The weather this year has been frustrating. Winter held on far longer than normal. While temperatures weren’t freezing in March and April, they stubbornly refused to get into a temperature range that would allow me to open the windows. Then, after a few nice days, it immediately jumped into the early summer. Needless to say, such turbulent weather is not good for observing and I’ve only been out twice this year1
However, temperatures have cooled back off and, should a clear night occur, it would be a quite pleasant night to observe. This past Friday, despite solid cloud cover all day, the weather reports said it should clear by around $8$pm, so I headed out to Danville with the quadrant to do some observing. Although there was some high humidity, an adjustment in how I select what objects to observe made this likely one of the most productive observing sessions I’ve yet had.
In truth, the new method is really the old method. When I first set out to create a catalog of stellar positions, I created a list of $\approx500$ stars that I wanted to include. Since, at that time, the quadrant didn’t have a way to read the azimuth, it was fixed to the meridian and stars could be observed as they crossed the meridian. We quickly found that this was excruciatingly slow for many parts of the sky in which there’s just not many stars and I added the azimuth ring in response.
However, that meant we no longer had to concern ourselves with what stars needed to be targeted when, and as a result, we usually picked a region of the sky and mapped whatever we felt like. This resulted in a strong bias towards bright stars and many of the stars that were challenges were often skipped.
To help make sure I didn’t do this, I’ve now printed out the list again, as well as printed out a star map for each star. Then, when observing, I’d pick a region of the sky I wanted to observe and could use these print outs to help me select my targets. This forced me to challenge myself with fainter stars than I would have otherwise gone for. In this observing session, I was able to target stars down to nearly $5^{th}$ magnitude although the faintest stars on the list were around magnitude $4.5$.
Ultimately, I ended up with $64$ observations2. A full half of them were stars that I’d not previously observed. This filled out a lot of Leo and Ursa Major3. Oddly enough, there’s still several bright stars that are part of the Big Dipper that I don’t have measurements for. I know I’ve taken them before, but I guess they were on a night where the data was bad and got tossed.
Speaking of data integrity, I’ve been refining my methods on making corrections there as well. One of the things I’ve been noticing in the past year is that levelling isn’t a process of setting it and being done for the night. Rather, it needs to be constantly adjusted. And the more I pay attention to this, the more reasons I come up with. The first was that the center of gravity isn’t in the center of the central column thanks to the big quadrant hanging off it. It pulls slightly to the side that the quadrant hangs from. Thus, as that turns throughout the night, it needs to be releveled for that. This seems to be the biggest one. It’s generally fine when doing a single constellation, but change more than $30-40º$ in azimuth, and it’s best to check.
Second, there frequently seems to be settling. While I have some hard wood blocks that I put under the corners with the leveling screws, the ground is often still somewhat soft and the blocks slowly settle in. Thus, even without moving the quadrant, occasional leveling seems to be prudent.
The worst issue, that I can’t really do much about, is stars near the horizon. In this instance, the quadrant gets lifted quite high and the center of mass pulls to that side. Since it’s a single observation, there’s not really any re-leveling that can be done.
The other consistent issue that I’ve faced is determining north to align the azimuth ring. This time, I lined up the north star at $9:02$pm and set the azimuth ring accordingly. However, after arriving home, I looked up the true position of the north star, since it isn’t directly on the north celestial pole, and found it was nearly a half degree ($0;29,25º$) to the west of true north. I also checked the alignment of the north star after observing and came up with a reading of $1.5º$ when it should have been $0;17º$ to the east of north.
So what to do with all this information? Well, I grouped the two observations. For the first, my alignment was $0;29,25º$ off from what should have been true north. For the second, I’d marked the north star $1.5º$ off from the first measurement, but $0;17º$ of that was the offset of the star from north, so really, it was only $1;13º$ from my first measurement. I then took these two values and averaged to come up with a correction of $0;51º$, or, since I do my measurements in decimal, $0.85º$ which I subtracted from all the measurements I took for azimuth.
How did that end up making the data? Ultimately, extremely clean. While it’s not the best night I’ve had observing, it was quite close. The average error in RA was… wait for it… zero. I’ve never had that happen before. My previous low was $0.02º$ which is indistinguishable from zero within instrumental uncertainty, but never has the math worked out so well. The standard deviation was also very good at only $0.29º$. The error in Dec was also quite low at only $0.07º$ with a standard deviation of $0.28º$.
So overall, this was a fantastic night of observing and has brought me up to $458$ stars in the catalog now. And here’s what that looks like:
- And really the last time was more of training someone as it wasn’t a great moon phase.
- Plus two that were so far off I tossed them.
- As a reminder, Ursa Major is a large constellation. The Big Dipper is just a small part of it and isn’t an official constellation in its own right. Such “not official constellations” are known as asterisms.