This page aims at presenting color images recorded with a T250F3.5 telescope and a thinned backside illuminated Marconi47-10 CCD.

For the first time, using a Marconi47-10 thinned CCD , I wanted to make some tricolor images of the sky and test the capabilities of a such a CCD. I know also that I am not observing in a very good place, I was wondering how far this place degrades the ability of making tricolor images.

I have used the Edmund scientific set of RGB dichroic filters, after many doubts about IR leaks of these filter, I have decided to measure them using a dedicated machine to measure filters and here is the transmission plots :

RGB transmission of Edmund scientific dichroic filters (through out all the CCD sensitivity)

This is simply astonishing how bad they are before 400nm and beyond 700nm.
- BLUE and GREEN filter suffers from large leaks above 700 nm, the use of a BG38 filter is mandatory.
- The RED and GREEN filter suffers from UV leaks between 300 to 400 nm, with large peaks and valleys as shown into the next close up :

RGB transmission of Edmund scientific dichroic filters (close up 300-520nm)

If compared to CCD that are sensitive to UV (300-400nm band), the UV leaks can be a real issue !! In any case do not use these filter for any photometric application !!

Do you wanna see what it does if you are not using a filter which blocks IR light ?

Left, star image magnified 10x, B filter with no BG38 filter, right B filter with the BG38 filter, since I am using a Ross field corrector made of glass, the IR image of the star does not match exactly with the blue one. This is fortunate because I could figure out leak problem related to the blue filter from the Edmund scientific set.

The transmission of the BG38. One can conclude, that for sure no IR leaks can arise IF the BG38 is added to the blue filter. And this is what I did and paid off. For the UV leak from the V filter, to cure it, use the standard Schott BG18+GG495 set instead of the Edmund scientific filter.

Demonstration that the Green filter suffers from a massive IR leak A landolt Field [18h42m47.600s, 00°14'38.79''] has been used to check the IR leak of the V filter.

Left the V filter image without BG38 filter, right with a BG38, the right image confirms : all red star dimmed means this filter suffers from massive IR leaks.

A BG38 or equivalent filter must be used to block the IR light. KG1, 2 and 3 filters will partially remove the IR leak.

The Red filter, for sure, is indeed a R and I filter. If the R filter has to be turned into a true visible R filter, the IR light has to be blocked with a KG3 filter. BUT the UV leaks prevent it from making any photometric measurements.

This is interesting to compare the Edmund scientific set with other set of filters like the SCHOTT one.

The next table gives THE photometric filter set (note that the WG305 is useless, as well the GG385, because the silicon does the job very well to cutoff these wavelength)...

This table gives the filter combinaison used to make the photometric LANDOLT catalog

The Landolt catalog provides with many 30' fields spread in the Sky, with stars very well calibrated for the UBVRI bands, with an accuracy of 1/1000 magnitude...

Anyway, I made some test with the famous M57 planetary nebula, using the following exposure time : R=120sec, G=120sec, B=240sec, all are individual from 60s exposures and the images are stacked. Using the thinned CCD featuring a very high blue sensitivity is really great, because the blue frames are really coming with high SNR (let's say almost over exposed !), Nevertheless, I was first puzzled by the following, and then realize what were the issues :

M57, approximate 1° diagonal field of view.

Having a closer look to the M57 spectral strongest lines of the ring, it can be noticed that the strongest light are located at :

Strongest planetary nebula emission lines, compared with the Edmund Scientific filter set, the OIII line can be either blue or green.

The inner of the nebula part is glowing at 4686A [HeII] Blue. Looking to the filter transmission, the blue filter looks to take over the 4959 and 5007A lines, that could explain the color of the outer ring think that the blue filter overlaps to much toward the V band. Also I did not block the IR light with a KG3 filter when using the R filter.

The M27 Dumbell Nebula, same exposure time as M57 (V filter left, V filter+IR blocking right)

The M27 Dumbell Nebula, here RVB Schott filters (described in Table 2) were used. The OIII line of the nebula has been taken over by the Green filter, because the Schott green filter lets 80% of the OIII light, whereas the blue lets only 5%. The figures are opposite for the Edmund Scientific filters, where 80% of the OIII line goes thru the blue filter.

The Ngc6992 Veil Nebula, same exposure time as M57 (V filter left, V filter+IR blocking right)

The Gamma Cygnus region HII complex, same exposure time as M57 (V filter left, V filter+IR blocking right)

References

Links