Magnification

[robmichael81]

I'm trying to find out what magnification I will achieve when using a 3x Barlow with my 40D and therefore if it will be too much for my scope. I often use live view to capture Jupiter at 5x magnification. My scope has a focal length of 600mm and an aperture of 130mm.

I'm lead to believe that its maximum useable mag is 260x but given that this is a modest entry level scope I'm guessing it will much lower than this.

Any help on this matter would be much appreciated.

Thanks in advance

Rob

[JamesF]

Just forget all about magnification when you're talking about imaging.  What you care about is "image scale" or "plate scale", which is basically how much of the sky covers some part of the camera sensor.  Generally that will be measured in something like arcseconds per mm or arcseconds per pixel.

For planetary imaging many people try to set their optical train up so that the resolution of the OTA matches the resolution of the camera.  The maths for that is fairly straightforward, but as a general rule of thumb it works out that the focal ratio of the entire system should be five to six times the pixel size of the camera in um.  Somewhat unexpectedly for me when I went through the maths this holds for any OTA regardless of focal length or aperture.  What it doesn't allow for is rubbish seeing, but hopefully it should allow you to get the best out of those moments of good seeing when they happen.

The pixel size of the 40D appears to be 5.7um, so that would give you a target focal ratio range of between f/28 and f/34 or thereabouts.  With a 3x barlow you'd only be reaching about f/14, so there's plenty of room to play with there.  (I've assumed you're correct about the 600mm focal length, though many entry-level 130mm OTAs I can think of actually have a focal length of 650mm which would make the figures slightly different.)

I'm not sure how the 5x magnification bit works I have to admit, but if the data isn't there to start with making it bigger isn't actually going to much to improve the image.

James

[PeterCPC]

This might help http://astronomy.tools/calculators/field_of_view/

Peter

[Astrosurf]

And this! 

http://www.12dstring.me.uk/fov.htm

Alexxx

[Scorpius]

Based on an OTA F/L of 600mm, a 3x barlow, 5.7 um photosites and a 22.2mm x 14.8mm sensor - the camera's FOV would be 0.71 degrees (42 arc minutes) x 0.47 degrees (28 arc minutes and the sampling rate would be 0.65 arc seconds per photosite.

[ollypenrice]

Magnification is meaningful in visual observing because we know what we are magnifying. The image formed on the retina. This is our starting point, but we have no such starting point in an imaging system.

Olly

[robmichael81]

Great advice guys, thank you. Ultimately does this mean by increasing the magnification with a 3x Barlow (compared to my 2x) will improve the detail in my image (everything else being equal).

Sorry if you're all throwing your hands up thinking why do we bother. 

Thanks again in advance and sorry for the stupid questions. Theres so much to learn.

Rob

[JamesF]

By changing the image scale with a 3x barlow (that is, making each pixel cover a smaller area of sky) you should be able to capture more detail compared to what you get with a 2x barlow.  You may have to wait for the seeing to play ball to notice though, and that can be somewhat random.  I have sets of three-minute capture runs taken one after the other where the seeing quite obviously changes from good to bad and back again over periods of ten or fifteen minutes.

James

[Mto]

Great advice guys, thank you. Ultimately does this mean by increasing the magnification with a 3x Barlow (compared to my 2x) will improve the detail in my image (everything else being equal).

Sorry if you're all throwing your hands up thinking why do we bother. 

Thanks again in advance and sorry for the stupid questions. Theres so much to learn.

Rob

In theory yes, but in reality no. Apart from the influence of seeing the maximum theoretical resolution for a 130mm aperture telescope is approx. 1 arcsec. So, no matter how much you increase the focal length you can not capture details smaller than the theoretical limit of the optics.

[JamesF]

In theory yes, but in reality no. Apart from the influence of seeing the maximum theoretical resolution for a 130mm aperture telescope is approx. 1 arcsec. So, no matter how much you increase the focal length you can not capture details smaller than the theoretical limit of the optics.

My understanding is that once you start applying deconvolution and similar image processing techniques to captured data then it is possible to resolve better than the Dawes Limit (which is what I think you're quoting here).  I was reading up on this a couple of days ago and I believe the Dawes Limit was derived empirically for visual resolution where such post-processing is clearly not possible.  I'm happy to have it demonstrated otherwise however.

James

[Mto]

My understanding is that once you start applying deconvolution and similar image processing techniques to captured data then it is possible to resolve better than the Dawes Limit (which is what I think you're quoting here).  I was reading up on this a couple of days ago and I believe the Dawes Limit was derived empirically for visual resolution where such post-processing is clearly not possible.  I'm happy to have it demonstrated otherwise however.

James

I might quite well be wrong, but I've understood that deconvolution and other enhancement techniques work to mitigate effects of seeing and other perhaps equipment related problems (ie. small tracking errors, focus errors). In these cases basically the data is there to be dug out but if the resolving power is 1 arcsec then there is no detail smaller than that to enhance.

[JamesF]

The other issue worth consideration is that resolution varies with wavelength.  I posted something the other day about this and calculated that the Dawes Limit appears to apply at around 560nm (I think).  Angular resolution is proportional to wavelength and the visible spectrum changes wavelength by something around a factor of two from one end to the other, so the largest resolvable detail at one end of the spectrum can be half the size of the largest resolvable detail at the other.

This is why I think there's always a certain amount of "wiggle room" here, because even in mono imaging there's quite a wide range across the passband.

James

[JamesF]

I might quite well be wrong, but I've understood that deconvolution and other enhancement techniques work to mitigate effects of seeing and other perhaps equipment related problems (ie. small tracking errors, focus errors). In these cases basically the data is there to be dug out but if the resolving power is 1 arcsec then there is no detail smaller than that to enhance.

Oh, and another thing... (Apologies for the multitude of postings.  The wind here last night was so strong it was making our roof creak and the tiles clatter and I didn't get anywhere near as much sleep as I need, so I'm not quite as sharp today as I'd like to be

Ideally we also want to take into consideration Nyquist's Sampling Theorem and sample each "smallest resolvable detail" at least twice.  So even if the OTA can only resolve to one arcsecond, you might want to increase the focal length to achieve a resolution of half an arcsecond per pixel.

James

[Mto]

Yes, I agree that over sampling is the the way to go if using advanced enhancement techniques. I remember reading somewhere just a day or two ago, that when imaging, it would be best if smallest achievable detail was sampled three times. Unfortunately, at least for me, the atmosphere is the worst enemy of resolution. I'd be lucky to get 2-3 arcsec seeing few nights a year   when the weather is otherwise bearable (above -20C).

[robmichael81]

Well, I've opted for a cheap 3x barlow and will compare the results with the 2x. Just got to hope this cloud clears some time soon but I won't hold my breath. 

Thanks for all your help

Rob