aleixandrus

Hi! I have an issue with my flats and, after many-many tries, I can resolve. I have a ZWO ASI183MM Pro with a Samyang 135mm @ f2.8. I also have a ZWO EFWmini with 6.5nm SHO Baader filters. My lights (both subs and master) seems fine. I recently acquired a circular flat panel (3D printed) with a slightly bigger diameter than the Samyang 135mm. What I found is, as I rotate and change the position of the flat panel, the light pattern of my flats also change orientation. Please, note that this screenshots are low quality as it is a remote connection (the subs SNR are as as good as expected). However, I think the "light to dark" pattern and how it change as I move the flat panel is clearly visible. As I was concerned about the quality of the panel and how even the illumination is, I also use a 10" tablet but the issue is still there. I tried the following: Double check camera parameters (gain 111, offset 8, temperature -5ºC) Double check the lens hood for evident tilt or other mechanical issues Visualize the subs with different software: NINA - PixInsight - ZWO Fits Viewer Change the exposure length: 0.1s - 1s - 4s - 10s - 15s Change the brightness of the panel/tablet (10% - 40% - 90%) Change the position of the histogram peak (20% - 40% - 80%) Change the filter: S2 - Ha - O3 - L - No filter Change room illumination: on - off - covering the telescope with a blanket I always get the same results: as I change the position of the tablet/panel over the lens hood, the light pattern in the flats also change. But for a given position, the pattern remains stable. When visualized with no histogram stretching, the flat subs look normal with very small vignette. I was concerned that as I'm virtually stretching too much the histogram of individual subs, making small differences too visible. So I stack 30 flats (same panel/tablet orientation) with WBPP and darkflats to check if the differences or if outlier data are averaged, but no, the master flat looks similar to individual subs (but calibrated, cleaner). Oh, by the way, as the ASI183 has a small sensor, the Samyang 135 has a flat field (support full frame cameras) and I tend to shoot large nebula, my master lights seem to calibrate "well" to my eyes. But background extraction techniques are somehow tricky to me (I use Pixinsight ABE or DBE and GraXpertt) and I'm not experienced enough to confirm the masters have no hidden issues. I'm ran out of ideas, I don't know what to try now. I suppose the issue is not what I do with the flats processing but the flats itself: why my flats are not even and change as I rotate my flat panel. Any help of hint is very welcome.

I like the new versions. The Spagetti is a nice target but, as you said, is quite faint specially in the O3 channel. I have some Ha data taken with my previous setup configuration I will have throw away as I won't be able to combine with the new one I hope to obtain in a near future (very different rotation and backspace configuration). I need some clear nights!

Weather is being awful but I had some data to work with in the meanwhile

Very nice images @Budgie1! May I point one thing? it seems to me that the California and the Spagetti pictures have a bright large spot in the middle... not a vignette but kind of. The other two seems fine to me. I may be wrong but, well, I think it worth pointing it out PS: Love that Spagetti!

Weather is being awful for so long I'm afraid to forget how the telescope works! In the meanwhile, I spend some time processing data from my last session, 10 October. Yeah, it has rained EVERY SINGLE DAY until today, and weather forecast shows rain in the next 15 days. Oh, dear. Well, this is the result after combining ~30h of useful data in SHO with my ZWO ASI183MM Pro and the Samyang 135mm f2 (@ f2.8). I use drizzle x2 and BlurXTerminator but this is a very undersampled setup so I cannot crop a lot to show the small details of the Bubble Nebula, an object I like a lot. However, I think the result is nice to see. I'm trying to leave some green in the image to show the transition from blue to red but... please, feel free o critique & comment because this is the 3rd time I edit this image as I don't get the colors right 😅 PS: the combination of 20MP ASI183MM with drizzle x2 is making files GIGANTIC and my hard drive is suffering as I'm keeping all the intermediate processing steps while editing 🙈

Thank you for your insights. I think you're right, I'll probably go KISS (keep it simple, stupid! 😅) and make an obsy for my current setup (and just a bit more for growing). If in the long term I have a second rig, I'll probably need to rebuild it from scratch anyway to fix details I'll probably mess up with due to lack of experience. I'll go to the design board and make an sketch, if things go well, I'll try to open a thread in this forum. As I said, thank you so much, very thankful.

