Unexpected results with larger format resolution targets (1 Viewer)

[4th_point]

I'm wondering if anyone has input on what I observed with larger format targets.

I've been interested in comparing the resolution of different optics at hundreds of feet rather than hundreds of inches. I do like close range viewing in the yard, but tend to use optics at longer distances in the field. Anyway, I assumed that larger format targets might be more useful than something like the 1951 USAF chart even though it has a wide range of element sizes. So I made my own simple targets with alternating black and white lines of uniform density that fill an 8.5"x11" sheet of paper, which seems to work fine. They are loosely based on the diffraction limits of the instruments.

What I didn't expect is that the resolution values that I am estimating are 110/D to 115/D which seems unrealistic. Even if I factor in some error in my measurement for lp/mm, I still get high resolution values.

These are good scopes, but I wonder if large format targets are not compatible with resolution estimates from the 1951 USAF chart?

It doesn't make sense to me. Unless the larger target provides more input to the instrument, eye, and brain?

 

[henry link]

It would help to know a bit more about your chart and your methods. Could you post a photo of the chart and tell us how you converted line pairs to arc seconds? What magnifications were used and at what distance?

 

[4th_point]

Hi Henry,

Thanks for the reply.

I don't have a picture of the targets handy, but they are just bar targets like a Ronchi ruling. And before I forget, I should state that I consider a target visible or resolved if I can tell the direction of the lines over the majority of the sheet. Orientation is random and unknown to the observer.

I convert line pairs to arcseconds by measuring the number of line pairs over a distance. For example, one target has 36 line pairs over one inch. I consider a line pair to be one black line and one white line. So each pair has a fixed width of 0.028". Or, 0.027 minute or 1.59 arcseconds at 100 yards.

I have been experimenting to 100+ yards using up to 75x. I have several targets, of different lp/inch, but have simply been choosing one that seems most appropriate for the optic and distance. I start at a distance where I think that the target should not be resolved and moved closer until I can detect the direction of the bars.

Any comments or suggestions would be appreciated. I have learned a lot from your posts.

Thanks,

Jason

 

[henry link]

Hi Jason,

Your experience reminded me of comparisons I've done between the USAF 1951 and the NBS 1963A seen here:

NBS 1963A Resolution Target | Edmund Optics

NBS 1963A Resolution Target ideal for precision optical testing is available in multiple sizes with either positive or negative patterns at Edmund Optics.

www.edmundoptics.com

The NBS 1963A uses the same black and white line pairs as the USAF 1951 but with much longer lines in groups of five black lines instead of three. Using the two resolution charts as eye charts just now I found that resolving the line pairs was much easier for me on the NBS 1963 than on the USAF 1951 even when the line pair widths were identical. Perhaps it's just the large number of lines and the long line lengths on your target that make it easier for the eye to perceive the line directions compared to the same width line pairs on the USAF 1951.

Henry

 

[Binastro]

Absolutely longer lines are easier to see than shorter lines.

I don't know about 5 lines compared to 3 lines, but 5 dots in a line are easier to see than 3 dots.

5 subvisual stars may be seen where 3 equal subvisual stars are not.

Regards,
B.

 

[4th_point]

Hi Jason,

Your experience reminded me of comparisons I've done between the USAF 1951 and the NBS 1963A seen here:

NBS 1963A Resolution Target | Edmund Optics

NBS 1963A Resolution Target ideal for precision optical testing is available in multiple sizes with either positive or negative patterns at Edmund Optics.

www.edmundoptics.com

The NBS 1963A uses the same black and white line pairs as the USAF 1951 but with much longer lines in groups of five black lines instead of three. Using the two resolution charts as eye charts just now I found that resolving the line pairs was much easier for me on the NBS 1963 than on the USAF 1951 even when the line pair widths were identical. Perhaps it's just the large number of lines and the long line lengths on your target that make it easier for the eye to perceive the line directions compared to the same width line pairs on the USAF 1951.

Henry

Thanks for the information, Henry. That helps me a lot, and I appreciate that you took the time to check.

Jason

 

[4th_point]

Absolutely longer lines are easier to see than shorter lines.

I don't know about 5 lines compared to 3 lines, but 5 dots in a line are easier to see than 3 dots.

5 subvisual stars may be seen where 3 equal subvisual stars are not.

Regards,
B.

It definitely makes sense now.. I think that I had an overly simplistic understanding of resolution, where the results were more dependent on the line spacing.

 

[Binastro]

Hi 4th point.

A skilled observer should trust his or her observations, and not base the results on conjecture.

Barnard, a very skilled observer with exceptional vision could see wires against the sky 0.57 arcseconds thick with unaided eyes.
I'll check the actual figure.

Then there are long lines either straight or curved.
I have a thin line drawing of a smiley face that I can clearly see at 4 metres and probably more.

Then there are black dots against a white background and white dots against a black background.
Coloured dots, grey dots.

Black lines.
Line pairs.
Printed charts.
Transparency charts.

Blue light gives better resolution than red light.

However, for some reason infra red lasers are more efficient the longer the wavelength perhaps to the fifth power, but I just can't remember.

The line charts are all different to star images and resolution of double stars of equal or unequal brightness.

The 1951 US chart must be an attempt to give results similar to double star separation.

Then there are subvisual observations, where two or more subvisual markings or stars are combined to give a visual image.

I found that the separation of subvisual stars or dots can combine at up to 4 or 5 times the size of the required resolution, but preferably 2 or 3 times.
Beyond 5 times I found no real gain.

I wrote an initial paper in the 1980s on stars, then later on sunpots with safe high optical quality filters, but I don't have copies of these.

Then one is testing ones eyesight as well as the optics.

As to double star separation, artificial double stars in daylight seem to give figures 2/3 of night time double separation.

