Jump to content

Stargazers Lounge Uses Cookies

Like most websites, SGL uses cookies in order to deliver a secure, personalised service, to provide social media functions and to analyse our traffic. Continued use of SGL indicates your acceptance of our cookie policy.



  • Content Count

  • Joined

  • Last visited

Community Reputation

195 Excellent

About ramdom

  • Rank
    Star Forming

Contact Methods

  • Website URL
  • Skype

Profile Information

  • Gender
  • Interests
    Science Art Philosophy Family
  • Location
    Youngstown, NY
  1. Thanks Dave - yeah, that's the case here but I'm actually integrating data from two resolutions and both are undersampled but one more so than the other. So it shouldn't be occurring IMO. But I will investigate further to make sure the same histogram is being applied. --Ram
  2. Hi all, I'm puzzled by this behaviour of Drizzle and LocalNormalization. Here are the two images I'm referring to - after Integration, only an AutoSTF has been applied to make the image: Drizzle only: http://ram.org/images/space/downloads/ou4_sh2-129_Ha_nf.v0_d.resized.jpg Drizzle with LocalNormalization: http://ram.org/images/space/downloads/ou4_sh2-129_Ha_nf.v0_l_d.resized.jpg I've never seen this behaviour before and I've experimented a lot with both these algorithms. The Drizzle data is actually a bit soft (which I've never seen happen - almost always it improves things) and appears to be missing/details/decreased. LocalNormalization for the first time actually brings out details/improves things in a noticeable manner - usually I find it helps even out gradients a bit but it's never as extreme as this. Does anyone know if this is fine (I'm actually quite happy with the LN image) or if I'm doing something wrong and if there are other things like this that I'm missing that could influence the signal levels. I'm using the default parametres for both Drizzle and LN (and for most other steps - I'm using GESD Test for ImageIntegration). Thanks a lot! --Ram
  3. Thank you all! I appreciate it. Also especially Olly, you probably don't remember but a few (3-4) years back, I had just gotten back into astronomy for the first time since my teenage years and I was dabbling in lofi-EAA and I saw your stunning image of M51 and I asked you managed to do something like that and you said a few things but the main thing that stuck with me was that you put in 20+ hours of time on it to start. That was one of the key exchanges that began my move from EAA to AP to eventually AP with long total exposures and complex images. So thank you, and thank you for the inspiration! --Ram
  4. Nope, not mine but thanks. I've only finished the Squid but not the entire thing yet. I'm struggling with combining exposures of different lengths (153 x 5 minutes and 204 x 7 minutes for Ha) to get the maximum SNR. I experimented with a bunch of options using 1/4 the data and here's a sample image with 1/4 data as to what the final image will look like: http://ram.org/images/space/downloads/Image40.jpg and you'd think the same parametres and choices that worked well for 1/4 data would continue to work well for 100% of the data but no, the whole thing has collapsed. Previously I was able to use GHDRComposition to produce seamless integrations with high SNR and now it just isn't working anymore. So I'm still playing at different ways of skinning the cat until I get something I'm satisfied with... --Ram
  5. Thanks, the Bat is coming - I have a ton of data but working to integrate exposures of different lengths (7 minutes vs. 5). --Ram
  6. Also includes: v419 Cephei. Total integration: 3280 minutes/~55 hours (*632 x 300s for O3 + **60 x 120s for OSC). Cameras: *QHY163M (16mp mono) and **QHY247C (24mp OSC) CMOS cooled to -20 and -15 degrees C. Telescopes: *Takahashi FC100DF Steinheil fluorite doublet and **Stellarvue SV70T triplet apochromat refractors @ f/4.9 and @ f/4.8. Reducers: *Takahashi FC-35 (0.66x) and SFFR70-APO (0.8x). Mount: Paramount MyT. Filters: 1.25" Astrodon 5nm Ha, 3nm O3, 3nm S2 and **2" Baader UV-IR-Cut Software: TheSkyX Pro, Sharpcap, PixInsight. Inline images with reduced quality uploaded to the forum. Full resolution images of all versions are available at https://www.astrobin.com/jdaa90/ and of the primary image at http://ram.org/images/space/scope/1.4+7.4.5+6/ou4_c_orgb_632x300+60x120_3280m_55h.jpg The Squid nebula (ou4) occupies over one degree of the night sky, representing the largest angular extent ever found in a planetary nebula. Known colloquially as the giant squid nebula, it is physically nearly 50 light years across. This bipolar nebula could be one of the nearest of its type known, though its possible planetary nebula nature needs confirmation. Even though it is big and close, you're not likely to find it easily. It is an extremely faint nebula emitting primarily O3 signal and resides within the larger Flying Bat nebula (not shown here, yet) which tends to overshadow it, making it a challenging target to image. The nebula is created by the outflow of material driven by a triple system of hot, massive stars catalogued as HR8119 (blue hypergiant) as well as the pulsating variable star v419 Cephei (red supergiant) seen near the center. While their shapes do their names justice, to me, the Squid seated within the Flying Bat more resembles a hand holding an infinity stone (viewed sideways). This is