January 31, 2010: Long
Time Since Last Update
Last winter I got a new DMK31AU03 planetary camera. I've
taken some pretty good photos of Jupiter, Saturn and Mars.
I also have a few new galaxy photos in the past year, the
best being a M-31. For Christmas this year, Santa got me
a DSI III Pro, but I haven't had a chance to use it yet.
March 23, 2008: Tough Year Weather-Wise
Winter is usually the best time for Astrophotography.
This year wasn't so good. It seemed like it was usually
clear during the day, then cloudy at night. I only got
three decent pictures since the fall, though I was quite
happy with the ones I got.
Last month, I stripped-down my mount, cleaned, and
regreased it. I found one of the bearings rusty due to a
lack of grease, and found a replacement for it. I also
replaced the mount's dovetail bracket (which holds the
telescope) with a sturdier, less wobbly-one. I'm not
really sure if these changes made an improvement in the
tracking, since I've only been able to test once and I'm
not sure I had a good polar alignment.
February 14, 2007: A Brief Lesson in Imaging
The Setup - Initial Setup:
For deep-sky images, long exposures must be taken to
gather a lot of light. Long exposures require very
precise tracking of the telescope to keep the object from
moving around and smearing the image. The first
requirement of good guidance is a good initial setup and
alignment. Care must be taken to have the mount precisely
leveled, balanced (I use forty pounds of counterweights!),
and aligned with the Earth's polar axis. Alignment is
tricky - the north star (Polaris) is actually close to a
degree off the polar axis. My telescope has a polar axis
alignment scope in the middle of the mount (the polar
axis shaft is hollow), with markings for finding the
exact position of the polar axis with respect to Polaris.
But that can only get you to within a few tenths of a
degree and that still isn't good enough. Careful tracking
of the motion of stars (there are several procedures for
this, and I'll spare you the details for now...) allows
for finer alignment. But even with a good alignment,
tracking isn't perfect...
No telescope can track perfectly on its own. Besides
alignment errors, tiny imperfections in the gears -
thousandths of an inch - show up as periodic error
patterns in the images - the image will move back and
forth on the screen in a repeating pattern. For my old
telescope, half of my images of 30 seconds would be
unusable and I couldn't go any longer than that. With my
new telescope, I can keep most of the images up to a
minute exposure. But that still isn't good enough. For
longer exposures, automatic guidance is required. I
recently started autoguiding with my my old-old telescope
as a guide-scope. A camera is connected to my laptop,
which is connected to the mount. The laptop tracks a star
near the object to be photographed and sends commands to
the mount to keep the star centered. With this technique,
exposures of any length can be taken, though things like
wind or vibrations in the mount due to an unstable
location (such as my deck) can still spoil an image.
Acquisition - The Camera:
I primarily use the Meade Deep Sky Imager II Pro. It
connects to my laptop, webcam style, and mounts directly
on the back of the telescope, with no lenses in it and no
eyepiece on the telescope. The DSI II Pro is a monochrome
(black and white) camera. To get color pictures with it,
I shoot separate images through filters for each color (red,
green, and blue) and combine them later in software. If
you see "rgb" under an image on my site, it was
taken with this technique. Additional detail can be
gathered by a black and white image, which pulls in more
light because there are no filters to block some colors.
This is called a luminance (brightness) frame. Images
taken with this technique and combined with rgb photos
later in software are labeled "lrgb" in my
There is noise in all cameras. A ccd camera works by
converting light to electricity, and that generates heat
- which converts more charge to electricity and shows up
in the pictures. This noise is very predictable and is
automatically subtracted out of the initial acquisitions
by taking "dark" frames. Dark frames are
literally what they sound like - the telescope is covered
and images of nothing are taken, equal in length to the
images to be taken of objects later. Since the noise
repeats, it can simply be subtracted out. Unfortunately,
this noise is temperature dependent, so when the outside
temperature changes by more than about 5 degrees, new
dark frames need to be taken.
