Resolution Question

[sheeny (Al)]

I did a practice run a few weeks ago with the ToUcam and a camera lens for the lunar eclipse. See this thread:

http://www.iceinspace.com.au/forum/s...hlight=eclipse

To get the image dim enough for the ToUcam I had to use an improvised objective aperture... a spray can lid with a 5/16" hole in it. That's fine, but I wasn't happy with the sharpness of the image, and did some calcs to find that at that aperture Dawes limit is 14.5 arc seconds, while the pixel resolution of the TouCam is 5.77 arc seconds.

I thought I could improve the resolution by making an aperture mask with smaller holes spaced wider apart, so the total area would be the same but the effective diameter for resolution would be the distance between holes rather than the diameter of the holes themselves - same theory as a radio interferometer... right?

Well, the theory doesn't seem to work, based on my experiment today.

I shot some video of the tree across the road with both aperture masks today. Focus is the same for each video. The results of my experiment tells me the single aperture is as good as I'm going to get - the idea of using an interferometer type mask (like a Hartman Mask really) is not a good one.

The photos show the results:
The first photo is a stack of the best 100 frames from the multiple aperture mask.
The second is a stack of the best 100 frames from the single aperture mask.
The third shows the two aperture masks.
The last two are the same as the first two images but with "Dennis wavelets" (1=20, 2=10) applied in registax.

Can anyone explain to me either:

why doesn't the multiple aperture mask work to improve resolution?

or what have I done wrong in implementing this?



Al.

[[1ponders] (Paul)]

Al, can't you reduce the ToUcam exposure and gain enought to compensate?

[[1ponders] (Paul)]

Ok just read the other thread. Does the lens have a manual aperture adjust or is it electronic

[sheeny (Al)]

No, been there done that, Paul.

I had the lens stopped down to f/22 (it's manual), gain = 0, exposure = 1/10000 sec, etc. Even tried auto exposure but without the aperture mask it's just a white screen!

I really expected the lens would be dim enough at f/22 with a 2X teleconverter, but it still needs the aperture mask.

Al.

[[1ponders] (Paul)]

You'd think so wouldn't you. When you stop it down and look down the throat do you see the aperture blades almost fully closed?

[sheeny (Al)]

Good on ya Paul!!!!

The aperture stays open for maximum view finder brightness and only closes when the camera operates... the Mogg adapter I'm using doesn't trip the aperture lever at all so I can fiddle with the aperture ring all I want and the aperture won't close.

Problem solved I suspect! All I need to do is work out a way of holding the aperture lever in position.

Thanks mate!

Al.

[sheeny (Al)]

Mind you... while you've probably solved my problem you haven't answered my question...

Al.

[bojan]

I will use neutral lunar filter in front of the sensor, even two if needed (I think they are $15 at Bintel, they have them with 30% and 10% transmission) to reduce the light and to avoid using aperture masks (they will always reduce the resolution)

[[1ponders] (Paul)]

Try setting your aperture while still on the camera and with the camera still turned on remove the lens. It's not recommended but it has worked for me on previous occasions

[sheeny (Al)]

Originally Posted by bojan

I will use neutral lunar filter in front of the sensor, even two if needed (I think they are $15 at Bintel, they have them with 30% and 10% transmission) to reduce the light and to avoid using aperture masks (they will always reduce the resolution)

Yeah I have an 18% ND filter that I can slip in behind the lens. I'm thinking about trying to maximise the aperture (and therefore resolution) now I've worked out how to get the aperture ring of the lens working.

Al.

[sheeny (Al)]

Originally Posted by [1ponders]

Try setting your aperture while still on the camera and with the camera still turned on remove the lens. It's not recommended but it has worked for me on previous occasions

Since the OM-1 is all mechanical except for the light meter, I don't think that will make any difference Paul, but I've had a play and I can get the aperture to stop down but sliding a cable tie over the depth of field preview button on the side of the lens.

Thanks for your help!

Al.

[Terry B]

Originally Posted by sheeny

Mind you... while you've probably solved my problem you haven't answered my question...

Al.

You are correct that resolution=diameter/wavelength and this is the principle used in radio interferometry but the imaging technique in radio is broadly named "synthesis imaging" and uses interference fringing not simple summing of the images.
Having said that I remember reading about using an aperture mask with lots of holes in it on the Keck telescope as an experiment on interferometry. Sadly I don't remember the results
Another point is that the pic you have of your mask still did not have much distance between your holes so this still made resolution worse than no mask.

[bojan]

Also, interferometry with apertures wide apart works OK for simple sources (like one or two close stars for example). It is relatively easy to reconstruct the equivalent image from fringe patterns which are simple for this case... but for more complex sources to resolve, it is necessary to apply math algorithms, which make this task very hard to do.

[sheeny (Al)]

Originally Posted by Terry B

You are correct that resolution=diameter/wavelength and this is the principle used in radio interferometry but the imaging technique in radio is broadly named "synthesis imaging" and uses interference fringing not simple summing of the images.
Having said that I remember reading about using an aperture mask with lots of holes in it on the Keck telescope as an experiment on interferometry. Sadly I don't remember the results
Another point is that the pic you have of your mask still did not have much distance between your holes so this still made resolution worse than no mask.

Thanks Terry.

It's obvious to me from the results of my backyard experiment that the diameter of the holes is dominant over the spacing between them in this set up.

I could've gone to 35mm pitch diameter on the holes but the holes would've been right on the outer edge of the lens which brings any optical errors in the lens into play so I figured 25mm was as wide as I was prepared to go (that should've been more than adequate to beat the pixel resolution of the ToUcam).

I'm not sure of the ins and outs of interference fringing, but another analogy is the large mirrors made these days in segments like the Cangaroo scope at Woomera. If the diameter of the hole (or mirror segment) dominates then large segmented mirror scopes have no advantage other than light gathering. I'm sure they would also have better resolution too...

Al.