Software Bisque

Take a virtual tour of the Software Bisque campus Watch Steve Bisque discuss the Paramount mount with S&T's Dennis Di Cicco Read the Paramount MYT Review Read about the Paramount Software Suite

Evaluating the One Star Polar Alignment Technique
Richard S. Wright Jr.'s Blog
"Make hay while the sun shines", or so they say where I grew up. Or, when the skies get good, pack up and head down to the observatory for some "field work". In Golden there is an observatory in the parking lot at Software Bisque. I have to make due with my back yard, and I have precious little light polluted sky to choose from between my neighbor's trees in my neighborhood outside Orlando, Florida.

Paramount MYT ready to roll.

So I've spent the last few days down at my dark sky site in Okeechobee county Florida. SQM of 21.03 Tuesday night.  Yeah, be jealous, just don't build a Walmart down the road...

I have a couple of educational projects I'm working on, such as explaining why you don't need a polar scope for people who didn't take all that vector math stuff in high school because it didn't apply to "real life" <vbg>, and improving on my wind-blown daytime polar alignment video.  Editing video takes forever, so here's a blog with something useful right off the bat.

There are some things you just know will work, but you really should "try" them, and try them more than once (the essence of science is repeatability!).

I decided to do a rigorous review of the one star polar alignment technique, and evaluate it's performance quantitively for both photography and visual astronomy. For night #1, I setup a 150mm f/7 refractor on the MYT and put a DSLR on the back. The refractor weighed more than the mount (35 lb.), and is probably representative of what a portable imager might be willing to lug into the field. I setup the tripod, leveled it... and was careful to not be too careful.

Imagine it's dark, there are crying bored children, and a wife complaining about being bitten by bugs (this is my own experience... I'm not being sexist, I know there are plenty of women interested in astronomy...just not the one I married ;-).

I cleared the sync using the polar alignment routine on the iPad, and centered Capella after dark using only the altitude and azimuth adjustments on the mount. In order to simulate a visual experience I did this all by eye though the view finder of the DSLR, I did not do an Image Link (other than the iPad, I had no computer hooked up at the time!).

At this point for visual astronomy, I should be ready to go, and if I'm guiding I should be ready to go. "What's that?" Didn't you run TPoint first and dial in the polar alignment?

Yes and no. I did in fact run TPoint, because I wanted to know just how accurate my somewhat clumsy, potentially rushed attempt at polar alignment really was. I did not make any mount adjustments however. A recurring theme in the forum and in personal emails and conversations is that running TPoint "takes too long". I want to know if I can tell people, "Fine... just don't run one then!". The Paramount's have homing sensors, which makes the one star alignment technique possible (just how with all the vector and coordinate frame stuff will be the subject of a future blog... but in layman's layman's (sic) terms... look, 10 - 7 = 3, and 7 + 3 really does give you 10 back.

Sometimes I think we do a bunch of TPoint samples and adjust our mounts out of some perceived need that isn't really there (I have the power, I MUST use it!). I've even seen some of our users bragging about how many samples they ran before they started imaging...

I'm not sure that's helping either. I'm only shooting at 1050mm focal length, and I'm guiding. How close to the pole do I really need to be? Well, let's see how close I actually am, and see how well it does tonight. I did an automated 97 point TPoint run. After running Super Model, I pulled up the polar alignment report. With Super Model I had 21.2 arcsecond pointing, yet I was off a whopping 27.7 arc minutes in azimuth, and 20.7 arc minutes in altitude. (18.8 ticks and 11.3 ticks respectively on the calibrated adjustments).

This sounds like a lot (about half a degree), but remember we are guiding. Running TPoint is great, it's even fun, and there is a satisfaction to knowing how close you are to the pole. But it can easily become an unnecessary expectation and burden if you're in a hurry. Again, a common theme on the forums is "how long it takes"... well it took about 15 minutes to setup the mount with the camera, and maybe less than five to futz with my one star alignment. The 97 point TPoint run was not needed, it was just so I could have a metric for how off I could get away with. What I did next will surprise a lot of people.

