Clear Night Skies

Water tightness and other musings

2009-09-01T13:34:00.001-07:00

After building the observatory, I eventually got around to making it water tight. I just got sheets of rubberized canvas and made skirts around the roof's 2x4 frame so that the skirts cover the gap and extend a bit over the walls on all four sides. Upon water testing this with a garden hose it held up pretty well. Unless we have a rain with winds strong enough to make the rain come close to parallel to the ground, there should not be a problem (and this never happens in california).

The base of the walls, where the frame meets the asphalt of the roof was also completely sealed using roofing cement on all sides. This makes sure that water does not seep inside from the bottom edges.

I also noticed that I needed a bit more ventilation and added two more of those solar fans from solarwholesales.org. This time I added them at the apex of the roof so the air is being circulated by two fans in the center and two fans near the edges. This seems to be working very well as well. We've been having a few 100 degree days lately and the insides easily stay 15 degrees or so cooler than the outside.

Also for a while now, I have been contemplating making the system I have (Celestron C8 on an Atlas EQ-G mount. QHY8 and Hyperstar III) a little more versatile. Of course I could always operate the scope in f/10 config by attaching the camera to the back, but with a f/6.3 reducer/corrector, I could also operate it in that mode effectively giving me 3 different configurations. A quote from the Celestron website on the f/6.3 reducer: "For the Celestron reducer with a focal length of 284mm working at f/6.3, the spacing is 105mm. For Celestron accessories simply screw the reducer onto your scope and view at f/6.3. For aftermarket uses the compression ranges from about 0.7 at 50mm spacing to about 0.5 at 225mm spacing." Given this, I got the following items:
1) Celestron f/6.3 reducer/corrector
2) SCT to T-Thread adapter. Got the 55mm version
3) 10mm T-Thread extension

I already had a 20mm T-Thread extension and from the QHY dimensions

I have a total spacing as follows:
55mm Thread Converter + 20mm Extension + 10mm Extension + 20mm QHY8 = 105mm spacing between Corrector and CCD Chip. The QHY8 has 7.8 um x 7 .8 um pixels with 3110x2030 total pixels and 3032x2016 active pixels. Using this calculator and the active pixels, I get the following:
At f/2 Hyperstar setup: 16" FL == 40.64 cm FL == 406.40 mm FL == 199.8' x 132.8' FOV
At f/10 Visual Back setup: 80" FL == 203.20 cm FL == 2032.00 mm FL == 40.0' x 26.6' FOV
At f/6.3 FR Visual Back setup: 50" FL = 127 cm FL = 1270.00 mm FL = 63.9' x 42.5' FOV

I could also change the spacing using the various spacers I have and get these additional combinations ( comp = 1 - Spacing/FR FL. For Celestron f/6.3 FR FL = 284mm):
75mm = 1 - 75/284 = 0.74x
85mm = 1 - 85/284 = 0.7x
95mm = 1 - 95/284 = 0.67x

Which translates to:
At f/7.4 FR Visual Back setup: 59" FL = 149.86 cm FL = 1498.60 mm FL = 54.2' x 36' FOV
At f/7 FR Visual Back setup: 56" FL = 142.24 cm FL = 1422.40 mm FL = 57.1 x 38.0' FOV
At f/6.7 FR Visual Back setup: 54" FL = 137.16 cm FL = 1371.60 mm FL = 59.2' x39.4' FOV

I am not sure if these additional configs are worthwhile to attempt, but it's interesting nonetheless. Maybe I can use the FOVI in Sky6 to see what suits best. Although, I should point out, the Celestron documentation indicates a different range for the same range of spacing.

Observatory

2009-07-08T13:03:00.000-07:00

Been on a hiatus for a while now, but meanwhile I have been scheming to construct an observatory. The detached room we have in the backyard provided for a very good skyline and height to build this. After researching on the web, I pretty much narrowed down my decision to something like this (or this). After talking to Richard Shell from STI, and looking around for the right shed, I finally decided on an Arrow Newport Shed from HomeDepot which was on sale for 230$. Richard suggested that instead of the 2x4 and 1x4 rails, he is now using garage door rails and it works much better. But this would add some more engineering to the project and so I decided to do this with 1x4s and 2x4s with castors.