Pre-Processing - Stacking:
All images have a certain signal-to-noise ratio. The
noise is the grainyness you see in some regular photos,
as well as atmospheric distortions of the light rays (like
the wavyness you see in the distance on the road on a hot
day). In astrophotos, due to longer exposures and higher
magnifications, sources of noise are greatly amplified.
But these sources of noise are largely random. By
stacking images on top of each other (literally what it
sounds like), random noise averages out to nothing, while
the clarity of the object is amplified. For deep-sky
images, 10, 20, even 50 images are stacked. If each sub-exposure
is 2 minutes, for example, and there are 20 of them, you'll
see 20x120s under the photo in the gallery. If it is an
lrgb image, that means I spent a total of 20x2x4=160
minutes acquiring images, for a single resulting image.
That's a long time - and I'll be going even longer for
dimmer objects. For planetary images, detail is
everything and since the exposures are short, they
combine hundreds of individual exposures for one
Post-Processing - Computer Magic:
Even after everything else, there is a lot more
processing to be done. Some of the steps:
--Wavelet Processing - Frankly, I don't
know how it works, but it is a mathematical way of
finding and enhancing detail.
--Stretching - Pretty much what it
sounds like. Colors are assigned a number from 0 to 255,
with zero being black and 255 white. By cutting the top
to 128, for example, and stretching the rest, blacks stay
black, but dim colors are made twice as bright.
--LRGB Combining - Discussed above. It
is how color photos are made with a black and white
camera and filters.
--Color Balancing - What it sounds like.
Sometimes, due to filter transmittance and camera
sensitivity, colors are balanced by varying the exposures
for different colors. Other times, it is done in post
--Background Noise Reduction - Sometimes
the background comes out grey (or even green). The color
spectrum is adjusted to cut it back to black.
Phew. If you got through all that......I won't believe
you - but thanks anyway!
|April 13, 2006 (taken from a post on www.physicsforums.com
): Selecting an Intermediate Level Telescope
a first decent telescope in the $800 range, you have the
Meade ETX, the low end of the LXD75 series, the Orion
Skyview Pro, or the Celestron C -- GT series.
If I had the chance to do it again, I probably would
not have gotten my ETX. I would have either spent a
little more for an LXD series or gotten an Orion for
about the same price. Part of the problem there was that
I already knew I wanted to go the next level up, but didn't
quite have the funds yet, so I wanted something to hold
me over until I could get something better (which I am
about ready to do). I actually bought a $400 Meade
reflector that was junk, decided I was willing to spend $800,
then didn't spend enough effort on the decision.
Some key issues:
Focal Ratio (ratio between focal
length and aperature):
This is probably the biggest issue. It is what determines
whether or not you can do deep-sky imaging with much
success. Lower is better and if you want to do deep-sky,
having a high focal ratio is an absolute killer. At f14,
my ETX requires eight times the exposure of the Orion
Skyview Pro (f4.9 - you ratio the squares) to get the
same brightness. Now mine can work with a focal reducer,
which brings it down to about f7, but that still means
double the exposures. Similarly....
Very close second. My mount has a tracking bug that makes
exposures over 30 seconds impossible (and even at 30
seconds, I can only keep about half). Some people have
gotten better, but the ETX is hit or miss. The Skyview
Pro is better, though even then people say getting 2
minutes unguided is difficult with any scope. Still, that
means you can get 8x brighter images with the Skyview
than I can with my ETX, without having to jump up in
difficulty to autoguiding. There are a lot of deep-sky
objects in range of a scope like the Skyview, not a lot
in range of mine.
Since most scopes in the same product line/family have
the same focal ratio, bigger scopes essentially just
yield the same brightness but a bigger image. For
planetary, though, aperature=resolution.
Real quick, reflectors or CATs (half refractor, half
reflector) are generally the most versatile and therefore
There are three basic options:
For planetary imaging, a decent webcam (Quickcam 3000 or
4000, or anything with a CCD, not a CMOS imaging chip)
will match or exceed the results of a more expensive
camera. Regardless of if you are going to move up to deep-sky
imaging, I highly recommend starting out with a webcam.
