Light Pollution Rate formula

[carlosgib]

Hi,

I am trying to find the exposure time, how to calculate the ideal exposure length for my CMOS (QHY268C) cameras with a CLS clip filter fitted?

For example, I have calculated my ideal sub length using the formula of: 10x (read noise sq/LP in electrons).

https://www.youtube.com/watch?v=ub1Hjvl ... ex=10&t=2s

Where I live I have a Bortle 4 to 5, my scope F is 7, the camera has a QE of 80%, and noise read of 0.7.

Apply the formula = 10 x ((0.7>2)/1.90) will result in an exposure length of 2.6 second.

There are three questions:

I am using a Color CMOS should I multiply by 3 instead of 10, or multiply the result above by 3 ??

Second question, the LP filter will reduce the light entering the sensor, maybe I can use the histogram level to workout the reduction of light by taking exposure with and without the filter, but I am more of trying to find the difference by working it out mathematically, so is there a formular that i can use to work it out.

Third question, my camera has a QE 80%, compare to the scale using 50% in the formula, how can I work to include the QE 80% ??

Thanks

Charles

[oopfan]

Hi Charles,

I visited QHY's website for your camera. It says:

0.7e- Read Noise at High gain
3.5e- Read Noise at Low Gain

In your calculation you mentioned using 0.7 for Read Noise. Is that the plan to image at the highest gain setting? If true then there is a knock-on effect that diminishes your Full Well Depth (FWD) from 75,000 electrons down to probably around 10,000 electrons. That leads to star saturation very rapidly. Take it from me you don't want this. My camera's FWD is 13,000 electrons and it saturates very quickly. So although your camera is capable of it, is this what you want?

Also, I noticed that you used the number "10" as the factor in the calculation. I see that you got that from Robin's YT video. Robin has an online calculator, which I can't seem to locate now, that uses "20". The difference is that using "10" means that you are willing to accept a higher level of noise per frame. I use "20". You can use a higher number if you wish.

So my off-the-top-of-my-head analysis is that:
a. You are willing to accept low dynamic range images.
b. A higher than normal amount of noise per frame.

For your camera I can't seem to find a Read Noise vs Gain curve but my guess is that you should be setting your gain such that Read Noise is approximately 3.0 electrons. Also notice that Robin's video's footnote says that if you are using a color camera that you need to divide the light pollution by 3. Therefore the formula becomes:

Exposure = 10 * 3 * 3 / (1.9 / 3) = 142 seconds

The 1.9 comes from Robin's video which is for an f/7 scope, Bortle 5 skies, and a MONO camera. Since you are using a color camera you need to divide that number by three. I am sure that tomorrow Robin will reply with a link to his online calculator. The numbers will change once you plug in your camera's QE of 84%.

Brian

PS: Using a CLS filter means that you need to increase the exposure or accept a higher level of per-frame noise.

[turfpit]

Charles

A search on Astrobin for a QHY268C turns up 2 pages of images https://www.astrobin.com/search/?q=qhy268c&page=1

On the 2nd page are a couple of Image Of The Day - these are both using 120s exposures in a Bortle 5 sky (using a UV/IR cut filter).

Have a look at the other images, particularly the Top Picks, to get some ideas of how the different equipment and exposures translate into real images.

A CLS filter blocks the light of the spectral lines of mercury and sodium-vapor lamps. If your local lighting is the modern LED type then the CLS may not help much. Adding any filter into the optical train must impact the desirable exposure setting.

For me, getting a histogram which looks like this tends mean the camera settings will produce decent results for a given sky. M42 is a bright object so 60s works for me at Bortle 6. For dimmer objects I would be thinking 120s. If I get to Bortle 4 I can increase that (for the right target, say M33).
YMMV.

M42-60s-Altair183C-single-FITS-frame.JPG (62.41 KiB) Viewed 3572 times

From my own experience a set of 100+ frames when calibrated in Astro Pixel Processor will have wildly varying quality scores even on a supposedly 'clear night' with good conditions, so I tend not to overthink this.

Dave

[admin]

Hi,

I have a calculator website that will calculate the sky background electron rate for you providing you enter your bortle number, F ratio and camera sensor statistics.

https://tools.sharpcap.co.uk/

This gives 0.56 e/pixel/s for bortle 4.5, f/7, 80%QE, 3.75um, colour.

At minimum gain with 3.5e read noise and 50000e FWD , that would give about 220s exposure.

At gain of about 2x, the read noise drops to about 1.5e (17000FWD), about 40s exposure

You would not use maximum gain for DSO work - the FWD would be tiny - in the 100s of electrons.

Robin

[carlosgib]

Will like to thanks all for your input , really appreciate it.

Best

Charles

[ViggenAJ37]

Thanks, that will be worth its weight in gold! (well, weight, that's a couple sentence on a site... let's convert bytes to grams for the sake of the comparison). The thing I'm struggling the most these days is light pollution, as I'm living in a city; fortunately, where I live, there isn't too much light pollution and you can see the stars at night when the weather's clear, but I still prefer driving to the hills nearby to get a clearer picture. I could drive further so that I don't get any light pollution at all, but well, driving on unlit countryside roads at night isn't exactly my favorite activity.

And since my wife's and my plans to get one of these Athens house https://tranio.com/greece/attica/athens/detached/ for the holidays has been delayed by the COVID pandemic, I have to make do with urban light pollution.