Tuesday, December 27, 2005


This one is the mosaic for 840/841 instead, the last one is taken by pan and expand.

Monday, December 26, 2005

Solar observation

840/841 mosaic at f/24

plus 838 at f/36

Sunday, December 18, 2005

Solar observation today

Woke up at 4:30pm for a good afternoon sleep, watched briefly without opening the window (!!), the view was not much affected.

Used the Unistar mount which is easy to setup and use, 20mm Widescan view is less great that the 13mm Nagler view, which is simply optimal I would say, at least for visual impact and contrast. Also taken some afocal shots, but it's going to be no good since my 12x DC is not good at that.

Two solar observation

17 Dec 2005: afternoon, very brief, taken some clips as follows, one at f/36 and one mosaic at f/24.

AR835 and AR822, 835 was also magnified.

Wednesday, December 14, 2005

Nice prominence on 20051127

No time to process until now, they're quite nice indeed!!!

an alternative way to gain more field with webcam

We know that webcam has a very small chip with limited amount of pixels, and if we want more field, we will usually make mosaic.

Making mosaic is quite tedius and hard, for examples, to fully utilize only 40mm of aperture, I will need over 40 raw frames stitched together to ensure enough overlapping. It's hard to imagine if one use a larger aperture, it would mean a lot of work.

Attached please find an image taken with a webcam, it has more than 640x480 and it's wider than a webcam can give, but I don't need to go through the stitching work. It's just a proof of concept shot, so don't let the poor quality to turn you down. :)

The key is to use "expand image" is registax.

In around the middle of the image, you can see a sunspot. To capture a wider area around the sunspot, I start the AVI capturing on the left of the sunspot, keep it there for a period of time, say 20s and then you start to slew your mount slowly to the right, so the sunspot is now on the left without stopping the capture, now stay sometime on the right of the same sunspot.

So, the above AVI contains both the images of the left, middle and the right of the sunspot. Notice that it's just a single AVI.

To stack it, make sure expand image is selected, and you align using the middle sunspot. The resulting stacked final will contain both the left and the right of the sunspot.

Such trick should be useable on the moon as well, the good thing is:

1. one saved time to stitch
2. one saved time to adjust brightness contrast and those stuff before stitching
3. one saved time without having to derotate the frames
4. one can 100% sure about overlapping

I shall try more to see if it really works, and also, I believe instead of from left to right, one can even go up and down, given a common patch is available for alignment during the stacking process.

Saturday, December 10, 2005

Short sort of observation

Has been cloudy the whole day, but sun breaks the cloud at around 16:00 and I can get a gimpse of it shortly before it's blocked by buildings.

The Autostar cannot locate it well, seems like it's not very accurate to put Sun as an asteroid, anyway.

The Sun is very quiet, with a small sunspot, but when tuning the t-max, I can see several nice filaments on the surface, I can also see a plague somewhere, and it should be the remain of 826 (?) forgot the number.


Still got to find a method to balance the SM40 on the ETX.

Thursday, December 08, 2005

Mosaic failed

But here it is.


1. Capturing: use all the same settings, make sure exposure are the same. capturing from bottom to top, to avoid backlash, start in the east and match to the west; after completing a series of the same DEC, adjust DEC to capture the next series.

Allow more overlapping. I was too greedy and aggressive at first and thus the first strip cannot be merged to the rest...

2. Processing: turn off predict track, since the subtle detail on the solar disc is better non-tracked than tracked. disable stretch histogram, expand image can be left on or off. Finally, wavelet process the stacked result briefly, but remember to use the same amount of wavelet profiles. This step will allow stitching software to work more effectively.

3. Stitching: stitch all at once, don't do it piece by piece.

Wednesday, December 07, 2005

AR826 on 1203 and 1204

Some results

Seems like I shall pay more time in this blog than in forum, here's one of the surface detail combined with prominence.