We were getting a new roof constructed on the detached roof and my roofer was kind enough to listen to what I needed and built me a counter leveled platform for my shed. He also installed braces for the 4x4 columns. The platform were built with these dimensions (this is from the shed manual):

The platform and brace for the columns eventually ended up looking like this:
Platform from the front
Platform from the back
4x4 column brace

The platform is counter leveled to correct for the slope in the roof and make it flat. And the braces are leveled as well. The height difference between the two braces (because of the slope in the roof) was corrected for by using different length 4x4 columns. Anyways before we get ahead of ourselves, I had to buy all the lumber. braces, castors etc. for the roof frame, the rails and the columns. I drew up this design before acquiring the lumber:

This changed a bit as I started building, but the roof frame was pretty much spot on. As you can see from the above, the roof frame and the part of the rails on which the castors ride are made of 2x4 bars. The side stops on the 2x4 rail is made of 1x4 bars and the rail support in the back is made of 4x4 columns (attached to the 4x4 brace in the bottom)

Basically, when constructing the shed off of the instruction manual, you just skip the parts where the roof is affixed to the walls through the wall channels. Instead the roof is constructed in completion and attached to the 2x4 frame using wood screws. The rest of the building (base frame, walls, wall support channels, doors etc.) are again constructed according to the manual. One the walls are all standing up, you build the rails on top of it and then place the roof on top of the rails. The rails were originally going to be build with a complete four sided frame with the two side portions running further back so the roof can roll off to the back. But then I realized that I did not need all four sides. In fact I only needed the two sides on which the castors would roll. So eventually I just ended up doing this. Also the length I used finally for the rail 2x4s was 139 inches as opposed to what I had in the diagram to begin with. On either side of these 2x4 rail supports (which are laid flat on the roof, screwed with wood screws through the side wall channels and supported and screwed into the 4x4 columns in the back), I used 1x4 bars to provide for wheel stops, so that the roof does not slip off the 2x4 rail by moving sideways. This creates a depressed channel in each rail in which the castors ride. Blocks of wood are also screwed in these channels on either end so the roof can only move so far in either direction
The rail design using the 2x4 for castor support and the 1x4 for wheel stop on the sides forms this depression in which the castors ride
Here are the end stops which make sure the roof does not travel more than it should. 4 such blocks on the 2 rails make sure that the roof only goes so far in either direction
Castors sitting inside the rails
The 2x4 castor supported is screwed on to the wall channels from underneath using wood screws

I then built a pulley system using nylon ropes and metal rope eyes so I could pull the roof on and off from below without having to climb up on the roof. Basically it uses two pieces of rope on either side so that there is no rope in the center obstructing the view and uses metal eyes and rope knots to acheive the right forces to move the roof in either direction. Each side is composed of 3 eyes. One of the roof frame (eye 2) in the front and two on the rail system one in the front (eye 3) and one in the back (eye 1). The rope passes through the eye 1, enters eye 2 where it's knotted and then passes through eye 3 and then back out through eye 1. These four ropes then pass through two additional eyes on a beam fastened between the 4x4 columns and reach the ground in the back where they can be reached and pulled on. When the top ropes from the two sides are pulled, because of the knot on eye 2, the roof starts rolling off. when the bottom ropes on the two sides are pulled, eye 3 directs the force forward because of the knot in eye 2 and hence the roof starts rolling forward. It is important to note that when the roof has completely rolled on eye 2 should be a little behind eye 3. The forces act in such a way that when rolling the roof on, the forces on eye 2 act horizontal only as long as it is behind eye 3. Once it gets on top of eye 2, the force becomes vertical and the roof won't move anymore. Some pictures might help:
A full view of the ropes inside the building. This is the right side looking from the front of the building
Eye 2 and eye 3 setup when the roof is completely rolled on. Notice the knot on eye 2. This is what lets the forces direct the roof on and off
Eye 1. This eye is used to just direct the ropes out of the building while staying on the edges
These eyes in the back take all four ends of the ropes and direct it down from where they can be acted upon. You notice that ropes from one side of the building go through one eye. The two eyes here are centered so that there are no lateral forces when yanking on the ropes. The length of the two pieces of rope from either side (top part or bottom part) are tied together so they form two coherent pieces, one which moves the roof off and one which moves it on
A view from the back of how the ropes travel behind the building
Another view
A view from down below where the ropes will be used to roll the roof on and off

And here's with the roof off and on:

When the roof is off there is a pretty good, clear view of the sky, and the telescope should be able to slew to it's limits through the zenith. The view to the back is not much limited either (the horizon is above the level to which the rolled off roof obstructs the view and the roof itself is not very tall, even with the castors and the frame. Atmost 6~8 inches)

Another idea I had (what with 100 degree days in NorCal these days) is to have some forced ventilation in this observatory. I found these solar powered vent fans which were a great fit. I installed two of these in the center roof panels after cutting the right holes. Caulked it afterwards to be water tight. They work great!
The solar panels can be seen on the outside
view from the inside

