ngc 2403

[alacant]

Hi everyone. Not seen this one here before so thought I'd have a go. Close to the pole, the mount balance seems to get more critical -especially DEC- something I hadn't realised before. I'm in two minds whether to have a go at a longer focal length; not sure if simply enlarging what I already have here would have the same effect. Any comments on the choice of fl or indeed anything and everything most welcome.

This was also supposed to be a trial to move the position of the filter. Alas I was about 30 minutes in when I realised... Duh.

Thanks for looking and clear skies.

 

[Filroden]

I'm not familiar with this one but it looks to be a lovely tight spiral. You can just see the arms so more time on the target would be worth it I think. 

Thank you for sharing. 

[gnomus]

Nice effort Alicant.  You've maintained star colour and the galaxy looks natural and not overprocessed.  I think seeng some 'air' around the galaxy is not a bad thing.  

As to longer focal length - well it all depends.  Try plugging your proposed telescope and camera into the Astronomy Tools website and see what imaging resolution you come up with.  Then take a long hard look at your mount, put on a serious face and ask it, man-to-man, whether it really is up to that task.

[alacant]

10 hours ago, gnomus said:

plugging your proposed telescope and camera into the Astronomy Tools website

Hi. I get 1.09" current and 0.74" proposed although I don't know what that tells me.

10 hours ago, gnomus said:

look at your mount,

Yeah. Good point, although with PHD2's new PEC algorithm I think it'll be OK. The worse thing about it is that it's a miserably dim f8.

[gnomus]

1 hour ago, alacant said:

Hi. I get 1.09" current and 0.74" proposed although I don't know what that tells me.

Yeah. Good point, although with PHD2's new PEC algorithm I think it'll be OK. The worse thing about it is that it's a miserably dim f8.

I read somewhere that you should be aiming to have a guiding error no greater than half of your imaging resolution.  That means you currently need around 0.5" RMS total error, and your proposed rig would need 0.37".  That is a tall ask for a mount.  My Mesu can do this most nights.  My EQ6-R could not.  You don't say what camera you have (a DSLR perhaps?).  With a CCD you can 'bin', of course.    

[EDIT:  Please note that I am just starting to get my head around these things and it is possible I have misunderstood something.  If so, could a wiser head please jump in and let me know?] 

[alacant]

Hi. OK, you've lost me! I have tried 1200mm before, it just takes so long to get anything decent. I thought maybe this target was brighter. Dunno...

[ollypenrice]

I agree with Steve.

The theoretical resolution of your system is measured in arcseconds per pixel, meaning, 'How many arcseconds of sky land on one pixel?' The smaller the piece of sky sampled by one pixel, the more detail you will theoretically resolve. (If there is an arcsecond of dark sky between two faint stars then I cannot separate them with our Twin Tak rig because it operates at a coarse 3.5 arcsecs per pixel. Both the faint stars and the gap between them land on the same pixel.) If I use a long FL system working at 0.6 "PP then I can separate the stars. *

All of this is theoretical because if the guiding is moving around by an arcsecond during the exposure, the arcsecond gap will be blurred out and will disappear from the final image. In fact, as Steve says, the accepted rule of thumb is that your mount needs to track with about half the pixel scale of the camera-scope in order for the theoretical resolution to appear in the picture. I don't know how this 'accepted figure' is arrived at but I just accept it! (Baaas like a sheep. )

Even if your mount can track without any error whatever you also need the beam of light coming down from the stars to be undistorted by the atmosphere, and it won't be. An imaginary beam from a point source can easily be stirred around so that instead of landing on one pixel, over a few seconds, it lands on several.

0.74"PP is not going to be easy to realize.

Also to retain your Quattro F ratio with a long focal length is going to need a BIG scope! (Of themselves focal lengths tell you nothing about exposure times. You can have a fast long FL scope but it won't go on an EQ6!

Good image. Working at a long FL it all gets harder and (very easily) very expensive...

Olly

* In fact stars will never land on one pixel but I'm using the one arcsecond gap as a straightforward way to explain resolution in theory and practice.

[gnomus]

48 minutes ago, ollypenrice said:

...  In fact, as Steve says, the accepted rule of thumb is that your mount needs to track with about half the pixel scale of the camera-scope in order for the theoretical resolution to appear in the picture. I don't know how this 'accepted figure' is arrived at but I just accept it! ...

A detailed and clear explanation from Olly (as usual).

As to the quoted point, I had assumed that this is because the PHD error is reported as 'plus/minus'.  Therefore, thinking in only one dimension, if the error is +/- 0.5" then the total potential movement is 1".  So to image at 1" you need a maximum guiding error of +/- 0.5".  

I don't know if my understanding is correct, but it is how I think about it, and remember it.

Steve

[ollypenrice]

3 minutes ago, gnomus said:

A detailed and clear explanation from Olly (as usual).

As to the quoted point, I had assumed that this is because the PHD error is reported as 'plus/minus'.  Therefore, thinking in only one dimension, if the error is +/- 0.5" then the total potential movement is 1".  So to image at 1" you need a maximum guiding error of +/- 0.5".  

I don't know if my understanding is correct, but it is how I think about it, and remember it.

Steve

It really is a bit complicated, this one. For instance, we actually know nothing about what is happening between guiding commands. We only know the position of the guide star on the chip as each new command is sent. Avalon claim that their belt drive mounts, having no backlash, return very quickly to the target position on receiving a command. They are, therefore 'on target' for longer between guide commands than conventional mounts. I'm prepared to believe this because our Avalon does seem to me to produce better results than the guide trace suggests. It's hard to quantify this, though.

And then we must remember that we also know nothing about where the mount is really pointing. We only know where the guide star is on the chip, and the guide star can be moving around due to turbulence. So if we take lots of very short and frequent guide subs we may get a better guide trace. This, alas, ignores the fact that what we have is just a good trace of a star dancing about in the seeing. We would get better real guiding by using longer and less frequent guide subs in which the turbulence has been averaged out.

But the rule of thumb (guide error to be no more than half the pixel scale) seems reasonable to me.

Olly