For quite a while, I've meant to learn how to take scientific data, such as variable star observations, exoplanet transits, etc. I got connected to the AAVSO (American Association of Variable Star Observers) by way of a friend of mine from my last astronomy club, Phil, who put me in touch with the Director of the AAVSO, Stella Kafka, because she's a female astrophysics PhD who he thought I might like to talk to for mentorship. I resolved after our Skype call to look at adding observation runs to the beginning or end of my imaging runs, but it never happened, for one reason or another.
Since I currently have two rigs up in the backyard, but only one good monochrome camera that I quickly grew tired of swapping back and forth between the two platforms, I decided I ought to put the second rig to good use as a scientific data-taking rig! Believe it or not, the gear requirements for taking useful observations for variable stars and exoplanets is less than for deep-sky astrophotography. The exposures are generally shorter, your stars don't necessarily need to be round nor in focus (actually it can be helpful for them to be slightly de-focused, so as so spread the light across multiple pixels and get a more accurate reading), your camera doesn't have to be as good, and your tracking doesn't have to be as good. As an example, on last night's The Astro Imaging Channel's presentation about amateur observing of exoplanets, NASA's Rob Zellem mentioned that he had a student who used a 6-inch reflector on a mount with such bad tracking that the star drifted 300 pixels over the course of the dataset!
The second rig I have out in the backyard, aside from the Paramount MyT and Celestron C8 that is my pretty-picture imaging rig, is my Celestron AVX with a Vixen 8-inch f/4 Newtonian that a very generous member of my last astronomy club gave me. I was going to use it for imaging, but I haven't yet gotten a configuration of coma corrector and the spacing between that and my camera sorted out to have the coma reduced enough to properly take pretty pictures with it. So it's just been hangin' out in my backyard collecting pollen and spider webs underneath its Telegizmos 365 cover.
The thought occurred to me to make it a science rig a little while ago, but I couldn't decide what to do about a camera. I definitely needed a monochrome camera. On the upside, I still have my Starlight Xpress electronic filter wheel after swapping it out for the ZWO version that has more filter slots for the 2-inch size. I got a carousel for it that holds 1.25" filters a while back, and I recently popped in my 1.25" Astronomik RGB filters and my Schuler BVRI photometric filters that another generous member of my last astro club gave me (I don't have the U filter, but it appears to be less important than the others) into it.
I found a camera!
So I went a-hunting through my spare gear boxes -- all the stuff that I've acquired, usually by someone giving it to me, that I haven't put to use yet, or at least, am currently not using. I figured there was a camera in there I had forgotten about. And lo, there was such a camera -- an Orion Deep Space Monochrome Imager II.
|Internet image because I forgot to take my own|
It's an old camera -- old enough that its manual references the fact that you have to use USB 2.0 instead of 1.1, says you need Windows 2000, XP, or Vista, and also actually says that a mouse is a requirement -- but to my delight, it has some key features for an astro camera, particularly one for doing scientific data-taking.
- It's a CCD chip rather than CMOS. While CMOS is arguably more sensitive (at least, at the consumer level), CCD has a more linear performance, which is ideal for science.
- It has big juicy pixels -- 8.6 microns! Compared to my pretty-picture-taking CMOS camera, which has 3.8-micron pixels. This might sound weird to the casual reader, since bigger pixels mean less resolution, right? True -- the resolution is 2.2 arcsec/px on my Newtonian, which isn't ideal (something closer to 1 is better in most cases). But! Bigger pixels means more light-collecting area, and thus much higher sensitivity. This means I can get more light in a shorter amount of time (particularly helpful considering how not-great the tracking is on my AVX).
- It's a cooled camera! This one was a shocker. I expected it to have larger pixels and to be CCD, but when I looked up its specs, I learned that it has a TEC (thermo-electric cooler) that can go down to 20 degrees C below ambient. Woo hoo! This is great news because it means I can have much less noisy images. Hopefully. I'll still have plenty of noise to deal with -- it has a ridiculously high read noise of 24 electrons/ADU, according to some guy who made the measurements and posted about it -- but it at least reduces one noise source.
- It has a 16-bit ADC. At least, I think it does, according to some posts I saw about it online. Now, a lot of lower- and mid-level astro cameras have 12-bit or 14-bit ADCs (analog-to-digital converters -- basically, how many brightness levels you can encode into the saved image), which gets saved as a 16-bit number, but is really only converted to 12- or 14-bit. Having a true 16-bit ADC is normally reserved for the higher-end cameras. The higher the bit depth, the more brightness levels your software can discern between, which makes for a higher-contrast image, or better science data.
- It has a small chip. 1/2" diagonal, 752x582 pixels. Ordinarily this would be less than helpful, but since my particular problem with the Newtonian regards coma, which stretches out stars close to the edges of the field-of-view, this actually works in my favor by only imaging the "nice" center region of the FOV, which should have good-looking stars. Also, the frames should download faster. If Sequence Generator Pro downloaded frames in a reasonable amount of time, which it doesn't! Even on my fast CMOS cameras, which can run 10 fps in SharpCap full-frame (or at least 2 when I'm in 16-bit mode), it takes like 5s to download the frame. Hmph. Anyway...
One weird thing about this camera is that the TEC runs on a 3V DC power source. 3V?? It came with a battery back that takes two D-cell batteries. I have no idea how long those will run it for, but the manual says "it's a good idea to bring a spare set with you." Also, I have no clue whether or not the TEC is running, since you can't turn it on/off, set the temperature, etc. You just plug it in, and it cools to whatever 20C below ambient happens to be. I don't want to be buying D-cell batteries all the time, so I've got an adjustable AC power converter coming in the mail from Amazon.