A full view of the observatory:

And a video of it in action ;)

Also to control the power on/off remotely, I am using this wireless power controller in the observatory. Using this I should be able to turn everything on/off remotely without having to get up on the roof

Also the insides of the observatory will be lit with this work light and red CFL spiral, in case I do need to go up there to tweak something

One final problem I need to figure out is the water tightness of the building for realistic rainy situations. The sides which have the rails should be fine:

But the front and back might not be:

One of these days I am going to climb up there with a garden hose and see where I need to do some work to block out rain. Most probably some plastic sheets will do the trick, but I need to test first. Of course I will have to do this before I put any equipment in there :)

All in all this project turned out pretty well.

Heart to accompany the Soul

2009-01-25T23:04:00.000-08:00

It's a beautiful thing. This is 16x10min subs. 60 Flat frames and 40 bias frames. I took the Flat frames with 0s exposures in Neb2 and so used the bias frames as flat darks as well. I think the flats are coming out nice, as evident in the lack of a central halo.

But this image just has too many stars in it!! I don't know how to reduce them. Also I wonder how can I do a better job of data acquisition for this. Should I go longer on each exposure?? or do shorter, more numerous subs? Light pollution is definitely screwing me over to some degree. But there is a glimmer of hope.

ReSoul

2009-01-20T13:25:00.001-08:00

After some feedback from Sander about there being enough data, and to just add processing, I decided to take another stab at this. With the same data as the previous post. And here's the result:

I used the Red Channel as a luminosity channel and some iterations of Histograms, Curves and Noel Carboni's "Enhance DSO" script. All in all very happy with it. Thanks for the suggestion about processing Sander.

Soul

2009-01-20T12:03:00.000-08:00

After cleaning the corrector, remounting it etc. I decided to check if everything was ok yesterday night. Tried to image Soul Nebula (IC1848). Here's what I achieved:
My flats have gotten much better as can be witnessed in this image. I also too Flat Darks this time around. So here's the skinny on this: 17x10min light frames, 60x0.1s Flats, 40x0.1s Flat Darks and 40 Bias frames. All in all a pretty decent image assuring me that things have not gone horribly wrong. The collimation looks pretty good as well. Although this image probably needs much longer exposures time on each sub frame. Will have to come back to this target.

Fiasco

2009-01-19T20:28:00.000-08:00

Yesterday, in an attempt to bring the sweet spot back to the center (as noted in the previous post about flats etc.), I tried recollimating the Hyperstar. After an hour or so of frustration, and lost collimation etc. I realized that the threads of the hyperstar had stuck to Hyperstar conversion kit in the corrector plate. I guess this was because of my attempts to get a better angle for the collimation screws on the Hyperstar. The collimation screws on the Hyperstar really need to be more accessible. They are hard to reach, especially when using a camera such as the QHY8 which has a profile which covers the screws from the top. Anyways, this whole thing meant that I now had to remove the whole corrector plate and rescrew the baffle tube to the conversion kit ring. What a hassle! But the silver lining is that since I had the corrector out anyways, I ended up cleaning it pretty good. One other thing that was stuck was the nose piece of the QHY8. I guess I lost whatever seal it was providing to the CCD. Now I will have to see if frosting problems will rear their ugly head. Finally as of tonight, I am back to my old IDAS-LPS-P2. And the Hyperstar seems collimated and I think I have the corrector centered properly. So the sweet spot should be back in the center. To collimate I started with the hyperstar completely flat and I need very little effort to bring it to collimation. One other thing that I ended up fixing in the process is the way the Robofocus was attached to the scope. I was using 3M mounting tape, which was rated for 2 pounds. I went to Home Depot and got Scotch mounting tape rated at 5 pounds and the robofocus seems to be a lot more stable now.

As I type this I imaging IC1848 (Soul Nebula). Hopefully it will turn out good enough to post. We will see...

Steep light paths and narrowband Light Pollution filters...

