Complete an Astrophotography Image Using New Telescope

Just a quick post-mortem to go along with my last post: I am so glad I decided to participate in this event. My project ended up being kind of strange because I had pretty much completed it day 3. Once I had the bugs worked out with the tracking mount, I didn’t need any motivation doing the rest because I was running on pure enthusiasm. That said, I can honestly say that without sitting down and writing these goals, all my equipment would still be sitting unused in my room here, which would be a shame, especially since I had saved up so long for it all. Anxiety can just be paralyzing. But I saw that two people had followed my project and I was determined to show them cool space stuff, and that was enough to finally get started.

I’ve also improved greatly on things from the beginning of the month. I had posted early on that I was having issues with polar alignment, but now it’s a breeze and barely takes any time. I’ve got focusing and balancing the telescope down to a science too. It turns out you get better at things when you practice them. Who knew?

Shout out to everyone on the discord for being such a delight, it has been great reading everyone’s updates. So many amazing projects. Hanging out in the project chat this month has been a joy.

Well, I’ve been done with this image for a couple days now, but I’ve put off writing an update until basically the last possible moment. It’s just like college!

This is Messier 101, the Pinwheel Galaxy. It is a spiral galaxy not quite twice the size of our Milky Way and contains a trillion stars. At roughly twenty million light-years away, it is furthest object I’ve ever photographed by roughly twenty million light-years (as the joke goes, what’s the difference between a millionaire and a billionaire? About a billion dollars). This image is a stack of 250 thirty second exposures for a total integration time of two hours five minutes.

This picture might not be quite as visually impressive as the previous image of M42, but as I said, the Orion Nebula is kind of like “easy mode”. It is four times larger in appearance and forty times brighter than the Pinwheel Galaxy. I’m still very happy with how this image turned out. I tried several things for the first time while working on it, including taking pictures over multiple nights, and I learned quite a lot.

One of the biggest issues I had taking pictures of M101 was finding the damn thing. This was the first object I’ve imaged that wasn’t naked-eye visible, or at least visible through the viewfinder of my camera. If you look at the exposure I tacked on to the end of the previous post, you can just barely tell there’s something there, and that’s a 30 second exposure. It took me 90 minutes to find the galaxy for the first time. I was trying to use fancy techniques like using “coordinates” and “computers” and “math” before finally going back to the tried and true method of star-hopping, which is where you follow a trail of brighter stars to the object you’re looking for. This only requires a star map. If you’re curious, to find M101 you just hop four stars away from the second star in the handle of the Big Dipper. On the second night, using this hard-earned knowledge, I was able to have the galaxy center frame in about five minutes.

My intention for these nights was to use my computer control the camera to automatically take pictures, however despite my daytime tests beforehand, I couldn’t get it to work. It might have something to do with the DSLR I’m using being state-of-the-art in 2007. The computer refused to take exposures longer than five seconds, which is way too short. This left me once again sitting in the cold clicking the shutter remote by hand. This also limited the amount of pictures I could take as I was relying on the camera’s storage instead of my computer’s. This restricted me at about an hour of exposures per night, when this image really could benefit from much much more. The second night I brought my computer out with me to watch Mostly Walking as I clicked the shutter, which helped pass the time.

I also found the tilt that had been in my system during the first night. Tilt is an aberration in an optical system that occurs when elements are, well, tilted, or otherwise off-alignment. I noticed in my M42 pictures that the stars on the left side of the frame we way streakier than the right side (you probably didn’t notice in my previous picture because of the strategic square crop). Having it only on one side instead of symmetrical around the center screams “tilt”. I noticed that that there was some play in the connector between the camera and telescope. I turned the camera around so the weight of it pulled it in the other direction and voila, it held itself in position with no tilt. This was pretty relieving because tilt can be extremely difficult to fix in some cases.

My goal for week 4 was to take it slow and really focus on get the processing as good as I could, and that is what I did. Have you ever said a word so many times in a row that it stopped sounding like a real word? That was how I felt from staring at this galaxy after a few days. I would be messing with it in Photoshop for a couple hours and get it to a point where I felt happy. Then I would compare it to my ten-minute-first-try (which came out pretty well) and realize that it looked terrible and start over from scratch. This happened seven or eight times. So yeah, I still have a lot to learn when it comes to processing. I should mention that when I say I’m “messing with it in Photoshop,” I’m not drawing or painting anything, or really changing the data at all. Everything you see in the image is what the camera detected, I’m just working to better show the details of the object while keeping the background nice and dark and the stars bright. There is some subjectivity, making it “”“look good”"", which is why I put this project in #Visual Art as opposed to #Other or something else, but the data is the data. I’m just moving sliders around.