the first image of my Flying Bat and Squid project/series, which became a huge undertaking for me, ending up with a total of more than 131 hours in the final integrated widefield image of both objects and a series of three main images with a few versions of each depending on the data set used for the integration. In the first image, I am showcasing the Squid nebula by itself using only the O3 filter data, with and without RGB stars. The mono Squid imaged with the Takahashi FC100DF consists of ~53 hours of total exposure in O3, making it my single longest total exposure of a target with a particular framing with a specific filter/scope/camera combination. The RGB data collected using the QHY247C with the SV70T adds another two hours to the exposure. The total integration for the version with RGB stars is ~55 hours. I also captured another 5.5 hours of O3 data using my SV70T which is present in the final image in the series but I haven't created a separate Squid only version with that integration yet but I might if I get around to it. There are six versions of the Squid only framing at https://www.astrobin.com/jdaa90/ --- the difference between the two monochrome versions of the Squid is the application of HDR Multiscale Transform which reduces the halo on the central star as well as the amount of data used: (A) is based on the full ~53 hours whereas (B) is based on the best 40 hours. (C) and (D) are corresponding versions that combine O3 in the blue channel with RGB stars along with changes in the amount of saturation and brightness. Similarly with (E) and (F), but the Squid is in monochrome mixed with the RGB stars background. After doing different versions of the Squid by itself, I worked on a narrowfield and widefield versions where it is situated between the wingspan of Flying Bat (coming soon). I started with data from the FC100DF/QHY163M combination which results in a total exposure of 78+ hours for SHO. The SV70T/QHY163M data used for the widefield framing representing another 51+ hours was also integrated in. The RGB data using the QHY247C with the SV70T adds another two hours. The total exposure of the final image with the narrowfield or widefield framing is 131+ hours, my longest to date! As always, thanks for looking! --Ram
  7. This is an amazing image - the best M101 I've seen... --Ram
  8. Also includes: Cr30/Mel16/IC1824/NGC1027, IC1831, LBN646-48,50,55-57, and LDN1356,59-73. Total integration: 43.75 hours/2625 minutes (136x7m = ~16h for Ha + 92x7m = ~11h for O3 + 147x7m = ~17h for S2). Camera: QHY163M (16mp mono) CMOS cooled to -15 degrees C. Telescope: Stellarvue SV70T triplet apochromat refractor @ f/4.8. Reducer: SFFR70-APO Mount: Paramount MyT. Filters: Astrodon 5nm Ha, 3nm O3, 3nm S2. Software: TheSkyX Pro, SharpCap, PixInsight. Inline image with reduced quality uploaded to the forum. Full resolution images of all versions are available at https://www.astrobin.com/97h2mp/ and of the primary version at http://ram.org/images/space/scope/ The Heart nebula region is about 7500 light years away from Earth in the constellation Cassiopeia. The 330 light year diametre complex contains the eponymous emission nebula (Ced7/Cr26/IC1805/LBN654/Mel15/Sh2-190) shaped correspondingly, along with a separate region labelled the Fishhead nebula (Ced6/IC1795/LBN645/NGC896), which was the first part to be discovered. The nebula's shape and distinct signal is due to a cluster of stars in the middle in an area that could appropriately be called the tendrils of creation. One of the more striking aspects of this nebulosity is the number of dark nebulae (~15) contained within it (LDN1356 and LDN1359 through LDN1373) many of which I've tried to highlight especially on the Ha only image. There's also a small open cluster to the right (Cr30/Mel16/IC1824/NGC1027). One of the cooler features is the ring nebulosity in the centre which you can make out if you look closely, which is like a second (or more likely separate) shell of gases being pushed out. This capture is part of my "make up for being a newbie" series, where usually once a year I revisit one of my very first images and redo it with the latest capturing and processing techniques that I've learnt. In this regard, the Heart nebula complex represented a lot of firsts for me: it was one of my very first astrophotography images, my first narrowband image, the first colour image with the SV70T scope, and it was done on my first mount. This version which I did back in 2017 had a total of integration of 4.7 hours, so this current version has almost ten times as much information going into it. I hope you'll agree with me that this current version is a far cry from the image I did using live stacking, 8 bit PNGs, and my AVX where I could only get two minutes unguided per frame, yielding a total integration of less than five hours (https://www.astrobin.com/305156/). There's ~44 hours of captured data going in this version and I think even though the gains are incremental, I was able to bring out additional background nebulosity and details, as well as reduce noise, particularly in the Ha only image which I thought turned out really good: deconvolution actually worked like I thought it would for the first time and I was able to achieve a balance between sharpening and noise reduction. When I first started redoing this, the moon was half full so I decided to capture S2 