2009-01-18T11:15:00.001-08:00

Recently I bought the Astronomik UHC filter to try and mitigate some of the light pollution problems I had been having. I have not posted a single image that was taken after obtaining this filter although I did do many runs of data acquisition. The problem being that in even data acquisition, after stacking them, I notice these diffraction color bands. At first I thought it was because of the flats, which I corrected as indicated in my last post. And even after that I notice the bands:

After some discussion on the QHYCCD yahoo forum, turns out that this is because of the steep light paths the the f/2 hyperstar system entails, combined with the large CCD chip used to image. The data is there as can be seen from the red channel:

The other channels show severe banding. Green:

And Blue:

As can be seen, the flat looks ok but is just not able to correct for these diffraction bands:

And the bands make it very hard to correct for it and bring out the details. People indicated that the Baader 35nm Halpha filter migt do a better job. The reason being that the steep light path basically shifts the wavelength by about 6nm and this leads to diffraction patters resulting in the color bands. The Baader 35nm, being pretty wide for a Halpha filter still captures the Halpha even though the shift happens and hence should preserve the data while cutting out all the light pollution. I tried stacking them without flats as well, to rule out the flats and the same problem exists. But oh well!! After 200$ for the Astronomik UHC, I think I am back to using the IDAS-LPS-P2

One thing that can be noticed here is that the chip seems to be offcenter. This was indicated by Sander after seeing my flats. I think the collimation effort pushed it a bit off-center. I can try and recollimate to bring it back to the center, but for now I think I am going to continue with this setup and the LPS to see what kind of results I get.

Analyzing the flats...

2009-01-17T14:51:00.000-08:00

After getting the Astronomik UHC, I spent some time training foxusmax for the new optical train, with dismal results. I realized at this point that the screws on the Hyperstar were a little loose and my collimation was way off. Spent about half hour day before yesterday collimating it with good results eventually. But then noticed with the data I captured of IC1805 that my flats might not be optimal and might be contributing to the destruction of my images:

The values are clipped and the flat is way too bright. The exposures were already 0.02seconds each so, going low on the exposures was not a reliable option. So I got milky acrylic sheets instead to dim the EL Sheet I was using in my lightbox. Subsequent to this the flats look a lot more well behaved:

Using the Flat ADU calculator on Starizona, I need to have around 26000 ADU for my flats. And in this image you can see that it is very close to that. The exposure now is 1.6 seconds and lot more reliable with headroom to experiment a bit.

More learnings...

2009-01-04T17:15:00.000-08:00

By default Neb2 debayers the fits before saving it. This makes the fits files 3 times bigger. This is definitely a problem. I do not know how this affects the calibration, alignment and stacking in DSS!! Anyways, I gave another go at grabbing the data of NGC2237 (Rosette nebula) with Neb2 saving the data as raw files instead. And here's the result:

Overall I am very happy. But a few things I need to correct next time around. This data acquisition was originally 48 frames of 4 minutes each. But I could only use 35 frames since the focus shifted pretty severely during the run. and beyond the 35th frame it was completely out of focus. I need to setup focusmax optimally and try and use the temp. compensation feature. Calibration frames were 60 Flats and a master bias breated out of 60 Bias frames. Also, although the IDAS LPS is helping a lot, there is still a lot of light pollution gradients in my image and I have an Astronomik UHC on order. Hopefully that will help. We will see in a few days...

Horsehead and Flame

2009-01-01T10:57:00.000-08:00

After a couple of more attempts I realized that there might be a problem with the way I was grabbing the data, so I bought Nebulosity 2.0 and tried grabbing the frames with it and I think it came out well. This is 25x2m light frames with 60 flat frames and 60 bias frames.

One more thing I realized is that I definitely needed longer exposures for most objects. I was doing 20s and 30s exposures but longer would definitely help. I stacked with DeepSkyStacer using AHD for debayering, sigma clip for calibration frames and auto adaptive weighted average for the light frames. I am pleased with the results. Obviously there is a lot of improvement to be had, but this is a great start. The optical train is Celestron C8 + Hyperstar + IDAS LPS-P2 + QHY8.

Horsehead Nebula is a dark nebula and the Flame Nebula is an emission nebula. Both very beautiful targets.

Post LPS and Light box

2008-12-19T10:21:00.001-08:00

I got my IDAS LPS-P2, mounted it on the hyperstar and started shooting. I also now had the light box for flats. And here's the first attempt:

This is the great orion nebula. 27 40s light images. 30 flats (with max pixel value at 22000 ADU) and 30 bias frames. I think the flats helped a great great deal!!! The lightbox works like a charm and is really easy to use. The LPS has made pollution gradients much more easy to handle, but there is still a local gradient visible. I think that for this target, given the trapezium, I needed to do a lot more light images (maybe 60 or 100) at 20 seconds each and that would have given the SNR that is desired. Also I am wondering if I should use a narrower LPS given the light pollution in my area. Pixinsight was used to do masked stretches and SCNR noise reduction (The image had a distinct green tinge after debayer). Stacking was done using DSS. All said and done, I am pretty pleased with the progress. Hopefully the next time I get a chance, I can correct some of these acquisition issues and come up with a much better image.

