2004.11.19 - Preparing for first light
I've always wanted to do some serious deep-sky astrophotography, but up until now it has been financially out of my reach. It is a truly expensive hobby. Between the telescope, camera, and accessories, one could easily spend several thousand dollars just to get a basic setup. So my desire to experiment with this hobby had been strictly back-burner for a long time.
Two years ago I bought my girlfriend a 130mm Newtonian telescope for her birthday, but it lacked a motorized mount and therefore couldn't really be used for photography of anything except maybe the moon. Then, about a year ago, I bought a Canon EOS 300-D digital SLR camera that was capable of doing reasonable astrophotography. I was able to do some simple wide-field tripod long exposure photography, but the options here were pretty limited.
Now, enter my friend Bob. He recently bought a behemoth of a telescope - a 14" diameter Meade LX200GPS-SMT Schmidt-Cassegrain Telescope. It is an amazing telescope. He is also interested in astrophotography, but lacks a camera! It took about two minutes for us to realize that by combining our gear we could do some serious photography without egregious additional expense.
We're both new to this, so we're going to start out simple (and cheap) by just doing wide-field piggyback camera shots. We'll mount my camera and lens to the side of his telescope and do manual guiding corrections through the main scope.
Bob had to buy a camera mount for the optical tube and an illuminated reticle eyepiece for guiding before we could take our first stab at it. These things are now in transit. The camera mount that Bob bought does not have a ballhead - it is a fixed-orientation affair and if it is too close to the body of the optical tube, the 28mm wide-angle lens I have (which has a 46.5 x 30.9 degree field of view) will image a big chunk of the tube instead of just sky.
There are several solutions to this problem. One is to get a longer lens, which I went ahead and did, just to get another option for frame composition (it's a 50mm f/1.4 prime with a field of view of 26.0 x 17.3 degrees). However, it would be nice to be able to also use the 28mm lens, so if its FOV intersects the tube we'll have to either build a stand-off for it or I will buy a ballhead which will give us enormous added flexibility in frame composition. Eventually I'll buy the ballhead anyway because once we get our technique down we will want that flexibility. For starters, though, we're trying to do as much as we can without spending much money.
Here is the equipment setup we plan to use for our first attempt, which will happen as soon as all of the new gear arrives and we have a clear night (which lately have been strangely infrequent):
Telescope:
Camera:
There is some chance that even the 50mm lens will be too wide for the camera mount. It might be possible to mount the camera so that it is not pointing down the tube, but out over the fork, and do some weird setup like that. Otherwise, there will be a delay in getting actual exposures made while the ballhead is ordered and shipped. Once the lens and mount arrive (they have both already shipped), we'll do some daylight test exposures. I'm going to bring my 28-105mm (optical quality: crap) lens over and mount it to see at what focal length the FOV is clear of the optical tube.
The Meade #07591 piggyback camera mount arrived at Bob's place tonight. I was very unimpressed with this camera mount. It does not elevate the camera off of the OTA more than about 1.25" and it has no mechanism for tilt, which means that wide-field photography isn't really an option using only this mount. It is so close to the OTA that any lens whose largest diameter is greater than 2.5" + the diameter of the lens mount on ther camera cannot be attached. So this mount precludes the use of very short lenses and very large lenses as well. In addition, the attachment knob is difficult to grip and in such a cramped position that it was basically impossible to get a really good tight attachment to the camera without using pliers. Bad design.
We did some testing with various lenses and orientations to see what our options were. For starters, we wanted to see at what focal length the OTA was no longer visible in the field.
|
|
28mm lens mounted directly to
the Meade piggyback mount. Considerable OTA intrusion. |
28-105mm lens set at 50mm mounted
directly to the Meade piggyback mount. OTA still significant. |
By zooming in with the 28-105mm lens, we were able to determine that the OTA is no longer visible at a focal length of about 70mm. This is unfortunate, given that I just bought a 50mm f/1.4 lens. However, we were able to pan the camera side to side in the mount and tighten it down at some off-axis angle such that the OTA was not visible. For the 28mm lens, this meant an angle of about 45 degrees. For longer focal lengths, this angle would decrease.