data first and I was off a bit in my framing which I only corrected after my first session (there's a seam barely visible if you can spot it). I was able to use it but it is why I have the most amount of data using this filter since I wasn't sure if the first session would be useable in the end. Overall, I would have to say this is the first image where everything came together nicely in terms of processing. Take for example the ring nebulosity, it's missed in many images of this target or not pronounced (my previous image doesn't have it at all) but it shows up here nicely. I've put up two bicolour and tricolour Hubble palette versions of this capture on AstroBin along with the Ha only version. Previously I had done a HOS combination but that was mostly Ha data since I had done a crappy job on the O3 and S2 data; this time the colours in the HOS were okay but not as good as the SHO version so it is not being showcased (just imagine the Ha only image in bright red and you get the picture). The framing is also the "right side up", looking at it from the Earth. As always, thanks for looking! --Ram
  9. Yes, both the broadband (which will let most light in, just not ALL) and the duo narrowband filters (these are not the same as regular narrow band filters) will work on a unmodded DSLR (so can real NB filters but the issue with a DSLR IMO is cooling and in your case, a slightly older model). This is what I mean by the duo narrowband filter: https://www.cyclopsoptics.com/astronomy-filters/stc-astro-duo-narrowband-filter-48mm-2-free-shipping-free-lenspen-set/ It's a 2" model and it's expensive (about $300+) but I don't know the prices - this is just the first link. It says on there it's suitable for DSLRs and OSC cameras. So if you do get a OSC CMOS camera that can be cooled, you can still consider this down the road. Bortle 4 is good though - that's rural/suburban transition if I recall so you shouldn't be having that much issues with light pollution unless you're imaging with the full moon or have a light source. So it could be your processing - I'm just wondering why your images look a bit washed out. --Ram
  10. What is your light pollution level (Bortle scale)? More time per exposure isn't necessarily good - it depends on your light pollution. If you are in a light polluted area, you may be better off with shorter exposures but more of them. The PacMan is a relatively bright object so you shouldn't need to spend too much time on it total. If you are in a light polluted area five minutes may be too long and one thing to look into is a broadband filter. Also I've seen some people use these so-called duo narrowband filters for colour cameras and they're producing really quality images with those - with most of the light pollution filtered out, these emission nebula will really shine through such filters. If you're imaging from a dark site then obviously your issue isn't light pollution, but I don't know much about your camera... --Ram
  11. Total integration: 37 minutes/2210 seconds (37m L + 7-8m for RGBSHO). Cameras: *QHY163M (16mp mono) and **QHY247C (24mp OSC) CMOS cooled to -15 degrees C. Telescopes: *Takahashi FC100DF Steinheil fluorite doublet and **Stellarvue SV70T triplet apochromat refractors @ f/4.9 and @ f/4.8. Reducers: *Takahashi FC-35 (0.66x) and **SFFR70-APO (0.8x). Mount: Celestron AVX. Filters: 1.25" Astrodon 5nm Ha, 3nm O3, 3nm S2, Orion SkyGlow. Software: TheSkyX Pro, SharpCap, PixInsight. Inline LRGBSHO and LRGB images with reduced quality uploaded to the forum. Full sized versions of both these and the LRGB and L only images are available at https://www.astrobin.com/d5zk98/ Full sized versions of the LRGBSHO and LSHO images are also here: http://ram.org/images/space/scope/1.4+7.4.5+6/neowise_c_lrgbsho_37m.jpg and http://ram.org/images/space/scope/1.4+7.4.5+6/neowise_c_lsho_37m.jpg Comet NEOWISE aka C/2020 F3 (NEOWISE) was named after the mission that discovered it using the Wide-field Infrared Survey Telescope (WISE). Initially identified on March 27, 2020 when it was an eighteenth magnitude object, it started getting brighter and brighter as its long period approached the sun, eventually reaching a peak of magnitude 0.5. This is the brightest comet observed in the Northern Hemisphere since Hale-Bopp in 1997. I had earlier observed the comet with my naked eye a couple of times, once when it was closest to the sun on July 3. But the early morning of July 12 was when everything cooperated in terms of seeing/transparency/clouds and it was the brightest comet of any I've ever observed with my unaided eye. It was also my daughter Maya's first comet observation. The first time I had chance to image comet NEOWISE was on July 24, 2020 when it was around magnitude five just a day after its closest approach to Earth. The issues I was having were numerous: I was using my normal high end mount for a long imaging session of the Heart Nebula IC1805 (again!) and I didn't want to move it or change its synchronisation. There were low level clouds and this comet hugged the horizon whether you saw it in the morning facing Northeast or at night facing Northwest. I had also moved my cameras around and had forgotten the spacing required to obtain the roundest stars with my QHY247C which I didn't have time to debug. I finally dragged out my original Celestron AVX mount and moved it an optimal viewing location, mounted