Difference in horizontal and vertical resolution: 20D, Canon zoom, Celestron C90

In my recent quest to get a decent picture of the moon with a 25 year
old Celestron C90 and a Canon 20D (I've given up, BTW), I decided to
see if the Celestron was really as bad as my lunar images implied. I
did a quick and dirty resolution comparison with a Canon 75-300 zoom
using a resolution chart I downloaded from the Web. The Celestron is
(no big surprise) worse, even though this particular Canon lens has the
reputation of being soft at full zoom. However, in examining the
results I came across something I didn't quite understand, and I
thought you folks might be able to give me some insight.

On vertical lines the Celestron was not dramatically inferior to the
Canon lens, though the contrast was substantially lower. On horizontal
lines, though, the Celestron image was simply awful. See

http://home.comcast.net/~jgates777/Canon_Celestron_Comparison.jpg

for a comparison of the images.

If anything, the Canon lens does slightly better on horizontal lines
than on vertical ones. The Celestron, on the other hand, is just awful
with horizontal lines. I repeated the comparison several times with
the same result. All tests done with lenses/camera mounted on a Bogen
3046 tripod with a 488RC2 Midi Ball Head, mirror locked up, remote
release after 10 seconds of MLU. I moved the tripod/camera so that the
image was about the same size for the two lenses (i.e., the Canon was a
lot closer to the target than the Celestron, which is about 1200 mm in
focal length). The only think I can think of is that I'm getting
vertical "nodding" with the Celestron. Any thoughts?

 

$On vertical lines the Celestron was not dramatically inferior to the
$Canon lens, though the contrast was substantially lower. On horizontal
$lines, though, the Celestron image was simply awful. See

How were these lines arranged within the frame? It's normal for a
lens to have different resolving power for radial lines than for
tangential lines, and in some cases, in some parts of the frame,
there can be a very large difference (which is why MTF graphs have
pairs of lines).

So if you arranged it so that the horizontal lines were (say)
immediately to the side of the middle of the frame, and the vertical
lines were (say) immediately above or below the middle of the frame,
and the same distance from the middle, you should see about the same
results for both. Put them next to each other at one of the edges
of the frame, and it wouldn't be unexpected to see one come out
sharper than the other, even though they're right next to each other,
because one of them will be radial and the other tangential.[/QUOTE]

 

The chart I used was the EIA Chart 1956. This is, I think, for video,
but for my purposes it didn't matter. The images I referenced were
form the vertical stripes above the center "striped square" and the
ones to the left of the center "striped square". The center of the
viewfinder image was the "bullseye" in the cinet of the chart. I also
checked the converging lines to the bottom and left of the "striped
square", and the same disparity was apparent: the hoirizontal lines
were very poorly imaged by the Celestron/20D combination.

The chart is at

http://homepage.mac.com/stevebagley/EIA1956wide.pdf

 

Lenses have radial symmetry and so can have different microcontrast in
meridional and sagittal directions. This is why mtf plots have M and S
lines.

Computer simulation suggests the average of the MTF M and S values is a good
indication of sharpness for a typical view with detail that is randomly
orientated.

 

You are comparing a 300mm lens with a 1300mm telescope.

Might be that your scope is out of alignment. Or might be that
your particular scope isn't one of the better ones.

Keep in mind that a 1200mm Canon lens runs about $12,000 while
the Celestron can be had for a couple of hundred. The tradeoffs
mean that the telescope isn't as good a camera lens. In particular
the secondary mirror results in lower contrast and the complexity
of the large aspherical front element means plenty of opportunity for
error.

 

Try re-orienting the chart and camera 90 degrees to see if you have vertical
shake. I've noticed my moon shots suffer from vertical blur but no horizontal
blur... and since eye-piece views are perfect I know its the tripod/mount.

 

C90s are generally...crap. Worthless optical corrections and
poor contrast. Try a Meade ETX or even one of the Orion Chinese
Maks, the bigger, the better, or an inexpensive 4-5" Chinese
refractor.
I used an ETX-105 to shoot Jupiter and although resolution
was poor due to the seeing at my site, the contrast and colour
saturation were excellent.
-Rich

 

Bob's clue did it. I decided to try really tightening up on the ball
head and weighting the tripod. The Celestron is now as good as the
Canon 75-300 wide open @ 300mm, and the horizontal and vertical
resolution are now mediocre (lack of contrast) but equal. I was
counting on MLU to solve the shake problem, but these pictures were
indoors under incandescent lighting at 1.3 seconds. I assume the
shutter trip was causing the shake.

Thanks for the tip, Bob. I guess I should have tightened and weighted
all along! Sorry to bother y'all.

 

OK . . . that did it. I decided to try really tightening up on the
ball head and weighting the tripod. The Celestron is now as good as
the Canon 75-300 wide open @ 300mm, and the horizontal and vertical
resolution are now mediocre (lack of contrast) but equal. I was
counting on MLU to solve the shake problem, but these pictures were
indoors under incandescent lighting at 1.3 seconds. I assume the
shutter trip was causing the shake.

Thanks for the tip. I guess I should have tightened and weighted all
along.

 

Well, yes, I can get much better images by spending several hundred (or
several thousand) dollars. That wasn't really the point, though. I
already **have** the C90, and I was just trying to get a decent moon
photo with it. And yes, I've seen some very nice lunar images taken
with smallish Meade scopes, and others. Roger N. Clark posted a link
to one a while back taken with a Canon 500mm "L" lens and a couple
of teleconverters that was also very nice. I guess more money would
help . . .

I was trying for something like this taken (I think) afocally with a
(presumably newer than my) C90:

http://forums.dpreview.com/forums/read.asp?forum=1010&message=14611149

Perhaps I'll try again when the moon reappears . . . this time with
a tightened up tripod and some weight.

 

$Keep in mind that a 1200mm Canon lens runs about $12,000 while

Actually, it's a heck of a lot more than that. The 1200mm f/5.6L USM
(Canon's only 1200mm lens) is a special-order item and a price isn't
generally advertised, but from what I've read, it's more like USD80k.

If that's a bit rich, you could get a used 600mm f/4L USM and a
2x teleconverter for less than that $12k figure [/QUOTE]

 

I just bought a big heavy tripod but it's still not enough... I need to add some
kind of damping to the whole affair... especialy in the wind!