Understanding CCD size and matching to telescope

[Montana]

Can anyone help me understand how you pick the right camera chip for the telescope?

I am trying to understand this https://astronomy.tools/calculators/ccd_suitability

However, every camera and telescope combination I own, it says I am hugely over sampling and it is not good. Therefore, are any chip sizes any good? I am a bit lost as to know how you know whether you have the right one or not.

Many thanks for your help
Alexandra

[dragracingdan]

Hello,

It's not about the size of the chip but the size of the pixels on that chip. The size of the chip will affect the FOV. The pixel size with affect the resolution

Best,
Dan

[Montana]

Thanks Dan,

Yes, I am trying to find which cameras are a match for the imaging in terms of over sampling and under sampling. I don't understand which ever combination I use all mine are classed as 'seriously over sampling' in the calculator. I have a 3, 4.5 and 5.6um camera and all are classed as bad in all my telescopes. So how do you get a camera that does work, it would have to be a 20um pixel size to be correct which can't be right. I don't understand what you look for. So any help would be appreciated.

Alexandra

[dragracingdan]

My thought is that calculator is more so for deep sky imaging than for solar. If you are over sampling with 5.86um pixels then really there isn't many options for larger pixel cameras out there for solar. They make ccds with 9um or more pixels but for solar frame rate becomes an important factor. You could BIN 2x2 the camera which effectively doubles the pixel size but CMOS binning is mostly done with in camera processing instead of hardware binning from CCDs. The way the tech is moving is smaller and smaller pixels unfortunately. I've seen your images and they look fantastic so I don't believe sampling is as important as its made out to be. There have been many a flame wars about pixel size and sampling. I say use the best matched and affordable camera and roll with that. Probably not the answer you were looking for and I'm sure others would disagree

Best,
Dan

[Merlin66]

Alexandra,
We really are more interested in the “lucky imaging” where we collect an AVI/SER file which hopefully gets us close to the AIry limit rather than the average seeing conditions.
A rule of thumb for those conditions would be to target a recording focal ratio of x5 the pixel size of the camera.
If your camera has say 4micron pixel this would infer a fr of f20.
The size of the chip should suit your target image. If full disk, the chip needs to cover the 1/100 the focal length.
For me, I’ve standardised on the ASI 1600 for my solar imaging.

[Montana]

Thanks guys, that is really helpful for me. I think I was getting hung up on the 'proper' maths and maybe I shouldn't. I was wondering whether this was why I was getting so much noise with the ASI174 as they were talking about noise if the sampling wasn't correct. I was getting very confused. I guess the best thing to do would be to test all the cameras in the situation and use the one which gives the best results. Simple I guess

Alexandra

[MalVeauX]

Hi,

I have a calculator with the math used and has user inputs so you can play around and see what matches up. Attached.

As others mentioned, lucky imaging for solar is the same as planetary, you match pixel size, wavelength and focal-ratio for critical sampling. Longer wavelengths have less angular resolution so they sample on faster focal-ratios (this is why a simple multiplier factor of a pixel size is not sufficient, CaK to HA for example is around 59% angular resolution difference). The basic idea is to differentiate the air disc from a single pixel so that detail isn't limited to being a small square, so you need more pixels covering the idea of a single airy disc so that you can generate more shape differentiation from a square, for example. So you'll find as angular resolution increases, wavelength decreases, and the focal-ratio will increase significantly to do just that.

You can also use this to get an idea of what kind of atmospheric seeing is needed, indirectly basically, for a wavelength. If your seeing is X arc-seconds, you can indirectly calculate what wavelength and aperture, at critical sampling, would be possible in those conditions. So then you can see basically what your seeing needs to be to critical sample with a specific instrument setup at a specific wavelength. This is useful when using an SSM to see your seeing and know what your practical limits may be for the day and what to target at the time based on your goals.

Beyond matching parameters for sampling, the other consideration is of course field of view and FPS. FOV comes from sensor size. I would not worry so much about the noise of a camera at this stage, because you will be stacking 20~40 or more frames, easily, and that eliminates most noise to begin with. Instead, getting good sharp data during moments of good seeing is far more critical and that is more likely to happen with very fast FPS. A full disc at a course image scale doesn't need fast FPS, but the finer image scales at higher resolution do need faster FPS (which supports lucky imaging). For example, if imaging at 170 FPS (IMX290 for example) even half a second of good seeing nets you 85 potentially sharp frames, which is plenty to stack. So out of a 30 second run, only half a second is needed for the final image with lots of room to spare. Faster is generally always better in this case (and of course you need to set exposure time to whatever it takes to allow the max FPS to be possible to begin with; it's ok to use some gain). Keep in mind, the larger the sensor with smaller pixels, the FPS will drop of course, we are limited to throughput of USB3 currently still on most cameras. Also keep in mind larger sensors can use region of interest (ROI) to decrease the pixel array read out and increase FPS this way (so a larger sensor with smaller pixels can be a good universal camera for both full discs and high res).

See attached excel calc. All calcs and formulas are available to see. Some basic instructions provided. Should be very intuitive, stick to the colored boxes and input your information or theoretical information and press enter or click somewhere to run all the calcs. I tried to lock cells to prevent it from being broken, but if there's anything you need, I'm happy to provide an unlocked version (just know if you do anything it's possible to break the calcs).

