Nikon D5300 vs QHY294C

In 2020 I bought a QHY294C and had lots of opportunities to test out this dedicated cooled OSC astrocam. Up until 2020 I had been using a Nikon D5300 with was HA modified by Kolari Vision. The D5300 and QHY294C are 14 bit cameras.

The Ha modification did improve the camera’s sensitivity and I noticed more detail in faint galaxies. I decided to start putting together some comparison images. The images presented here used master DARKs, FLATs and BIAS. The Bubble Nebula was my first comparison.

This is a cropped 2019 image of the bubble nebula taken with my D5300 and my 8″ EdgeHD at 2032mmFL. 17 images taken at ISO200 200 second exposures. Total integration time is 2 hours 8 minutes. I always thought that the Bubble Nebula was a dim object.

No filters were used here simply because I never invested in filters for this camera. I figured that anything that would attenuate the light getting to the camera would push the exposure times out to impractical values. I also found that the 4000 x 6000 pixel RAW images always benefited from being binned 2 x 2. I use a RAW binning technique that achieves the same noise reduction as a MONO CCD camera (50%).

I was able to pull out some faint nebulosity…Enough to tell that there was more nebulosity hiding in the dark. I couldn’t get any more out of the D5300 because the star saturation would be terrible to process. As a test I decided to redevelop the image only this time I was really going to stretch the data.

I was able to see more nebulosity but the stars were getting bloated.

I spent 2020 imaging the same DSOs. This time I had the ability to use filters and the Bubble Nebula just cried out to use a narrowband filter.

This is the image I took in 2020 with the QHY294C and my 8″ EdgeHD at 2032mmFL. The imaging chip in this camera is a Back Side Illuminated (BSI) which significantly increases it’s sensitivity. It is also a cooled camera and, for this image, the camera was cooled to 0 Celcius.

I also made the very wise decision to buy the bits and pieces so I could use filters with this camera. This particular image was obtained using the Optolong L-eNhance Narrowband filter. This filter does cut down on the amount of light reaching the camera so longer exposures are possible without saturating the stars. I used a camera GAIN=1600 and 600 second exposures. A total of 18 photos were taken yielding a total integration time of 3 hours.

There is 50% more exposure time invested in the QHY294C image which is a starting point. The QHY294C is a much more sensitive camera. The quantum efficiency of a stock D5300 is roughly 55%. The QHY294C has a QE of roughly 75%. The QHY294C is a cooled camera which means the DARK frames will have low noise. I believe this allows me to stretch the image further before noise stops me. The narrowband filter tames the stars which also allows me to stretch the image further before the star saturate. The LeNhance filter also gives me a wonderful rich palette of colours to work with.

I think I have to give the Bubble Nebula to the QHY294C. Lovely rich colours that processed easily. The L-eNhance Narrowband filter also tipped the scales towards the QHY294C. Maybe I might decide to try the D5300 with the L-eNhance filter…I think I might be able to cobble something together that still respects the need to obtain critical focus.

Another DSO subject that worked out very well for me is the Pelican Nebula. This is the image I obtained in 2019 using the Nikon D5300 with no filters and my BK80ED refractor at 510mmFL. 27 images taken at ISO200 452 second exposures. Total integration time was 3 Hours and 23 Minutes. The D5300 certainly pulled out lots of data. I re-processed this image for this comparison only because development techniques also change over the years and I wanted that caveat ruled out.

This is the image I obtained in 2020 using the QHY294C with the L-eNhance filter and the BK80ED refractor at 510mmFL. 23 images taken at GAIN=1600 600 second exposures. Total integration time was 3 hours 50 minutes.

As with all of the photos I have taken with the L-eNhance filter…the colour palette is very rich. Without the ability to use a narrowband filter, the D5300 colours tend to be more pink. I also noticed that the stars were much harder to control with the D5300. It took a lot more work to tame the brightness of the stars. The L-eNhance filter pushes back the stars just enough that they are very easy to dim so they don’t dominate the image.

I always bin the D5300 images 2×2. The 24Mpixel images tend to make processing very time consuming. Binning the image 2×2 makes the process faster and also improves the SNR. The comparison images shown here are all downsized so the X axis is 2000 pixels. The calibrated D5300 images are 3008 x 2008. The calibrated QHY294C images are 4164 x 2794.

The D5300 has 3.9uM square pixels which, when binned, makes them effectively 7.8uM square pixels. The QHY294C has 4.63uM square pixels. I’m not entirely sure of this “insight” but the 7.8uM size pixels with a quantum efficiency of 55% is roughly equivalent to the 4.63uM size pixels with a quantum efficiency of 75%. I’m thinking here about the number of electrons being collected per second per pixel in each camera.

The D5300 has an APC size sensor. The QHY294C has a smaller 4/3 inch sensor so the field of view of the QHY294C camera is smaller than the Nikon D5300. I thought long and hard about this smaller field of view but I have not seen it as a compromise.

I am still curious to see what is possible if I use the Nikon D5300 with the L-eNhance filter. I know that it will require very long exposure times. The cooling and lower read noise of the QHY294C will always be seen as an advantage. My gut feeling is that my first choice will be the QHY294C for the foreseeable future.

I think the QHY294C wins hands down. I am thrilled with the QHY294C and look forward to many hours of enjoyment. I wasn’t sure if a narrowband filter was going to be a wise purchase. Now it’s my favorite. I have been putting off purchasing a dedicated astrocam for a few years. Now I am glad I did it.

Peter