As far as improvements for next time go, I sound like a broken record but the biggest one is getting more exposure time. I’d love to get more details and definition in the arms of the galaxy, and more exposures is how I’m going to get there. Another thing I can do is actually following through on my week 1 goal of learning how to take “flat frames”. These help decrease vignetting and make the background more even, as opposed to darker around the edges, which this image suffered from a bit. I probably could have taken the pictures on darker nights. As you can see from the picture in my last post, the moon was getting pretty full as I was taking these, and it makes a pretty huge difference. The clouds in week 2 kind of forced my hand on that front though. I’m excited to get outside again as the moon wanes and hopefully as it gets a bit warmer.

Here’s a quick picture of the moon I took last night just because I love you all. I happened to be pointing my camera at that side of the sky as I was focusing and couldn’t help myself and snapped a few pictures. I love looking at the moon, and you should all stop and take a look at it some time too. Even just a small pair of binoculars will give you a great view. Don’t take our closest neighbor for granted! No stacking for this and minimal processing; just a quick contrast adjustment and crop.

I’m well overdue for an update. Here’s actual footage of me for the past two weeks:

It has been completely overcast for two weeks straight. If you ever buy a new telescope, it will be cloudy for weeks before you are able to use it. It is called the Cloud Tax and it’s Real. I was able to get outside two nights ago, and hopefully I’ll have something to show from that soon. In the meantime, I said I’d talk more about my last image and then didn’t.

I’ve mentioned trying to get longer exposure times a lot, but the last image of M42 wasn’t a single long exposure; it is 48 minutes worth of 50-second exposures stacked on top of each other. This effectively sums the exposure times, making it akin to a single 48 minute exposure. This is almost five times the amount of “total integration time” as any of my previous photos. More integration time means more details and less noise (or, increases the “signal-to-noise ratio” if we want to be a little more technical). I would have liked to get a bit more time with it, but Orion is setting earlier as we get into spring, so it dropped below the trees pretty quickly. I had said that M42 is good for benchmarking, so lets compare to some photos I’ve taken of it in the past. I took this one a couple years ago. It is a stack of 300 two-second exposures for a total integration time of ten minutes. This was with just a regular telephoto lens, as opposed to a telescope proper. That lens had a three inch diameter objective lens, as opposed to my telescope’s six inch mirror, and since we know that area of a circle goes as radius squared, that means that the new image is from 20 times the amount of light as this one (5x exposure time, 4x area). Size doesn’t not matter.

The stars in this image look oblong even at just 2 second exposures at a lower magnification. The tracking makes a huge difference, which is why I have been talking about it non-stop every post. Orion is right on the celestial equator, where the stars appear to move the fastest; the metaphorical edge of the record moving quicker than the center, even though the angular velocity is the same. Also, notice how all the stars seem to have a purplish halo? This is chromatic aberration. Different wavelengths (ie. colors) of light are bent different amounts when they change mediums, like going from air into the glass of a lens. This difference in refraction based on wavelength is called “dispersion” and it’s how Pink Floyd prisms work: white light goes in and each color is bent a slightly different amount, and they separate as they pass through. Due to that, all the light from stars aren’t going to quite line up as a point when they reach the camera detector because different colors are going to be bent differently by the camera lens. In this case, the purples are more out of focus than the rest of the colors, creating a ring. There are ways to get around this problem with clever use of lenses made from specific types of glass, but I sidestep the issue by using a reflector telescope. There’s no dispersion with mirrors. That said, reflectors have their own separate issues because everything in optics is a trade-off of some kind. This image I captured a few months before the last one and is actually the first astrophotography image I took of anything besides the moon. At only 75 seconds integration time, you can see there is a lot less detail in the nebula, even compared to the ten minute image. What’s much more striking to me though is the noise: the graininess of everything. “Noise” is pixels in the camera’s detector saying “hey there’s light here” when there isn’t. This happens for a number of reasons touching on the nature of electronics to just a smidge of quantum theory (which I’m not gonna go near). The more “signal” you have (the pixels detecting light that is actually there, aka your data), the more you can drown out that noise. Also, wow, those stars look terrible. What even is a circle.

This turned more into a physics lesson that I perhaps intended, but hopefully someone finds it enjoyable (and that it all makes sense). I am just overwhelmed with how well this image came out. I keep opening the file on my computer just to look at it some more. Here’s a sneak peek at what I was doing two nights ago: just a single 30-second exposure. I’m hoping to have everything processed and a finished product to show off in the next week.