My Imaging setup

2008-12-10T10:45:00.001-08:00

I finally have all the components I need for a decent imaging setup:

Everything mounted on my Atlas EQ-G. The main imaging scope is a Celestron C8 with Hyperstar and QHY8 and the guidescope is an Orion ShortTube 80 with Orion Starshoot Autoguider. The dew shield helps greatly! Especially since the camera projects outwards and upwards. It helps aids in mounting other things such as a light box, since there is no other easy way to do this with a hyperstar setup.

This setup is still much lighter than what I used before. I can getaway with two counterweights. With my XT10 setup I would have to use three 5 pound counterweights and they would be all the way at the bottom of the counterweight shaft.

The dew straps are mounted on both the short tube and the C8. On the C8 I am mounting it right behind the plastic endpiece holding the corrector. Contact with the metal should give better results overall.

All the cables have been split into two groups: Power, which is one bundle, and Control, which is another bundle. Show above is the power bundle and the control bundle. And below, both cable groups are secured by tieing them around the handle bar of the C8, with enough play so that the cables don't get taught when the scope moves around.

Focus control is through Robofocus. The new C8s have a curved back and as a result mounting the robofocus was a pain. Eventually I had to get the plastic wedge and the expansion bracket, seen here, from Technical Innovations. The expansion bracket can be extended in height and I collapse it all the way in when storing the scope. When on the mount, I extend it all the way which is just enough height for me to mate the robofocus to the fine focus knob of the FeatherTouch via a sprocket and a timing belt. I would much rather have the mating be done with the Coarse focus knob of the Feathertouch, but I don't have the right sprocket. I bought what was touted to be the right Sprocket from Technical innovations, but it does not fit and they have not sent me a replacement yet (4 weeks and waiting...). Maybe I will just trim the inner surface of the replacement Sprocket I got and use that on the coarse focus knob (the inner diameter of the replacement I got is just a tad short and hence does not go over the coarse focus knob). This should let me achieve faster focus since with the finefocus knob the step size has to be set too high for any resonable speed in focus.

Without the dew shield you can see the Hyperstar setup. I does look wicked cool!! All the cables around the tube are held losely using a welcro strip. Also I secure the power box for the QHY8 and the Robofocus control box on either side of the C8 using welcro. They stay pretty secure this way.

All the power bricks sit currently on the eyepiece holder in the center of the tripod. This way there is no excess tension on the cables as the bricks are pretty heavy. But I need to figure out a better solution for this. Maybe a bag which hangs on one of the legs in which I can keep the bricks as they are plugged in. Yet to devise this

Light Box

2008-12-10T10:23:00.000-08:00

So I found out about these things called ElectroLuminescent Panels on the QHYCCD Yahoo forums and it provided for a convenient way for me to build a light box. Bought an A3 size EL panel from glowhut.com, which was shipped quick and courteously. Went to Michaels and bought two foam boards (16"x20") and made this:

I cut the foam board crudely, to be slightly bigger than the A3 EL Sheet. Cut a hole in one of the foam boards such that the dew shield fits through it snugly and then sandwiched the EL sheet in between the foam boards so the black side is outside and stuck the foam boards together with super glue. Voila! there's a light box!

It provides nice even illumination...

through the dew shield :)

The depth provided by the foam board with the hole is good enough to hold the board even when the dew shield is tilted and hence I should be able to mount this on the dew shield while the telescope is mounted on the mount. Just have to make sure that the telescope is pointing towards the zenith. I am excited to try this out. My primary problems so far have been light pollution gradients and bad flats. This should definitely solve the latter.

Light Pollution Filters

2008-12-04T10:16:00.000-08:00

Trying to figure out what filters to get is like pulling teeth!! So many opinions on the net. And so much data as well. Mind numbing stuff. Anyways, after enough digging around I had sort of narrowed down to two: either the IDAS LPS P2 or the IDAS LPS V3. I eventually found this article on a spectrograph analysis of the various filters and also a live comparison on M42. Seems to indicate that the V3 is narrower than P2 (which is apparent from the product brochures) and might work better in situations where light pollutions contributors include high pressure sodium lamps. I live close to San Jose CA where, as you can see from the wikipedia entry, a conscious effort has been made to use only low pressure sodium vapor lamps, because of the proximity to Lick Observatory. As a result, at this point I am leaning towards the P2.

New look, new setup...

