Dawes limit or pixel scale?

[M40]

I am trying to understand aspects of cameras and telescopes and like most, I am trying to get one thing to do everything, so looking for a few clues please.

Matching a camera to a telescope, would it be better to have the camera pixel scale below or above the telescope Dawes limit? Put slightly differently, which would be the best limiting factor, the telescope or the camera?

Thanks for any help.

[newbie alert]

You can drive yourself insane and then come to the conclusion thst it doesn't really matter..

Ron sums things up nicely here

 

[vlaiv]

What do you intend to do, planetary imaging or long exposure imaging?

In either case - Dawes limit has nothing to do with pixel size.

In both cases, it is bad to go with smaller pixels than are needed - question is, how do you determine size of pixels that are needed?

In planetary case, it is simple - there is maximum resolving power of the telescope and minimum pixel size is dependent on that.

found here:

https://en.wikipedia.org/wiki/Spatial_cutoff_frequency

You need to sample with two pixels per minimum wavelength corresponding to max / cutoff frequency.

With planetary imaging this is done by changing F/ratio of your system - for given pixel size you can determine critical F/ratio and then use barlow lens to produce that F/ratio.

With long exposure imaging - it is star FWHM that determines optimum sampling rate and that is not something you know in advance. You can only take average FWHM that you are likely to get for your conditions and base pixel size decision on that.

Star FWHM will depend on aperture size and optical properties of your telescope (off axis aberrations and correction of those), your mount / guiding performance and of course, the most important factor - seeing.

Once you establish likely FWHM that you will be getting with your system - then sampling rate should be FWHM / 1.6.

[M40]

2 hours ago, newbie alert said:

You can drive yourself insane and then come to the conclusion thst it doesn't really matter..

 

🤣 Sums it up for me, thanks for the video link I will look at that later 👍

[CraigT82]

2 hours ago, vlaiv said:

You need to sample with two pixels per minimum wavelength corresponding to max / cutoff frequency.

In you opinion Vlaiv, when using an OSC camera, is it better to sample blue correctly and over sample green and red? Or would you aim for correctly sampled green and under sample blue/over sample red. 

I guess that as blue is usually quite poor anyway, being effected more by seeing, it won't matter if its undersampled a little... And so it's better to aim for a correctly sampled green or even red? 

[M40]

2 hours ago, vlaiv said:

What do you intend to do, planetary imaging or long exposure imaging?

Plan is to do planetary on the skymax and EAA on the refractor. I can't see me doing many hours of imaging at the moment.

As a first step I have been looking at the ZWO ASI244 or 462. Both work really well on the skymax for planetary and then EAA on the stellamira using something like sharpcap for live stacking. I got my first EAA picture under my belt using a lodestar x2 on the refractor which was great but severely undersampled, giving a very pixellated image. So while the ZWO sale is on, I thought maybe now is the time to buy a more suitable camera. Without using either a Barlow or reducer, both give a similar field of view with the 462 having the edge. Again, without a Barlow or reducer, the 244 is just over the Dawes limit with the 462 well under which led to the question.

Looking at your answer vlaiv, I might be going about this the wrong way (no surprise there :D) so many thanks for the info as that was what I was looking for in the question. 

I must admit that I was angling down the line of the 462 and if it worked well, leave that for planetary and look at the 290 mono in the future for the EAA side of things.

 

Edited August 2, 2021 by M40
correction

[CraigT82]

Based on the spatial frequency cut-off method of calculating the correct sampling posted by @vlaiv, here is a spreadsheet of F ratios per colour and pixel size.

 

 

Planetary Imaging Sampling Calculator.xlsx

Edited August 2, 2021 by CraigT82
noticed error in sheet

[vlaiv]

53 minutes ago, CraigT82 said:

In you opinion Vlaiv, when using an OSC camera, is it better to sample blue correctly and over sample green and red? Or would you aim for correctly sampled green and under sample blue/over sample red. 

I guess that as blue is usually quite poor anyway, being effected more by seeing, it won't matter if its undersampled a little... And so it's better to aim for a correctly sampled green or even red? 

Indeed, I think it is best to go by 500 or 550nm - which would be teal or mid green as short wavelengths are the most affected by seeing and arguably will be most distorted.

Difference in between 400nm and 500nm is only 20% and if you under sample blue by 20% - it will not matter really.

11 minutes ago, CraigT82 said:

Based on the spatial frequency cut-off method of calculating the correct sampling posted by @vlaiv, here is a spreadsheet of F ratios per colour and pixel size.

Excellent spreadsheet.