3nm Custom Scientific Halpha filter scan provided by vendor with filter purchase (individually measured)
Goal was to limit blooming to a reasonable level, capture as much nebulosity as possible and to examine the background level to determine which filter offered the best signal to noise optimization
As predicted by theory of interference filters, AstroDon 6nm Ha filters have greater passband throughput than do Custom Scientific 3nm Ha filters. Early subjective tests indicate when compared to an AstroDon, about 50% longer exposures need to be made with Custom Scientific 3nm filters to reach approximately the same amount of passband signal.
To better quantify signal to noise characteristics of the two filters I shot the same target using each filter. However I used a 15 minute subexposure with the Custom Scientific versus a 10 minute subexposure with the AstroDon in order to get similar amount of nebulosity signal with approximately the same level of blooming of the bright star in the image.
The Custom Scientific filter was exposed for 15 minutes but had a background ADU count that was significantly lower than the background from the AstroDon filter that was exposed for 10 minutes.
The amount of signal in the nebulosity that is above the background level is similar: 211 ADU for the AstroDon and 225 ADU for the Custom Scientific.
Since the sky background noise is a function of the ADU count, the higher background ADU count of the AstroDon filter exposure indicates a higher level of noise. When the nebulosity signal delta is divided by the background level a figure of merit can be computed.
My tests indicate the nebulosity signal level compared to the background was 25% for AstroDon and 47.3% for the Custom Scientific filter. I therefore conclude that the narrower Custom Scientific filter has a better signal to noise ratio when being exposed with a bright sky background (near full moon). This means the Custom Scientific filter will give a better contrast than the AstroDon if exposed for a longer period (about 50% more). Each filter's exposure had similar amount of Nebulosity recorded and had similar level of star blooming.
I also subjectively noticed that there is less of a light gradient in the Custom Scientific image. None of these results speak badly of the AstroDon filters, it is exactly what the theory would predict.
The Custom Scientific filter needs about 50% longer exposure time than the AstroDon to get about the same level of nebulosity signal over background. Due to the sandwich construction used by the Custom Scientific filter, internal reflections can and do occur, particularly with bright stars and long exposures. The Custom Scientific results show lower sky gradients from my comparisons and have lower noise backgrounds.
My view is the Custom Scientific filters are the better choice in bright sky conditions such as shooting from the city or with a bright moon. If exposure time is a problem the AstroDon will have the edge. If shooting from a dark sky background, I would prefer the AstroDons. If using a slow focal ratio telescope, then I would prefer the shorter exposure time of the AstroDons.
There is no perfect filter for all conditions
Raw data follows:
Frames are calibrated and no further processing was applied.
ASTRODON: Sky Background ADU count = 843.5 ADU
ASTRODON: Nebulosity ADU count = 1054.4 ADU
Custom Scientific Results
Custom Scientific Sky Background ADU count = 476.4 ADU
Custom Scientific Nebulosity ADU Count = 701.6 ADU
Images for comparison (one hour total exposure per image)
click to enlarge
Image 1: Astrodon 6 x 10 minutes. Registar aligned, sigma combined, DDP w/o sharpening, curves and levels adjust. Note gradient.
click to enlarge
Image 2: Custom Scientific 4 x 15 minutes. Registar aligned, sigma combined, DDP w/o sharpening, curves and levels adjust.