Building Sky Models, Why?

[russellhq]

I've read a few times now about folks building sky models duing their session. I don't really know what this is but I'm assuming it's a more rhobust version of a 3 point star alignment?

Since I'm a lazy astronomer and clear sky time is precious, I've never looked any further into this subject. Rather I use plate solving to find and centre my targets.

But maybe there's something I'm missing. Is there any extra value building these models brings except for improving pointing accuracy?

Also, I work off a tripod I take inside each night so I assume the modelling wouldn't work for me?

[bingevader]

Sorry, pleb alert!

Sky models?

Do you need sticky backed plastic for that!?

Plate solving?

I've heard of plate spinning, but not solving, mine tend to just go on the table with food on them.

I'm sure some more enlightened folk will be able to help you with this.

Cheers

[IanL]

Depends on what you are using to run your mount. In your case EQMOD, so if you read this you will learn everything you ever wanted to know (and more) about how it builds and using pointing models:

http://eq-mod.sourceforge.net/tutindex.html

Basically it has two modes:

- A single alignment point mode in which case it calculates how to position the mount on a given RA and Dec by offsetting from a a single known point.

- A three alignment point mode, in which case it calculates how to position based on a triangle of points (which is much more accurate).

You can define up to 1,000 alignment points in EQMOD, and it will intelligently:

- Choose the single alignment point nearest your target if you have fewer than three points defined or your target is outside all possible triangles of three points;

- or choose the best three alignment points by finding the triangle of points that has its centre nearest to your target.

EQMOD has another useful feature which is 'append on sync', so basically every time you slew to a target and centre it in the field of view you can issue a 'sync' from your planetarium to tell EQMOD where you are actually pointing so it can update it's known position, but at the same time create a new alignment point in the model. You can therefore build up a better and better model as you go through the night adding more points and making each subsequent GOTO more accurate.

If you are using AstroTortilla to do plate solving, you can just tick the 'Sync Scope' option and assuming you have 'Append on Sync' enabled in EQMOD the whole process will just happen automatically with no intervention, so you should find that you are much closer to your target on the first GOTO after you have solved mutliple times in different parts of the sky.

(Plate solving is a process whereby you take a picture of the sky using a camera, and then your computer finds all the stars in the image, spends a few seconds playing 'join the dots' and then tells you exactly where you are looking. Take a look at http://astrotortilla.sf.net/ for software that can do it on Windows; there is a big thread running under the Discussions - Software section about it. Or take a look at http://nova.astrometry.net/ for a web site where you can upload images and have them figure out what it is you've captured for examples).

If you have a permanent setup, you can build a pointing model and save it to disk from EQMOD and re-load it the next time out. If you are portable, you have to start over the next time as any saved model would be useless.

There are all sorts of other ways of building pointing models from the simple one/two/three star models in Skywatcher and other handsets to super-sophisticated stuff like Tpoint (http://www.tpsoft.demon.co.uk/) which is used with the Paramounts and high-end professional scopes to build permanent pointing models that can compensate for mechanical issues like gears, flexure of the scope and mount under gratvity, etc. to get super-accurate gotos all the time (like a couple of arcseconds on pro scopes).

Your point about plate solving is well made though, for amateurs I think it is the best solution since you can just dump the scope on the tripod, do a quick polar alignment and get on to your target in a minute or two without doing any star alignment. Just as those of us who learned our way about the sky the hard way like to complain about how easy it is for the kids of today with their goto scopes, the next generation will be grumbling that people don't even know how to do a three point alignment.

I suspect big pro telescopes will always use complex pointing models since you don't want to be wasting precious and expensive imaging time whilst your several-hundred tonne scope slews about the sky looking for stuff (even assuming you have an instrument that produces an image suitable for plate solving, which you won't if you are doing spectroscopy or something similar).

[Gomtuu]

the kids of today with their goto scopes

I'm a 47yo kid and I love my goto mounts

[bingevader]

Ah, enlightenment!

We're talking about goto!

Just as those of us who learned our way about the sky the hard way like to complain about how easy it is for the kids of today with their goto scopes

Less of the past tense please!

I'm not complaining, I like star-hopping.

Cheers

[russellhq]

Thanks Ian, great response. Think I'll give it a miss and stick with AT!

[chrisshillito]

Thanks Ian, great response. Think I'll give it a miss and stick with AT!

But all AT tells you is where you are pointing - it doesn't fix your gotos. To do that you still need some 'modelling' element to be present. You may not be implicitly aware of it, but everytime AT solves and issues a sync, the mounts point model is being refined.

Chris

[Tim]

The other night I traveled out to get a pic of the comet and the site had more or less 360 degree horizon. By using Sirius, Mirfak and Arcturus, each of them more or less as far apart as could be, I got a brilliant model right off the bat with the synscan handset and heq5, putting targets dead centre of my C9.25 every time. I guess the wider the triangle the better the model?

My current mount has a GPS dongle which helps with accuracy, I suppose because the time is very accurate.

[Tinker1947]

This is a nice clock to have running when setting up, the time shown is spot on to my Radio Frequency Clock.......

http://www.atomic-clock.org.uk/atomic-clock.php

[russellhq]

But all AT tells you is where you are pointing - it doesn't fix your gotos. To do that you still need some 'modelling' element to be present. You may not be implicitly aware of it, but everytime AT solves and issues a sync, the mounts point model is being refined.

Chris

Thanks Chris. I was thinking about the difference between modelling the whole sky versus just the local point I'm interested in. Sounds like there's no point doing a whole model of the sky if I'm just going to image 1 or 2 targets in an evening and then take everything indoors when complete.