Thoughts from Starlight Cascade

This was the single best imaging run of 36 runs on a cold evening.
The shadow transit is from Io, the top right moon is Ganymede. Io itself is on the surface of the planet, not easily visible.
Seeing was poor, transparency was poor, and a more than quarter moon was nearby.
Exposures were about 2ms and this was the best 15% of 15k frames. The workflow was: firecapture, autostakkert!, registax, imagemagick, gimp. Not included here is the further processing using winjupos. That will come along shortly.

Below is an animation of the 36 runs of 180 seconds each, totalling 108 minutes

and after even yet more time processing… this is the result of running 36 imaging runs through winjupos to derotate. It was also put back through gimp for black levels, colour saturation and unsharp process (right image).. the left image is the closest timestamped image before winjupos processing.

These results are inconclusive regarding better signal to noise and a better image.

We have a new-to-us Celestron C9.25 Schmidt-Cassegrain OTA and as everyone knows, the front corrector plates on SCT’s are extremely prone to dew. So we have had an external strap heater around the front end (image c925a.jpg).

Unfortunately this strap heater moves around, comes loose, prevents the metal cap from going on, and makes it very difficult to add in the foam dew shield. So we got the Celestron Dew Heater Ring, which mounts on the inside.

Step 1: add a paper safety on the glass (image c925b.jpg)

Step 2: remove the six Phillips head screws

Step 3: remove the existing ring, the gasket underneath

Step 4: install the dew heater ring, adjusted so the power cable is near the 7’oclock position and put the screws back in.

Step 5: run the cables into the cable holder which clips on the edge.

Step 6: Bob’s Your Uncle.

Previous to the work, I watched THREE youtube videos on the process to make me an expert:

and my favourite: https://www.youtube.com/watch?v=2blj_PDthcI

All three video demos went smoothly and the job was done in less than 5 minutes.

I made it to step 2. Step 3, removing the ring and gasket, totally failed. After 20 minutes of trying to pry out the ring without touching the glass (or breaking the glass!!!!) I stopped, reversed and put it all back together again.

I am suspecting that maybe it was that all three demos were done in very warm climates, I was trying this at -4C and that everything (metal bits) had contracted and was jammed together. The alternative is perhaps it was stuck together with adhesive. In any event, I will do some more research and wait for a much warmer day (say +10C or better) and try again.

The last image (image c925c.jpg) is what the dew heater looks like (measured 7 ohms which calculates out to about 20 watts max)

Out last evening (Thursday Nov 7) imaging Saturn.

This is the best image of the best 10% of 10K frames exposed at 18ms. Saturn was only 35 degrees in altitude, seeing was poor, transparency was poor and the was a crescent moon nearby in the sky.

Saturns diameter is approx 18 arcsec. This was processed with autostakkert, regixstax, gimp and imagemagick.

I can’t make out any belts or bands in the southern hemisphere and see only 3 regions in the north. Saturns shadow shows on the rings on the left side of the planet and the rings are dim.

So.. I was having to manually adjust the tracking about every two runs of 180 seconds, as Saturn (in an 800×800 cutout box, in a 1200×1200 region of interest) moved from one side of the FOV to the other. How bad is the tracking I wondered?

I know I have planned another polar align followed by a 2 star align in my TODO list.. but still what is a number for drift?

I went to http://astronomy.tools/calculators/field_of_view/ and entered my specs:
C9.25, x1.5 barlow asi585mc camera.
System FOV 0.18 deg x 0.1deg = 0.18 deg/3600arcsec/deg by 0.1deg/3600arcsec/deg = 648×360 arcsec

Camera: 3840×2160 pixels

So.. the drift of the cutout box (800×800) in 1200×1200 ROI moves 400 pixels in about 360 seconds or ~1 pixel/sec
3840 pixels/648arcsec=6 pixels/arcsec

at ~1 pixel/sec it will take 6 seconds to moves 6 pixels or
it drifts at a rate of 1/6 arcsec/second

Is that good? bad? Can others run the process for comparison?

Saturn is now an evening object, and since I have been using my time for the morning imaging of Jupiter, Saturn has been awhile.

Wow.. a cold cold night! was -5C and a bit more at various times.

So.. This is Saturn from just after the CFL West semifinal where Sask prevailed over BC. Yay!

This is the best 10% of 10K frames of 20ms exposures with Saturn very near the meridian. Earlier that day I also did a firmware upgrade of the skywatcher synscan hand controller… jumped up three versions and the upgrade process went well. Always a dicey decision.. firmwares have been known to brick items.

The dew strap around the corrector and the dew shield worked well. no fogging or condensation of frost on the glass all night (I packed it in around 1am.. more on that in the next note about Jupiter and the Ganymede shadow transit). This processed image was taken around 2146 EDT. The altitude was fairly low at 36 degree but I did not really see any atmospheric dispersion effects. The rings are getting dimmer and harder to make out. The moon Titan was well out of frame last night as well..

autostakkert! registax and gimp processed.

This is the 54th imaging session of 2024 in the Serenity Observatory at SCGO.
This mornings three imaging runs of Jupiter occurred in nice warm 15C weather but with a wicked wind of 7kph gusting to 20kph.
It was supposed to be over 50% cloud but we got lucky with a clear hole in the clouds for an hours or so.
Today is also the first foray into more advanced image processing.
So far we acquire a video file of thousands of ms exposure frames, open that in Autostakkert! v4 which sorts them by quality and aligns and stacks them. We generally go with the best 5% or 10% of 15K frames.
Then we take that output and put it into Registax for wavelet processing and RGB colour balance. We get a much sharper image out of that process.
Today I downloaded Gimp 2.10.38 (a free image processing program similar to photoshop) and brought the images (.png format) in one by one and applied some very basic operations on them:
1) colors, levels, black drop, applied to the background and then to the faint processing artifacts of a light ring around the planet. It disappears.
2) colors, saturation, scale up to x1.5
3) filters, enhance, unsharp mask, scale up to x1
export as .png
THe first image on the left is the “regular” autostakkert and registax output. The 2nd image, on the right, is the GIMP enhanced version.

I can certainly see a difference in the two and consider the GIMP enhanced one a value added image. However there is sooo much more to learn now!

Mars is now high enough to be out of the goop and is starting to show some nice detail.

This is the best 5% of 15K frames exposed at 1ms each using an ROI of 1200×1200 and a firecapture cutout of 800×800.
The northern polar cap is seen with the planet showing 9 arc seconds of diameter. Some darker area north of the equator and lighter areas south. Using Sky & Telescopes Maps previewer, I could not actually match any features.
Mars altitude was 65 degrees high.
These annotations are incorrect as they state the OTA was the Vixen VC200L. This was actually the Celestron 9.25.. have to update the scripts and reprocess the annotations.