The unaided eye double epsilon Lyrae has a separation of 3.5 arc minutes.
I could rather easily separate them from town when young, but many can't.

So we have resolution results that vary by more than 200 times.

If everyone uses the 1951 US chart in the same lighting and especially testing two telescopes side by side, that is a reasonable test.

But to then talk about the Dawes limit is gobbledegook.
Dawes only talked of 6th magnitude equal white stars using fine 6 inch refractors. Both by him and his experienced friends.
He nowhere talked of 1951 US charts.

What would be interesting is the difference in percentage terms of your chart and the 1951 US chart side by side.

Regards,
B.

 

[jring]

Hi,

on a lighter note but not completely off topic here... did you know this one?

As some quite knowledgeable chap over on cloudynights remarked about a 2m surplus mirror blank bought by for spare change another member who planned to grind it and make a giant dobsonian... this blank was not made for looking up... no further comment...

Joachim

 

[Binastro]

Many observers can resolve a black line on a white background at a width of 1.0 arcsecond without optical aid, except possibly normal glasses.

Strangely, a white line on a black background is more easily seen, maybe because of irradiation.

Barnard could see a wire against the sky where the wire was 0.44 arcsecond wide without optical aid.

I must have seen 0.57 arcsecond for someone else.

Black spots against a white background are commonly seen at 34 arcsecond diameter, which is similar to my result with unaided eyes or glasses.

Regards,
B

 

[Binastro]

Thanks for the Youtube video.

It seems a very coarse test.

The SR71 could photograph golf balls from 85,000ft maybe obliquely also.

The U2 apparently could not fit the 240 inch focal length camera as the fuselage was not quite wide enough.
It used the 180 inch version.

However, the Canberra could fit the 240 inch lens for oblique photos.
In early days from above with smaller lenses.
The 60 inch Ross lens was common. I have the 50 inch f/8 Ross.

The Japanese used a 100 inch lens and a mirror.

The US 144 inch f/8 was bought ex gov and was used to photograph asteroids to 1/50 arcsecond accuracy.

The 70 inch diameter mirrors resolved about 4 inches with adaptive optics from 250 miles.

I think there were larger versions made also.

Aerial platforms could lock on car door handles at about 90km.

Regards,
B.

 

[jring]

Hi binastro,

I think that USAF 1951 resolution target was not for the blackbird or the U2 with their telephoto lenses but for those spacecraft which took their job and used mirror telescopes from a bit higher up... probably not fine enough for the latest generations but those in the know about their resolution could tell us but would have to shoot us afterwards...

As for the 144" f/8 - you mean 144" aperture, which would be truly enormous... do you have more info on that one? What type, probably some cassegrain variant?

Joachim

 

[Binastro]

Hi Joachim,

The 144 inch f/8 was from memory a refracting lens of 19 inch aperture that ran along the fuselage and had a mirror to direct the view.

I think it took 28 inch square negatives.

Perhaps designed by Baker and very fine quality.

The later use was for photos of asteroids and by measuring offsets from stars the position could be determined to 1/50 arcsecond.

There was a US 34 inch aperture lens, probably folded optics that showed up in Finland but disappeared.

The B47 aircraft relied on speed to take photos as it did not fly very high, usually around 35,000ft. In fact most jet fighters couldn't catch it.
The B36 was slower, enormous and had ten engines, including four jet engines. It flew at a similar height.

I presume the 144inch f/8 was used in either these aircraft or perhaps somewhat smaller aircraft, US Tornado??

I used to see these aircraft regularly in my 3 inch refractor.

The British Victor flew much higher, about 50,000 or 55,000ft, although I recorded one at 63,000ft.
The photo versions had multiple large cameras including those with the Wray 36inch f/4 night lens. I had about 4 or 5 of these lenses costing £10 to £60 ex gov.
These lenses were used with very bright flash cartridges that dropped down to illuminate the target.

I also used to see the Vulcan and perhaps the Valiant.
All were painted bright white.

I would think the USAF test chart in the video was something that a K24 camera with 7 inch Aero Ektar or the larger camera with 12 inch Aero Ektar could use.
Even a 35mm camera could I think use it.

There were no satellites in 1951, although rockets were going up to about 100 miles.

Regards,
B.

 

[jring]

Hi Binastro,

thanks for the info on the aerial imaging stuff and the 144" focal length long lens... most interesting.

As for no satellites in 1951- certainly not, but 1951 was the time when USAF introduced the pattern... not sure when this one was painted... but the KH-11 series started in 1976 with a 2.4 mirror like Hubble (but probably no production QA problems) and has a theoretical resolution of 0.05 arcsec or 6" at 250km. In reality probably double that due to aberrations and seeing. So probably still too fine for that pattern... but the 60s models were mostly in the 0.5 to 0.6 resolution range and for this it would make sense... some of those had long lenses with flat mirrors and others mirror telescopes.

Joachim

 

[Binastro]

It would be strange if image stacking wasn't used.

Horace Dall and others were doing this a long time ago.

In 1951 the highest aircraft were flying at 45,000ft to 48,000ft for the Canberra.
Not much higher than Spitfires or Arados.
Lenses used were normally about 5 inch, 8 inch, 14 inch, 20 inch, 24 inch, 30 inch, 36 inch, 40 inch, 48 inch, 60 inch and a few 100 inch focal lengths.

Regards,
B.

 

[4th_point]

Hi,

on a lighter note but not completely off topic here... did you know this one?

As some quite knowledgeable chap over on cloudynights remarked about a 2m surplus mirror blank bought by for spare change another member who planned to grind it and make a giant dobsonian... this blank was not made for looking up... no further comment...

Joachim

Thanks for the video, Joachim. I didn't know about it.

Jason