The obvious fix for this lackluster mount was to buy a ballhead mount to place on top of the piggyback mount. I went ahead and ordered an FLM Centerball 38 FPR ballhead which is overkill in terms of load capacity, but this will probably prevent a certain amount of flexure during long exposures. The ballhead gives us complete freedom in terms of frame composition and raises the camera 106mm (4.2") off the piggyback mount, which should allow for the use of any lens.
 |
The Canon EOS 300-D on the piggyback
camera mount. The inset picture shows a better view of the mount itself.
|
The next order of business was to set up computer control for the telescope. Bob had recently acquired a reasonable laptop for just this purpose and loaded it with the Meade control software. Unfortunately, the cable connecting the 6-pin connector on the telescope to the DB-9 serial port on the laptop had gone missing, so we frobbed one together out of an old serial cable and my RJ-45 crimper. It worked well and allowed us to replicate the functionality of the hand controller from the laptop, as well as doing direct GOTO control from a graphical star chart.
Despite everything the NOAA had said about tonight's weather, the sky was totally clear except for some thin clouds in the SE. Given that we could get a clear view with the 28mm lens if we were willing to shoot 45 degrees off-axis, we decided to drive the scope up into the mountains and take a first stab at getting everything working.
The site we choose was a parking lot just off of NM 4 at the intersection with Forest Road 289. The coordinates of our setup were 35.8481°N, 106.4243°W (WGS84/NAD83) and the altitude was 8980'. This was a nice location because it is only five miles away from Bob's house and offers over 1,700' of elevation gain. The fact that it is right next to highway 4 is a downside as cars were going by every few minutes the entire time we were there. The sky view isn't perfect; there are small mountains to the west and north, but it was pretty good.
Neither of us had used this telescope in a polar-aligned configuration, so we were figuring out how to set this up as we went along. Some manual-reading was definitely in order. There were issues getting the tripod level, getting the fork pointed north (I was unsure of the exact magnetic declination at our observation site), and aligning the scope with Polaris. In the end, we settled for a scope that was sort of level and sort of aligned. It was tracking well enough for visual work and the GOTO was getting us to well within the spotting scope FOV every time. It was not as accurate as the Alt-Az mounting that we set up last week, so we need to work on this.
 |
Bob using the laptop to control the
Meade 14" LX200GPS-SMT
|
Unfortunately, while we worked all of the problems associated with the polar alignment, the clouds in the SE moved in and covered the bulk of the sky except for a small clear window to the NW. Before the clouds covered the sky completely, which seemed imminent, I set up the camera on the mount. In doing so, I discovered that my cable release was missing from my bag. This meant that the only way to do bulb exposures was to hold down the shutter release button on the camera and this was unacceptable. Instead, we were limited to the maximum shutter time of the camera, 30 seconds. I tilted the camera off-axis and then we turned the telescope until the camera was pointing towards the clear part of the sky. We took a series of four consecutive 30 second exposures at 28mm, f/2.0, ISO 100. I should have bumped the ISO up to 400 but forgot. I also neglected to set the data format to RAW instead of JPEG. Additionally, I failed to fully get the OTA out of frame. One last error... the autofocus didn't work because of the cloud cover, though looking through the viewfinder I couldn't tell.
So we ended up with a series of dark, out-of-focus, compressed images with a corner blocked by the OTA. Basically I screwed up a bunch of stuff with these shots. However, since the reticle eyepiece hasn't arrived yet anyway and we therefore were not able to do any guiding, we weren't expecting anything amazing. A processed composite of the four images is below:
 |
Canon EOS 300-D, 28mm lens at f/2.0. Composite
of four 30s exposures. Levels and unsharp mask in Photoshop CS.
View is NW from observation site. Bright star in left-center is Deneb.
|
As you can see, The clouds were sweeping in from left to right. The reduction from 3072x2048 pixels to 500x333 pixels basically eliminates problems introduced by the slight lack of focus. The levels adjustment in Photoshop corrects for the darkness and the compositing removes the dark current noise, so this small image is representative of what we could have gotten if I hadn't made so many mistakes.
A couple of interesting notes about post-processing these images. First, I took a 30 second dark frame shot after the four regular shots to subtract away the noise. However, after compositing the four regular frames, subtracting the dark frame only added to the noise. This could be because there was JPEG compression artifacts, or because the dark frame subtraction is unnecessary when the noise-reduction is on in the 300-D, or because I should have dark-frame subtracted from each individual frame before compositing. I'm note sure yet.
Second, There was an odd feature in the corners where I noticed that the aberrations around each star were always aligned radially:
 |
Full-resolution segments from near the
corners of the previous image showing radially-aligned aberration.
|
I'm not sure if this is an effect of the lens not being focused properly to infinity, or sub-optimal optics in the lens, or something else. Probably the former.
After the last frame was shot with the camera, we shut down and disassembled the telescope because the cold was getting to us. There are a lot of things about our setup and routine that need improvement, but I think this outing demonstrated what most of those are.