2008-12-02T19:42:00.000-08:00

I know, I know! It's been a while. Too many other things happening in life and I got a little busy and distracted ;) Anyways, now I am back, once again, and have a completely new setup!! You might remember that initially I used to use the following:
1) Atlas EQ-G Go-to mount controlled via EQDIR and EQMOD
2) Orion XT-10 telescope tube with Feathertouch focuser and Robofocus
3) Starlight Xpress MX7C single shot color for both imaging and guiding (in interlaced mode)

I upgraded a bit :) The mount is the same, but the rest changed:
1) Celestron C8 with Hyperstar modification, Feathertouch Micro SCT focuser and Robofocus
2) Hyperstar
3) QHY8
4) Guiding with Orion Short Tube 80 and Orion Starshoot Autoguider

I will upload some setup pictures soon. I feel that I have a much cleaner setup at this point (even though it's more components). All the cables are neatly tied up etc. As far as software goes, only one additionl piece of software: PHD Guiding. But of course the laptop setup is completely new. The old one blew up, and I don't have a windows machine anymore. So now I am running everything on a copy of Windows running inside Parallels on my Mac Book Pro. It works just fine! In fact in some ways it's better!

Initially I had some problems getting the QHY8 to work. I kept thinking I could not focus, but turns out the QHY8 was not capturing at all. After some emails and web searches, I figured out that the 4-pin DIN (S-Video) connector supplying power to the camera electronics and CCD was not seating in properly (QHY8 has to power supply cables coming from the adapter to the camera. One for the cooler and the other supplying power to the camera electronics and CCD). After some cable swapping and jacket-trimming (Shaved the plastic jacket on the power cable and now it seats properly), it started capturing. Set it up that night but took some time, reestablishing parameters for robofocus, Maxim DL etc. The data I acquired was too noisy and I didn't have flats. So abandoned them. The next night I set it up again and images a few other targets:

M33. Leaves much to be desired in many areas. 52x30s. Flats, Bias and Dark were used but flats were not too good. Also after calibration I used the "Auto Flatten Background" Filter in Maxim DL and it messed it up a little more...

This is the Pleiades star cluster. Also called the seven sisters. The cool thing about this is that the solar winds are blowing back the gases and you can actually see the ripples in the gases around these stars!! Really cool...The image was 49x30s. Flats, Bias and Dark were used but flats were not too good.

The processing is a little hasty and also I need a better mechanism to take flats. With the big chip and the fast system, there is some severe vignetting. I might even benefit with a Light Pollution filter in the optical train. The next time I image, I need to have better flats. Hopefully I will get some more time this week.

United we stand, divided we...

2007-03-21T10:01:00.000-07:00

This data was collected over the previous weekend. After the end of this data collection my laptop conked out and lost it's mind. Won’t boot anymore. The image is 7x10 light and 12x10 dark frames. The flat and bias were collected on my Mac then using iCCD but I screwed up and the data is invalid now. Cannot open them anymore. The processing is very basic and I took a new approach and am very happy with it. Instead of splitting the frames into LRGB and then processing each channel individually, I calibrated and combined the light frames without splitting. I then converted it into RGB in Maxim and brought the tiff file into Photoshop and processed the various channels. The results are very pleasing! There is still some coma in this image which should be taken care of by the MPCC which I am yet to receive. Also I would probably benefit with some stretch and DDP in this image to tease out more of the details. All said and done, I think I will be processing my future images this way as opposed to splitting them before working on them.

Return of the Astro-Jedi

2007-03-15T20:02:00.000-07:00

So, I have been away for almost 2 months. I was off getting married and on my honeymoon etc. A much needed break from the daily grind. Now I am moving in with my wife and long-time girlfriend into a single bedroom apartment :( Which means I won’t get as much face time with my tele anymore. But, over the span of these two months, I did get robofocus and feathertouch focuser installed on my trusty reflector. Make a huge difference in the focus. First, it’s a much shorter and more pleasant experience. And then, it’s spot on!! Just look at some data I gathered while playing around. I am using focusmax to achieve critical focus. Some images of the focuser and the motor will come soon.

This image is 12*10min light and dark images. No flats. I was playing around with robofocus and wanted to get some data to see what improvement it brings and wasn’t really trying to get really good data. But it’s obvious with some better data and calibration this image would shine. The focus is really good and very little coma. Although a coma corrector and a light pollution filter would really help. This also happened to be the march challenge for the tac-imaging mailing list.

Maxim DL+EQMod = Flame

2007-01-15T10:27:00.000-08:00

I got my EQDirect mod from Shoestring recently and I now have the capability to control the telescope completely from my Laptop and doing PulseGuide through EQMOD Ascom driver. Took a bit of experimenting with the settings on Maxim DL and a bug fix to EQMOD before I got this working, but once I had it working, this was the result. This is a 3 hr exposure (18 * 10 min exposures) with equal darks. 50 4 sec flats were taken with 50 bias frames to go with it. The clarity and details are far better than my previous attempts.