Angular Resolution, Seeing & Critical Sampling Calculator (Martin Wise, 01272021).xls

(13 KiB) Downloaded 234 times

Very best,

[Montana]

Thanks Marty!

Alexandra

[Creativspelerr]

Thank you for the spreadsheet Marty!
Do you suppose you could post it with edit permissions - there are some cells that calculate numbers that don't fit in the cell and show "######" and in "protected" mode I can't even resize the cell.

[rsfoto]

Can anyone help me understand how you pick the right camera chip for the telescope?

I am trying to understand this https://astronomy.tools/calculators/ccd_suitability

However, every camera and telescope combination I own, it says I am hugely over sampling and it is not good. Therefore, are any chip sizes any good? I am a bit lost as to know how you know whether you have the right one or not.

Many thanks for your help
Alexandra

What is wrong with over sampling?

Does it degrade the image quality?

[FRAZ]

Mark posted this on my goto website for years, man that brings back some memories.


https://brierleyhillsolar.blogspot.com/ ... pling.html

[marktownley]

What is wrong with over sampling?

Does it degrade the image quality?

Hi Rainer.

Imaginge you are reading a newspaper / magazine at arms length, this is correct sampling - the text / photographs are the correct size that they were intended for viewing.

Now, view the same newspaper / magazine from the other side of the room this is undersampling - you can still see the same pictures / text as you could before but these will be smaller and you won't see the same amount of detail as you can reading at arms length.

Now, if we bring that newspaper / magazine as close to the end of our nose as we can and read, this is oversampling - the text and photos appear larger than in the previous two cases, however there is no more actual detail visible, the original detail is just larger.

I know the example above is a simple one, but if we now jump back to the world of solar then oversampling will only do one thing - increase exposure time, and, as there is a reasonable chance that if we are oversampling we are already running at a long effective f-ratio in the first place this extended exposure time is probably not the best thing.

However, there may be a scenario when a bit of oversampling could be of benefit. Normally seeing conditions are everyones limiting factor, they dictate the finest detail that can be made out. However imagine a situation where seeing is perfect, and now it is aperture that is the limiting factor - the biggest telescope is just not big enough! In this case it might be wise to oversample - collecting data with very high signal to noise ratio, then stacking and post processing may well work.

Spotting ineffective oversampling is easy though - it is those closeups that are soft, lacking definition and contrast. In this case it offers no benefit that simply doing an enlargement of the image and viewing the enlarged image.

Just my morning thoughts over a coffee!

Mark

[astroshot]

Agree 100% Mark.
Good explanation.

A little over sampling is a good thing, as you said, to capture the data during those moments of good seeing.

The only downside of excessive over sampling is the large data files, as the camera is recording data it can never use.

[TareqPhoto]

Nice topic and back old eternal subject.

Now i am trying to manage my setup so i can have all cameras and scope at least at ideal sampling if possible or at least good enough over sampling so i can have nice results, because i will do the three types/bands of solar imaging then i prefer to have three cameras as i will try to have them all together rather than swap, but i am sure i will try to not good with a bit expensive third camera like asi432 or 174, but i will think about using one of my already cooled cameras available for that so i won't buy another one.

[Alto]

Relating to this, which is better (solar imaging that big bright thing that occasionally appears) masking the objective, or adding macro. The latter, of course, adds at least another bit of glass and dust.

[marktownley]

Relating to this, which is better (solar imaging that big bright thing that occasionally appears) masking the objective, or adding macro. The latter, of course, adds at least another bit of glass and dust.

I think both has its place

[cybermayberry]

https://www.wilmslowastro.com/software/formulae.htm

Is a handy resource as well.
The CCD Planetary Critical Sampling section is basically the same thing that Marty provided.

[Alto]

Chuckles. Mark's on the fence - but thanks.

Aye, the astronomytools site gives a similar tool for sampling. I was just casing the preferences on way or another for how to slow the scope down from the more experienced guys and gals.

[marktownley]

Chuckles. Mark's on the fence - but thanks.

Maybe I should have expanded more If i'm imaging in CaK I like to stop down as this is an effective way of tuning the filter. At short (full disk) focal lengths this is fine, but then realistically about 1200mm in a refractor is the longest easily commercially available focal length you can get, so, in this instance I might use a barlow to increase the focal length and get more appropriate sampling. In CaK I like to keep glass in the optical path to a minimum. There are other benefits to using a stopped down scope in CaK too.

If you're talking Ha front mounted etalons I use a barlow to get correct sampling - front mounted etalons are expensive and so want to use them to the max. Rear mounted Ha etalons I will use different scopes to match the conditions, sometimes stopping down to reach effective sampling for the seeing.

If i'm doing WL full disks (Continuum or G Band), for full disks, I will often stop these down to match the seeing. Closeups I just barlow.

This certainly won't be applicable to everyone, but in my case seeing is often the primary governing factor, with it being rare I get <1" seeing. To give a perspective in CaK the following aperture values correspond to the following seeing conditions:

40mm = 2.5"
50mm = 2"
60mm = 1.6"
100mm = 1"

So for me, stopping down is easy to match day to day seeing condition. If my seeing was <1" (sub arc second) more often I would have to be using image amplification (telecentrics) and less or no stop down to image critically sampled to what the conditons allowed.

I certainly haven't got off the fence but I have wriggled about on it a bit more