resulting movement so every object actually appears as a little spectrum with the blue end towards the zenith. This effect is called atmospheric dispersion.
If we image a sky object with a color CCD cam we usually record Red Green and Blue channels. The color CCD features a filter matrix with R, G and B filters for each pixel to record R;G and B for each final pixel separately (in fact the ICX098BQ CCD of a ToUcam pro has a Bayer filter matrix but we can live with the simple RGB model here...). The R filter lets only red light pass, the G filter green light and the B blue light. Unfortunately the RGB filters let also infrared light pass. Near infrared light and blue light reside on "opposite sides" of the visible light spectrum so they will be moved quite noticiable from each other by atmospheric dispersion. The B sensor of the CCD will record blue light and the shifted near infrared light resulting in an unsharp muddy blue channel or ghost images in the blue channel at worst. The same applies to the green channel with a smaller shift between near IR and green.
The solution for this problem is an IR-cut filter to block (near) infrared light. The standard lens of a ToUcam has an IR-cut filterfoil on the backlens. If we use the cam in primefocus of a scope we have to use a separate filter like the UV/IR-cut filter from Baader:
Baader UV/IR-block filter
Transmission of Baader UV/IR-cut filter
The problem of shifted R, G and B channels due to atmospheric dispersion will remain. This can be solved by realigning the color channels in image postprocessing.
It's a good idea to always use such a filter when imaging through non-apochromatic
scopes like FH refractors to cure the scope's color dispersion problems
. For apochromatic scopes like newtonians, MCs, APO refractors... an IR-cut
filter is only needed to cure atmospheric dispersion. Since the dispersion
is dependend on the altitude of an object there should be a "break
even" altitude where the positive and negative efects of the filter
compensate each other for the resulting image quality. The filter "consumes"
a small but noticable amount of light even in the visible RGB spectrum.
Leaving it away allows the usage of fewer gain and/or shorter shutter
times resulting in less noise and better "seeing freezing".
In addition each air/glass glass/air transition in the optical path worsens
image quality even if the glass surface is coated very well to reduce
The "break even" point should be the altitude where the shift between blue and near IR is well below one pixel on the CCD so that IR-light can't reduce sharpness of the B-channel. Discard an IR-cut filter for imaging equal or above this altitude, use it below this altitude.