Oh god, you made my mosaic bite the dust! Impressive work. I'll save that image (for personal use) as a map of that region of the sky. Love it!

Thank you for the data. Based on this pic, I though it was bigger, it doesn't feel as narrow as 1.9m externally! I initially planned 2x2m inside just for imaging (i.e. 1m free radius). I also consider 3x2m to allow a second pier (for a small wide field refractor I don't have and I won't have in the mid term) but I think it is quite optimistic. As you actually build it that size, if you remove the desk... what do you think?

Impressive work, love it! Thanks for posting all the process, I'm planning building in the very short term and this helps a lot. I may missed that info in the previous posts but... could you please share the dimensions of the obsy? The pier is not centered, isn't it? Thanks a lot.

Weather was awful in the past month so I couldn't advance as much as I hoped. However, I get a unexpected row of three clear nights. Seeing was not wonderful but with my pixel scale (3.67"/px), even with poor seeing I think it worth shooting as I won't 'upresolve' the seeing never ever. So, I get some data. And I couldn't resist to stack everything and build a second test mosaic, this time in color: I know, I know, there are many things wrong but... I like it a lot! At least, I think the project is going along the right way, which is peace of mind for me. Some details of the above pic: This is a 3x3 mosaic instead the 4x3 I planed. I don't have enough data to add the 'Tulip Nebula' column and weather will be awful the next weeks so I stack what I have. This mosaics add 25-26h total (1h/filter/panel approximately) after removing bad frames, captured during 5 nights in 7 weeks (I also have a side project so I reserve some night time for it). I calibrated and stacked the data with WBPP, combined SHO with ChannelCombination (no normalization) and merged with GradientMergeMosaic (no DBE applied) just to make sure everything worked together. There are two large 'pinched stars' I need to fix and some noise differences from panel to panel, but nothing completely wrong, which is good news. I add some curve adjustments to stretch data, SCNR to remove and some saturation. Colors are very wrong, background is not neutralized, there are some gradients specially in the right side... but anything I think I couldn't fix. I didn't any further processing as the SHO master file is around 3GB and, with my old computer, a simple SCNR takes up to 4 minutes to run. I couldn't imagine how long will take StarXTerminator! 🙈 I'm afraid I'll have to consider binning the image... Any thoughts?

Quick update with a more stretched mosaic. No further processing besides better orientation (for my taste), edge cropping and curves transformation, using the previous shown data:

Some time ago, I comment in the Samyang 135 Mega-Thread that I had resolved most of my issues with this lens, specially those regarding backfocus. Not perfect but well enough to deserve some fun imaging. So... I planned a large mosaic of the Cygnus region: As this is a large-term project, I've planned to open this thread to comment on my progress so I can ask for advice and critiques (and you can gossip if interested 😉). So... there we go! This is a proof of concept shoot to test if the whole idea is feasible and make sure I can fix anything prior to the actual shooting. Some details: Main equipment: EQ6-R Pro + ZWO ASI183MM Pro + Samyang 135mm f2 (@ f2.8) + Baader SHO 6.5nm Narrowband Filters Other equipment: ZWO EFWmini filter wheel + ZWO EAF focuser + ZWO ASI120MM Mini + William Optics UniGuide 32mm + Astrodymium Ring System Imaging: 4x3 mosaic with 1x300s Ha + NINA for planning and data acquisition. Conditions: Bortle 4 backyard according to LightPollutionMap (probably Bortle 5 now) with some streetlights pointing towards my telescope placement. Processing: data calibrated with darks using WBPP and merged with GradientMergeMosaic with Pixinsight. Simple STF applied, no further processing. My OTA doesn't have a rotator so adjusting orientation is a bit of a pain due to the belt-system focuser I use. So I had to carefully select the orientation and panel disposition so I can get individual panels as 'interesting individual pictures' while adding more integration to the mosaic. My initial plan is to build a 'base' mosaic with 12x300s per filter (SHO) per panel: (4x3) x (3x1h)=36h total. Then, I can add more integration up to 3h/filter: 3x36h=108h. I know this is quite ambitious for my sky conditions -clouds and rain most of the year- and my skill but, well, it will be fun and I won't have to rack my brain choosing targets for a while 😅 Feel free to ask, critique, correct or whatever, any comment will be welcome!

