Jupiter 9-12-11

[Carole]

Still struggling to get decent colour and any detail on Jupiter.  Every image I take of Jupiter seems to be slightly better than the last one, but making VERY slow progress.

I decided to try out my ED120 with a 5 x Powermate this evening, but the seeing was VERY wobbly.  Worked better without guiding which seemed to make it even more wobbly.  

This is the best of a number of AVIs I took over a half hour period.
Captured in Sharpcap, 30 f.p.s. (I tried various settings), Philips SPC 900NC PC Camera
ED120 5 x powermate, unguided
approx 1200 frames about 50% of the best
Stacked in Registax, wavelets between 25% and 40%
Final twiddles in Photoshop CS3

[Mac]

Looking good,

Dont forget that in half an hour, Jupiter would have rotated quite a bit ~15 degrees.

So it might be the fact the the images are poor due to the fact you are stacking images that have rotated,

have you tried cutting down the length of stacked images,
say in 5 min blocks to see if that improves the detail.

Or even 1 min.

30fps for 60secs = 1800 frames, which should be more then enough to get the detail you need.

Another thing to think about, is the x5 pushing your optics above their design limit.

http://en.wikipedia.org/wiki/Dawes'_limit

Mac.

[Carole]

Hi Mac,

I didn't actually use the images spread over the half hour period as I knew about rotation.  I only stacked images taken within the same 5mins, but did various batches over the half hour stacked small groups of about 1200 or 1800 frames and posted the best, sorry to confuse.  

I tried the 5 x powermate because some-one on another forum suggested it would be comparable, to using my ETX125, so I was just really trying it all out.  

Yes the images are poor, but so was the seeing.  
Didn't understand how to work out the maximum resolving power from that link I'm afraid.

I'll have another go at Jupiter sometime when hopefully the conditions will be better.  

Carole

[MarkS]

A x5 Powermate will push the F-ratio from f/7.5 to f/37.5  but, in fact, depending on CCD spacing etc. it is likely to be more.  Carole, assuming Jupiter was 46" in diameter last night, you seem to be getting a focal length of 5020 which translates into an F-ratio of f/42

The Dawes limit is one factor in determining resolving power but we are now living in an age of CCD digitial spatial sampling and the important thing is to ensure that everything that the diffraction limited optics are capable of resolving is captured by the sensor.  Generally speaking we need around 3x oversampling.  For the CCD pixel pitches we tend to use, f/25 - f/40 is more or less optimum for resolving detail.  Just look at the imagery produced by Damian Peach - (Mac - as an exercise, try calculating the F-ratio he is using from his C14!)

However in practical terms, we are not imaging in pristine Barbados skies and are not using top of the range sensors.  In terms of overcoming sensor noise, you want a bright image so you want to reduce the F-ratio.  My best Jupiter image was done with an SPC900 at f/25.

So, Carole, to turn the above argument into practical advice, a x3 Barlow might serve you better for Jupiter imaging with that scope.

Mark

[Carole]

Thanks Mark for that summary.  I don't have a 3 x barlow, It is possible I might be able to do 2 of 2 x barlow, (by using a combi of a 2" and 1.25" one if the step down adapter will work with a barlow) otherwise I will have to acquire a 3 x barlow.  

On the basis of your working out above, the 4 x barlow will work out to f/30 which is within the scale you recommended

For the CCD pixel pitches we tend to use, f/25 - f/40 is more or less optimum for resolving detail

I'll give that a try next time if it works, otherwise will have to stick with the 2 x barlow until I can acquire a 3 x barlow.

Carole

[MarkS]

You might also find the following thread interesting, where I did some practical experiments without atmospheric seeing interfering.

http://forum.orpington-astronomy.org.uk/index.php?topic=4201

However, when I did those experiments, I did not take into account the extra resolution that can be gained using wavelet analysis.  I will repeat this experiment some time and see what extra can be gained.  It may be worthwhile to push beyond f/45.

And don't forget what the technique of super-resolution can offer:
http://en.wikipedia.org/wiki/Super-resolution  

In practical terms this would allow you to image at f/20 and then simulate f/40 by super-resolution.  I think the "Drizzle" technique is much the same thing.  They both rely on the fact that dithering is performed between frames.

Mark

[Carole]

Thanks for that information Mark, I don't remember that thread.
Does any-one know of a good tutorial for Registax, as I think my Registax processing skills could do with some refining.

Carole

[Mac]

That super resolution article looks pretty good,

Mac.

[RobertM]

That's an excellent attempt Carole, I cartainly haven't managed to get one that good yet (though I keep trying !).

I did something akin to super resolution with the H9 images.  Dithered, double sized then stacked to give better resolution.  It looks like a very interesting teaser article, especially the bit about breaking the diffraction limit.  I found this page http://blog.everydayscientist.com/?p=354 which seems very informative.  There are also super lenses - it's all very impressive stuff.  Thanks for posting Mark.

Robert

[Carole]

Thanks Robert, I did it with the Webcam I bought off you. 
I must say I was amazed that the mount kept such a magnified object in the field of view without guiding, but fiddle with the controls as I may, I could not get anything clearer than this. 

Gonna have to keep trying and buy that 3 x barlow and wait for better seeing I think.

CArole