I did NOT make the recommended adjustments to the mount. I left it be nearly 1/2 a degree off the pole. Is that close enough to guide? Of course it is! I shot 5 minute subs with an f/7 refractor on an uncooled  DSLR. See for yourself the image of the Pleiades, including the close up crop, but not resampled image showing very small round tight stars. I did a single 10 minute sub, but decided for 5 as I was overexposing too much around the bright stars. I've also included a snapshot of the guider graph... the guider was not working that hard (to be fair, note I did have a PEC table loaded - thanks Tom Bisque for doing this for me ;-), and aside from a few spikes from the dither operations, it's pretty decent.

The Pleiades.

Zoomed in.

So if you think it takes too long to run TPoint when you're mobile and you're shooting at a reasonably short focal lengths, then just don't do it. Use Closed Loop Slew to acquire your targets, and guide. You're done. What if you're shooting at something longer, say 2500mm or more? What if I'm really not very level, and the one star technique is way way off? Fine, do a short model.


Okay, listen... you know how when you watch something it feels like it takes forever? Let's be rational for a moment. It takes 15 seconds for my DSLR to take a 5 second exposure (use 2x2 binned in the newest Canon plug-in!), download, plate solve, and then move to the next target. This means I can do a short run of 16 points in four minutes.

If you don't have four minutes then you're obsessing over your imaging run far more than I am, and that's really saying something! The polar alignment report with only 16 points will not be the best, but it will get you as close or closer than my best and careful one star alignment technique (and I've gotten to within 5 arcminutes using a sun spotter during the day). 

Don't forget you're going to get to a rough focus with your camera and take a couple of test shots anyway, so don't add that to perceived time it's taking you to do a short TPoint run either.

Graph of autoguider log.

So, why do I need TPoint for mobile imaging? Well, an extra 5 minutes sure would have gotten me a lot closer to the pole, and I might not have needed to guide. My setup was pretty good too, I did not have any severe misalignments with the scope on the versa plate, etc. TPoint can make a sloppy setup perform like a good setup, and it can make a good setup perform like an almost perfect setup.
What too, if you don't want to, or can't guide? For unguided imaging you are limited by if nothing else atmospheric refraction to about 5 minutes or less at the most common focal lengths we shoot at. 5 or 10 minutes of an automated TPoint run will get you close enough for this.

It's hard to respond to someone who wants to show up at a site after dark, setup in 10 minutes, and be shooting unguided. You can get there with practice (visions of the NASCAR pit crews are dancing in my head), and only if you're shooting at very short focal lengths and a lot of other factors line up just perfectly... such as your optics temperature already matching the ambient temperature.

Here's something you can try for fun. Show up during the daylight. Setup and do the one star polar alignment using the sun or a bright planet. Ten minutes after the sun sets, you can start getting stars on your chip. Do a handful of points manually picking the brighter stars in the darker parts of the sky opposite the setting sun. As soon as you can Image Link, anywhere, start an automated run. By the time astronomical twilight gets here, you've got a pretty decent sized model built up, and if you let your scope cool at bit more (I'm finding my really good refractors need almost an hour to cool down) you can actually get enough points in to turn on ProTrack. Unguided for 10 minutes or more is quite reasonable, and there's no guide star to lose if a cloud comes through, or to reacquire on the other side of the meridian. You can even dither with the attached script, which auto-magically does a meridian flip by virtue of each dither being a tiny slew.

I don't have to experiment to see if that works though... I do it all the time.

Just say'n. ;-)


Posted 12-17-2014 8:01 AM by Richard Wright
Filed under:


© 2019 Software Bisque, Inc. 862 Brickyard Circle, Golden, CO 80403-8058 USA - phone: +1 303 278 4478 fax: +1 303 278 0045