Last night was beautifully clear, so I took everything outside and took some pictures. So without burying the lede, here’s what I ended up with: This is the Orion Nebula, M42. I kind of look at it as baby’s first deep sky object. It is big, and bright, and very easily found. In fact, if you’ve ever looked up at night during the winter and found Orion, you’ve probably seen this object without knowing it. It sits right in the middle of his sword, just below the belt. All of these points make it a great benchmark for photography. This isn’t my first time photographing M42, and comparing to my previous attempts, this one is way way way way way better. I’m currently exhausted because I was up until 3 am last night processing the image (which I then trashed and started over from scratch this afternoon because my first try looked like garbage because I was doing it at 3 am), so I’m going to leave it here and perhaps tack on more notes later. I was too excited to not post anything.

Well, the weather last night wasn’t perfect, with scattered clouds and gusts of wind, but it was good enough for my purposes of a little practice and experimentation. My main goal was to test the limits of the tracking and to see what hiccups were encountered along the way. Also, I updated my weekly milestones, just to be a little more specific about some goals.

After getting everything set up, I pointed the telescope in a direction and started taking pictures of the first bright stars I found. I tried to make sort of a contact sheet, taking longer and longer exposures and seeing the point at which the background stars started to change from circles to ovals. This gives an idea of an upper bound for exposure time with this equipment. This point happened around 60-65 seconds, meaning I can confidently take exposures up to a minute long without losing too much data. Here’s one of the photos, just to keep things interesting. This is a 30 second exposure (which I thought looked the best), and wow, there are a lot of stars you can’t normally see. If you zoom in you’ll notice how nice and round they all are. If it seems like I’m somewhat excited by this, I’ll offer the following photo as a comparison. Here is a 15 second exposure with the tracking disabled. Can you see the difference? I am very pleased with how well this is working, and it’s going to make a huge difference in future photographs.

Lets talk about some things that need work: I had a lot of trouble with polar alignment. In order for the tracking to work, I need to align the axis of rotation of the mount with the axis of rotation of the Earth. You essentially do this by precisely pointing it at the north star. It can be a pain, but it is also a thing that will get easier the more I do it. I just have to put in some more practice.

Another thing I need to do is get everything set up earlier, ideally when it’s still light out. Once it gets dark, it becomes a lot harder to do things like collimate the telescope (align the mirrors) and align the finder scope. I’ve known this one for a while, but there is a difference between knowing and doing. Preparation, in general, is a focus for this project.

Also, I was asked for a picture of my setup, so here’s how it looks in the daytime. It looks the same during the night but is harder to see. To explain a little, the mount turns on two axes: “Right Ascension” and “Declination”. You can think of these as projections of Earth’s longitude and latitude onto the sky, where RA is longitude and dec is latitude. The RA axis is the one you align with Polaris for tracking. The other notable, non-self-explanatory item in the picture is the coma corrector, which is a small lens system attached to the camera which, as the name says, corrects for the coma aberration. This optical aberration would otherwise make the stars towards the corners of the frame look like little comets and is common in fast Newtonian telescopes like mine. In addition, I just bought a cable to connect the camera to my laptop. This will allow the computer to take the pictures through software, as opposed to me standing out in the cold trying to work the shutter remote and a stopwatch while wearing winter gloves like a dummy.

Okay! The tracking on the mount now works. It was a frustratingly easy fix, so much so that I almost wish it had actually been broken so I would have an excuse. Almost. I suppose you would classify the issue under “user error” but suffice to say, the manual could use some revising, and this particular point wasn’t mentioned in any of the multitude of calls I made to the company’s tech support. I am equal parts elated, aggravated, and exhausted.

Figuring out the tracking issue was always going to be the hardest part for me. Getting out there and taking pictures is the fun bit, so needs a lot less motivation. I could mark this as completed right now and consider it a roaring success. Any images I come away with in the next four weeks are absolutely cherries on top.

I genuinely don’t know how much technical stuff I should be explaining in these updates, ie. what I’m doing specifically and why. But to shine a little light on things (ha), the tracking on this mount will allow me to keep the shutter of my camera open for much longer before star trails appear. Without it, I could only have an exposure time of 1-2 seconds before the stars that should be points start looking like streaks. With the tracking, I’ll be able to keep the shutter open for closer to 60 seconds. To state the obvious, it’s going to make a huge difference in the amount of light I’m able to collect.

As far as possible objects to photograph go, I have a few in mind sorted into categories of “easy,” “medium,” and “hard.” I may update my weekly goals based on them seeing as it took so little time to get to this point.

Anyways, thanks for reading, I guess. Hopefully I’ll have pictures of cool space stuff to share soon.