But does the camera work?
One pitfall of older gear is that getting a given device to work on newer computer systems is dicey, as is finding drivers for it. Fortunately, the Orion website still had both the camera and ASCOM drivers. The ASCOM driver installed no problem, but the camera driver file just unzipped some system files and no executables or anything that looked helpful.
So I tried opening the camera in SharpCap just with the ASCOM driver (and it did show up there), but it said it couldn't connect. I tried a couple other programs, but they all said the same thing. So I hit up the internet and found out that you have to install the driver in a bit of a roundabout way: if you go to Device Manager (this is Windows 10 by the way), it'll show up in the Unknown Devices list. Click Update Driver, and choose the "browse local files" option. Then go to the folder where these driver files are located. The wizard installed the driver from those files, and bingo, it worked! I was able to open the camera in SharpCap, and it did show changes in light level when I took the cap on and off. Success!
Connecting things together
The next step was to get it installed on the telescope. I currently had an M42-M48 converter ring on the focuser so I could connect the coma corrector, but I needed to take it off since my filter wheel has an M42 connector (M42 is the same as T2 or T-thread). Unfortunately, it was pretty well stuck on there, and after many attempts, I was unable to remove it. So I grabbed one of the spacers that came with my ZWO camera, a 16.5mm guy, which has M48 female on one side and M42 male on the other. It effectively works as a converter. I got it all put together, but the telescope is unfortunately too close to the fence to be able to see any of the things in the distance I could use to test focus. So I had to wait for nightfall.
Once it was dark, I went outside into the blessedly cool night air -- my house had warmed up considerably in the sun to 85 degrees! I had the windows open and fans running, but it took quite a while to cool down. After getting my main rig hooked up and ready to go, I slewed the scope to Arcturus for focusing. Luckily for me, the giant de-focused star appeared in the camera's tiny field-of-view, so I was able to center it before focusing. Also luckily for me, it did come into focus! Not a whole lot of backfocus on this scope. I think if I'm going to put a motorized focuser on it in the future, I'm going to have to replace the whole focuser. We shall see. I still need to measure how parfocal my photometric filters are.
But, mount issues abound
If you've been following my blog, then you probably know that Celestron mounts have brought me little but woe. While I did have this AVX mount working quite nicely last fall, it has again turned against me with a horrifically high amount of declination backlash. Two months ago or so, I popped open the casing to see if I needed to adjust the gears again, but the gear meshing and tightness looked fine. So I don't know what's causing the backlash now. I didn't re-balance it after putting this gear on because I didn't want to have to re-do the alignment before I got it focused, so I'll do that in the morning.
I tried re-calibrating the autoguider, since it was really bad last time I tried, but it was just as bad this time.
As you can probably guess from that super-weird-looking calibration, guiding was nuts-crazy. So was tracking -- plate solving to get the first target of the night centered just was not working. Partly because sometimes the stars were streaked, although it still seemed to be finding those okay and not grabbing noise pixels. But it still couldn't plate solve, and I'm not sure why. Maybe it has to do with the other weird aspect of this camera -- the pixels aren't square. They're slightly rectangular! They're actually 8.6 x 8.3 microns. SGP only has one dimension for pixel size in the plate solve settings that it feeds to the plate solve software (I'm using Planewave's PlateSolve2 software, which is free by the way!)
Earlier in the day, I worked on a target list. The AAVSO has a target-selection tool that you can sort in a number of ways, including by how high-priority the target is. So I scrolled through to find targets scattered throughout the night that I estimated had magnitudes within reach of my gear and were within the field-of-view I have between my house, my plum and lemon trees, and my neighbor's garage. I put the targets and their info into a spreadsheet so I could keep track, since they each had different requirements. For some stars, the primary researchers wanted specific channels, like just B or V. For others, they want as much of a full set of photometry as you can take. Some are known to be long-period variables and only need observations every week or month. Some are changing fast and needed observations every minute for a couple hours throughout the night. Some are currently super-dim, but are expected to erupt soon, so they only want your observations if the star gets above a specific magnitude. So many different things! And that's just variable stars -- I'd like to do exoplanet light curves as well here soon.
You might recognize the name of one of those stars -- KIC 846-bunch-of-numbers. It's the famous Tabby's Star! Named for the Louisiana State University professor who discovered its odd behavior, this is the one that was showing dramatic dimming and brightening cycles that spurred the discussion of the chance that an alien civilization was constructing a Dyson sphere. The real explanation is probably less fantastical, but is still being investigated.
As you might guess, with the mount issues, the sequence did not collect any data last night. In fact, when I checked in the morning, it had apparently tracked past the meridian enough for the loose dec axis to do the thing where it falls a bit due to gravity, which causes the motors to go crazy. It wound up on the wrong side of the mount and pointing at the ground. Luckily, no harm was done, although I'm going to have to re-align it tonight. I also re-balanced it, which might help?
And the primary rig?
My main imaging rig, the Paramount MyT with my Celestron C8 and ZWO mono camera with my other filter wheel and my fancy PrimaLuce Lab Esatto focuser, ran flawlessly. I tweaked the aggressiveness values a bit more in PHD down to 40 in RA and 70 in Dec, which seemed to help even more with the problem of huge fluctuations for M51 and M13. The M51 guide graph still showed huge spikes in dec as high as 4 arcseconds, but the images did not suffer as much weirdness. I was able to keep the majority of the images. *sniffles* So beautiful!!