Is this a focus problem?

[almcl]

For a little while now I've been attempting astrophotography.  Started when I didn't have a telescope, continued using a

home made barn door tracker and stopped when I got my Skywatcher 200P, but now feel inclined to try again.

The images below were done with a fixed lens bridge camera (Panasonic Lumix DMC FZ30 - if you want to have a laugh). 

Now I appreciate this is not going to produce results like Olly Penrice showed me at Les Granges in September - what a voyage of discovery that was!  His tutorial on post processing was a revelation too.  But I still want to see how far I can take this, before I splash out loads of cash on new equipment.

To that end I've been attempting to image M31 Andromeda, gradually pushing the zoom lens on the camera towards its limits.  However the latest results have been rather unsatisfactory and I am not sure what's wrong.  The stars, which have never been *that* pinpointy are now excessively bloated. 

I am using DSS 3.3.2 and stacking as many 1 minute RAW subs as weather, dew, battery &c will allow, working at ISO 200  (because 1 minute is the max exposure on the camera and ISO 200 is almost the maximum and 400 is just too noisy).

So here are two examples one at 300mm lacks detail and suffers from noise and light pollution but has round stars.  The second at 20mm has less noise but looks awful.

Is it just a focus issue?  The camera's screen is miniscule and telling whether or not it's in focus even with MF assist is problematic, but each star seems to be a doughnut.
 

This effect doesn't seem to be present when working at shorter focal lengths. 

Or is this an artefact produced by inadequate tracking which only becomes apparent at the extremes of the zoom lens? (I have the 6 volt Skywatcher dual axis motor drives and am reasonably confident that the polar alignment is OK.)

Any comments or suggestions would be welcome. (I've got some other questions to ask but will save those for other threads.)

[Kropster]

I would think that the 20mm shot (second one?) had been enlarged a lot more to get Andromeda a similar size to shot 1.

In which case you are looking at a much smaller part of the original image and hence a smaller part of the imaging chip.... less pixels, so less information and softer.

Akin to getting too close to the TV screen and the picture looking soft.

Maybe

[FunkyKoval35]

Hi almcl Yes, it is looking like focus issue - black dots in the center of stars(see zoomed stars from your image). To avoid it you can use Bahtinov focusing mask. Also, for better results you need longer exposures = better SNR.

Adam

[Carina Lass]

take some darks as well if you can - subtracting the noise will help a lot.

Focus - do you have a screen shot that gives you a 10x view ?

That might allow you to take a look at the focus.

Failing that, you will have to make a bahtinov mask - not difficult - I used the back of a plastic folder and it's lasted me years.

[almcl]

Thanks for the thoughts!

Apologies, I managed a typo in describing the images (despite editing the post for 15 minutes off-line)  the first one is 300mm, the second 420mm.  Although cropped, I think they are both approximately the same enlargement.

I hadn't realised a Bahtinov could be used directly on the camera lens.  As the optical bit is only 38 mm across, is it possible to generate one at this size or should I use the 55mm filter screw thread diameter?

Thanks for the advice on darks.  I did shoot darks but could not  see any difference between the stacked images with and without them.  ( a couple of examples below from a previously unsuccessful session).  Could I be doing something wrong in Deep Sky Stacker?  Or is the difference more subtle?

[glowingturnip]

no reason you can't make a bahtinov mask for a camera lens, but they are focal-length specific I believe, so not going to work with a zoom lens

[Carina Lass]

Oddly enough, they work on the same principle as the secondary mirror spider giving an optical spike.

you can have a very simple mask that looks like a Y it will work fine.

[michael8554]

I think that given the equipment involved,  the very first image (ISO 200, no darks, no Photoshop) is  surprisingly good.

Keep it up!

Michael

[almcl]

Thanks for the kind words, Michael!

I had hoped to improve but am still puzzled as to why DSS doesn't seem to improve the image when darks (and/or my 17 bias frames) are included.

I think I may have got a line on the focus problem.  There are some bright lights on the horizon about 2km away and manually focussing on those may help, if we ever get a clear, moonless night!