People make adapters, but all you really need to do is
remove the lens and hot-glue a 35mm film canister to it.
Then it goes in place of your eyepiece.
By shooting videos and stacking hundreds of images
with software like Registax, you get about double the
resolution from the final pic than in any individual
Low-end deep-sky imagers:
Meade makes a camera called the Deep-Sky Imager ($300)
and Orion makes the Starshoot ($400). They come with
software that takes the pictures and they have image
processing suites with them. They are a good start, but
they are low resolution and low sensitivity. One big
drawback of the color DSI - it is actually a monochrome
camera with a grid of filters on it. That means its
output resolution is really only a quarter of what the
CCD is (and therefore advertised as), and it is
noticeable in some of my photos. I don't think the
Starshoot has that issue.
Midrange CCD Imagers:
The next step up from there is the monochrome DSI II ($700),
or competing products in the up to $1500 range. The
obvious drawback is the need to shoot three(or four) sets
of images to then combine the colors via software.
Benefit: sensitivity and resolution.
On a similar pricerange is a regular digital SLR camera.
Their benefit is bigger chips (for wider fields of view)
and higher resolution. Drawbacks are that they are
unsuitable for planetary imaging and their sensitivity is
lower than the dedicated astrocams at similar prices.
You can also piggyback (means exactly what it sounds
like) a dslr with its lens to take really wide-field
pictures. I coupled an slr lens with my DSI for a few of
|December 3, 2005: Intro/History
figured I'd start off with a little history on me and my
I'm 29 years old and I got my first telescope for
Christmas when I was in high school - I'm not sure what
year. The telescope is a Tasco 60mm refractor on an
equatorial mount. My dad actually got a slightly cheaper
version on an Alt-Az mount, but recognized that an
equatorial was better and we got it exchanged. Though I've
actually heard differing opinions on that, I'm glad we
traded it in: the equatorial mount provides much easier
tracking and finding of objects. Anyway, with that
telesope, I could see the major planets, the moon, and
the sun. It isn't a great scope, but it kept me
interested - and wanting more. And I still have it - I
may yet use it as a guidescope in the future.
A little over a year ago, after reading about
techniques and software using a webcam for
astrophotography on the web, I tried it for myself. It's
astonishingly easy: take a webcam, disassemble it to
remove the lens, and hot-glue a 35mm film canister to it.
It fits perfectly into the eyepiece holder of a telescope.
The first pictures I took were terrible. But it re-lit
the fire, and before I even tried to do better with that
old refractor, I bought a new scope...
I bought a Meade DS2114 4" Newtonian reflector,
with computer control for $400. I should have realized
that it was too good to be true - to get that price, they
had to cut corners everywhere. So I sent
it back and bought an ETX-105 Maksutov-Cassegrain for $800.
It's a decent intermediate scope - though not without
limitations, which I will get into more later. For about
the past year, I've gotten about all I can out of this
telescope. I've used it dozens of times - occasionally
driving to the Poconos to set up in a nice dark, random
parking lot. Mostly, I use it at a local golf course. I've
even set it up for parties - with its guidance, I can
point it at something and let it go for a long time. My
friends get annoyed, though, if I leave a poker game to
check on it ;)
The camera I'm using right now is a Meade Deep-Sky
Imager ($300). It's not bad, but it also has limitations,
which I'll also go into later. I started with a webcam,
but webcams don't do long exposure, so they are pretty
much limited to planetary photography. Unless, that is,
you can modify them to take long exposures... People have
figured out how to do just that, and instructions and
schematics are available online. The problm is, modifying
them requires lifting and soldering pins on the circuit
board. At maybe a milimeter thick, with a milimeter
spacing, it's tough work for someone who has little
experience with such things: I destroyed two webcams
before deciding to buy the DSI.
Because of the limitations of my scope, I'm starting
to contemplate an upgrade. However, I've promised myself
I won't buy a new telescope until I buy myself a house.
So now I'm starting to look for both ;)