NGC2403

2006-12-31T17:06:00.000-08:00

This is not a very good representation of this object. But this is the best I could given the data I collected. I was struggling with Maxim DL and it’s settings when I was trying to image this. This is a 55 minute exposure (5 min * 11) with equal number of dark frames but no flats. I need to revisit this object sometime soon. This is a fairly bright and big galaxy in the local group that Charles Messier missed!

Eskimo Nebula

2006-12-30T10:33:00.000-08:00

I just don’t have the right FOV for this object with the telescope/camera combination I have right now. This is a 3 hour exposure (5 min * 36) with an equal number of dark frames. No flats. This is a beautiful planetery, but you can’t see any detail in this image. Maybe I should use a barlow or something.

Horsehead Sequel

2006-12-25T02:35:00.000-08:00

I gathered data from scratch. This time there was no moon, and I did this in an attempt to participate in the december challenge for tac-imaging mailing list. I think it worked out well. The processing worked good too. 20x3 min exposures totaling 2 hours. Equal number of darks. Flats were collected the next morning. Alignment and calibration in Maxim DL and final RGB composition in PS CS2. Another thing I realized while processing this is that the three AA plugins I had to split the exposures into LRGB were not all doing it the same way. One produced better RGB frames and one produced better L frames. I used both plugins. Deconvolution scripts used in Maxim DL to make the stars sharper. The color balance is probably also closer to reality although this is still a learning process for me.

M1 Reprocessed

2006-12-21T14:37:00.000-08:00

Since I could not manage to do any imaging for almost two weeks now, I decided to reprocess the M1 data I had with the IR filter. Same data as the one posted on December 11th, but this time, I used better black and white point setting and also deconvolution scripts on the L master frame using the scripts developed by Robert. This did wonders. I used to cycles. The first cycle was gaussian kernel with 5 iterations and 4 pixels. The second cycle was again a gaussian kernel with 3 iterations with 2.5 pixels. In both cases the Standard deviation on the background was 6 (derived from the information panel of Maxim DL in aperture mode). This gave me a much sharper L master. The LLRGB was combined in PS CS2. Using some basic curves on R,G and B frames (from Ron Wodaski’s tutorial) and also black and white point settings. I also used DDP script for photoshop. Layer 1 was 40% opaque. You also notice that with the deconvolution all the doubles are also split nicely. There is some halo around the big stars and that’s probably because I pushed the deconvolution too far. I think the color balance is also better on this one. Although I still have no clue on how to do the color balance right.

M1 with IR Filter

2006-12-11T17:23:00.000-08:00

Got me an Astronomik IR cut-off filter. And I think it improved things a lot! As a first try, M1 was targetted. This is a 20x150 sec exposure with equal darks. Flats were taken the next morning at dawn (this is one area I don’t know if I am doing the right thing yet) 130 x 15 sec flats corrected for Bias calculated from a median of 10x0 sec bias frames. Again I had some field rotation. I need to do a Drift alignment one of these days and at least nail the altitude bolts. Lateral adjustment should then be ok. Anyways, the subs were aligned with Maxim DL after calibrating in Astroart 2.0. I like Maxim DL, I may end up buying it after the trial period. Need to investigate if Maxim DL will work as my image acquisition software too.
Notice that even though focus is off just by a little bit (I think I am getting better though) the stars are are nice little circles :) This is a big improvement over my previous images. You do notice a lot of noise in this image, partly because of poor seeing conditions and also because there was an 80% moon out when I did this. With dark skies, better seeing conditions and better focus this will get better.

M57 Once More

2006-12-11T14:35:00.000-08:00

With a slightly better understanding of the techniques involved. I attempted M57 again. This is by far my favorite object. Biggest bang for the buck so to speak :) As you can see it’s a much better capture composed of 20x2 min exposures and an equal numberof darks. 150x10 sec darks and dark flats.

But notice the stars. They are not focussed properly and also are a smudge with a lot of bloom. Two things: Robofocus and IR Filter :)

Horsehead

2006-12-11T00:04:00.000-08:00

On the same night that I was experimenting with my new Astronomik IR Filter, I tried the Horsehead nebula (IC434) after shooting the M1. This is a 25x150 sec exposure with equal darks. Flats were taken next morning and consisted of 130 x 15 sec flats corrected for Bias calculated from a median of 10x0 sec bias frames. Again the stars look good. There is even natural diffraction spikes :) There is a lot of noise and graininess here, because the moon was very close to the Orion Constellation when I shot this and seeing and transparency was not great either. And I KNOW that my flats are not good at this point. You can see some dust leftovers on the left of the image. One thing that I would like to change is the FOV. My current setup has a very limited FOV.