my FC100DF and SV70T scopes on it, and managed to grab some frames using two cameras. At first I was excited to go back to my old mount, doing something different, but then it was such a pain to use that I realised what a difference the MyT makes! In any event, I was committed now. I initially started with my FC100DF and used my Baader VariLock T-thread extender to adjust the spacing between the QHY247C and the reducer to get the stars as round as possible and grabbed several frames at 1s, 2s, and 15s. Then since I knew my SV70T was set up more correctly with the QHY163 (this combination is what I was using on a nightly basis to image the Heart), I decided to use that to capture several frames using just a luminance filter. Since the Ha, O3, and S2 filters were present in the filter wheel, I imaged several more frames to highlight whatever was being captured based on the bandpasses of these filters at 4s, 15s, and 30s. I would argue this is a fairly unique combination and in addition to seeing where these elements (or other substances that overlap these bandpasses) are highlighted, I wanted to get some colour into my mono images in case my OSC images didn't work out. I then prepared all the necessary darks, calibrated, registered, comet and star aligned, and integrated these frames separately and in combination using PixInsight. In terms of processing, I first created a luminance (L) master of only the comet that included all frames, including the narrowband data (I experimented with creating both a L-only master and a complete master). I then created separate OSC, Ha, O3, S2, R, G, and B comet masters (in part because I wanted to see/show the data separately). Since a comet moves much faster than the stars in the sky, and this was my first time processing comet data, I had a hard time with removing the star trails and getting clean comet only and stars only images: only the main LRGBSHO image (A) is the one where I carefully clone stamped out the star trails; in all other cases, I didn't bother since I should've done this earlier in the process. The solution that worked for me was to create a starless image of the star master created from aligning on the stars (as opposed to aligning on the comet, which are supposed to be done separately and combined but that didn't work for me; I even tried a star mask based approach to add in the stars but again, no luck) and then subtracting it from the parent image, leaving me with only the stars and without any comet signal. Armed with L and RGB stars only masters, I then used the SHO-AIP script to figure out which combinations looked the most aesthetically pleasing. Overall, given the timing, the time spent on capturing (less than an hour), the rush in which it was done, and the older AVX equipment, I thought it turned out okay. I rarely see people taking narrowband data of comets, so I thought that was interesting. You can see both tails in the narrowband only image (B) are mostly green indicating hydrogen signal but the comet head is purple indicating a combination of sulphur and oxygen and the gases right behind are cyan indicating some combination of hydrogen and oxygen. The LRGB image (C) on the other hand shows the typical green halo of comets that's due to the presence of carbon. There's some internal structure to the body (i.e., the actual ball of rock and ice is clearly distinguishable). Finally the different sets of images combine to make one that is clearly brighter/better in my opinion even though I have a soft spot for the narrowband only image. That said, having processed my very first comet, I do wish I had spent more time capturing additional data. I started my astronomy hobby in all seriousness after seeing Halley's comet in 1986, a time when I could barely tell the difference from a comet or a random star, even with my small 3" telescope. Fortunately I had much older people than me who were able to work magic with this scope and what I saw was pretty decent, especially compared to what I saw in a 16" homemade reflector (the construction of which was completely done by hand, from the cutting of the wood to build the tube, to the grinding and polishing of the mirrors; I never did this myself but I went to every class and watched a bunch of people do it and saw a bigger and more refined operation of the same process, i.e., a three metre mirror being automatically polished going by memory). So working with this comet was a fun trip down memory line and while these pictures are not as great as they could be due to the timing, I can't wait 6,800 years to see and image it again! As always, thanks for looking! --Ram
  12. That is a stunning image! --Ram
  13. I've now linked to the bigger sized images - so if you click on the small inline images it goes. Thanks for suggesting this! --Ram
  14. You know, whenever I tried to upload something I can't get more than 10 MB - 1 GB for each forum post/image would be amazing but I think unrealistic. Because I wanted to upload two, it was easier to just upload the inline images. But glad you were able to get it - I think the astrobin or telescopius is the only place where you can get the multiple versions. I could put them up too but I've not figured out how organise them like they have. Thanks everyone for the kind words! --Ram
  • Create New...

Important Information

By using this site, you agree to our Terms of Use.