IC342 - Better Alignment

2006-12-01T10:49:00.000-08:00

The same data. Alignment is done using CCDStack instead of Registax and combined using Sigma as opposed to Sigma. As you can see, the details are better defined. Again, this is a 20x10 min light and 20x10 min darks stack. 150x10 sec darks and dark flats.

IC342

2006-12-01T10:48:00.000-08:00

This is a hard object for me. A galaxy in the local group. It’s fairly big, but very low surface brightness. And the light pollution in backyard is high enough that I cannot get a decent data set to build a good image. This exposure consists of 20x10 min light and 20x10 min darks. 150x10 sec darks and dark flats (although they may not be that good). In this exposure run I was plagued by field rotation and it was hard to align the images and retain the information. Hence you see the left is lighter than the right. In fact I did another exposure set with 4x30 mins and it turned out to have too much noise.

NGC2024 - Take 2

2006-12-01T10:47:00.000-08:00

Got it centered this time around. It is really a bright object and beautiful to image. 20 x 3 min shots with equal dark frames. 150x10s flats with equal flat-darks. All sub were combined in Ray Gralak’s Sigma and LRGB composition in PS CS2. This is in fact an LLRGB image where Luminance frame was used twice to improve details. Although it might be a little dark because of incorrect S curves. Sharpening and Guassian blur was used with some success to improve details. The color contrast adjustments also seem to be getting better.

NGC2024

2006-11-25T11:53:00.000-08:00

I was actually shooting for Horsehead and ended up on the flame nebula :) Also the pointing was bad, because you can only see a part of the flame. The coloration might be un-true too. This is 20x10 min exposures with a 20x10 min darks. 150x10 sec flats and flat darks. I processed three different times and this one is with Sigma, Astroart and PS

Ray Gralak's Sigma

2006-11-25T11:47:00.000-08:00

Same data, except I used Ray Gralak’s sigma to process the subexposures. The earlier image was with astroart processing the subs. The improvement I believe is primarily because I handled photoshop better when adding the RGB frames to the luminance and not because of Sigma, although, the data I have has enough flaws in it for Sigma to make any significant difference. I feel Sigma is a surgeons instrument, while my skill level at this point in collecting the raw data is akin to that of a butcher :)

M42 in Color!!

2006-11-25T11:30:00.000-08:00

Picture of M42. After significant processing here’s what I have. My processing skills are improving but still have a long way to go. You can still see a lot of blooming stars and also the fuzziness is probably because my R&P focuser does not do a good job. And maybe my collimation was off too :(
This is a composite of 20x10 min exposures. 10x10 min darks were taken for noise and 150x10 sec flats and flat-darks were used. Didn’t use bias. But maybe next time I will use them and see if it makes things any better. Bias information, though should be contained in the darks itself and should be sufficient since the darks are of the same duration as the light images.

Second light - Take 4

2006-11-08T21:27:00.000-08:00

Another failed attempt to locate an object. I was looking for M33. Couldn’t find it. It may have been too big to fit into the FOV of my camera. This looks like one of the spiral arms of it though, doesn’t it? 3x10 min exposures with 1 10 min dark. No flats.

Second light - Take 3

2006-11-08T21:07:00.000-08:00

Can you see the small galaxy there? I was actually searching for Helix nebula. Didn’t find it, but somewhere along the way I found this :) I have no idea what this is or if it is even cataloged. I think though, that this might be NGC7184. This was 3x10 min exposures with 1 10 min dark. No flats again.

Second light - Take 2

2006-11-08T21:04:00.000-08:00

What should follow M1? M2, of course :) This is a 3x10 min exposure with 1 10 min dark frame. I think I should take as many dark frames as exposures and for just as long. Color balance is again off. I have to learn how to do that right...

Second light - Slightly Better :)

2006-11-08T21:00:00.000-08:00

Read up a little and I think it paid off :) Found M1 after some hopping and slewing. 2x20 min exposures, 1 30 min exposure and 1 40 min exposure. 1 10 min dark frame and no flats. I should start taking flats. I think they will help. I have been using the artificial flat plugin for astroart from www.astrochris.com. I should probably keep the exposures lower too. The stars bloat up because of the long exposure. Then again that might be an artifact of poor focussing and improper collimation. The more I do this, the more I realize that everything has to be perfect. When I combined this in PS, I think I screwed up the color balance. Oh!! BTW, this is the first time I saw M1 ;)