I really like the image, I don't think the OSC is imposing a limitation here, specially with that sensor. Just keep adding more light :) Regarding processing (and I'm not the one you should ask for advice), the background is a bit darker to my taste, specially in the bottom left. From the compositional point of view, that left area don't add to the image so I would try to crop the image a little or pull out a bit more info in the dark areas. However, I like the nebulosity a lot, specially those reds, it feels quite natural. But if you lift the dark areas you can try to increase a little bit the exposure in the light areas to match the overall contrast. Maybe you can try to add some overall contrast, not sure if this will lead do a better image but you can try and compare. Finally, while I like the star reduction and the no-halo around them, the brighter stars are a little bit misshapen with some faint halos. This also happens to me with the Samyang as I'm way undersampled so they are a bit blocky and the BlurXTerminator can't fix them 100%. I don't feel this is a problem as it only shows up when you zoom 1:1 but, well, it worth to point it out. Comments beside, I really like the image. Oh, yes, I love those reds to the point I'm thinking to re-process my SHO Pelican and North America and do it again, this time HOO 😄

To be fair I also see that rotator but it feels quite cheapo so I quickly dismissed it. Well, based on your words, it could have been a mistake. Do you feel it may introduce some tilt or it seems strong enough? I will explore that option again but after all the backfocus/tilt issues, I'm very reluctant to make any change in the assembly for a while. I'm also in the middle of my "Cygnus mosaic" project I've shared some post behind so I won't change rotation for some time. Thank you so much for sharing, I am very grateful.

Oh, that thumb screws mislead me, I've already checked that rotator but I can't use it as it uses too much backfocus (I added a filter wheel so I'm close to the limit of total backfocus). Well, although it may happens it is not that frequent. And carefully checking the equipment from time to time is always good idea even with metal parts. I found my camera loose after one session and even today I don't understand why. Of course it didn't break anything but it ruined one imaging night. Simply check printed parts from time to time and don't over tighten the screws and you will be fine. But if your RedCat rig works... fabulous!

A vote to Astrodymium ring system. I'm quite happy with it despite I had to make some modifications to attach more elements as I'm not using the AsiAir but a MiniPC + PegasusAstro Powerbox. Oh, and I had to contact with the manufacturer (I broke a screw head, totally my fault) and he was very helpful and considerate providing a replacement, worth to highlight this. If you plan to attach additional devices to the lens (specially the EAF), one of these 3D printed rings really worth it. If not, a solution as the RedCat51 ring may be more robust in the long term. However, when using 3D printed accessories is very important to not over tighten the screws as the printed material will suffer from microcracks with will lead to major fails without notice. Tightening enough is... enough. Side note: Nowadays my major complain is not being able to rotate the framing easily as the focuser and the lens are coupled to the 3D rings. I would like to rotate the lens after loosen up the rings but the EAF do not allow turn the motor by hand. To be fair, this is a common issue with most 3D printed rings for the Samyang. I've 'resolved' this issue by moving the focus ring to the middle with the EAF, way out of focus, and then I loose up the rings to freely rotate the lens while keeping the focus ring and the belt carefully in position to do not damage the EAF motor. If I need a precise rotation angle, I move the focuser close to infinite again and do a plate solving... I have to repeat this procedure until I get the desired angle. Despite I know approximately the camera rotation checking the back of the camera body, it is quite annoying. I know I can attach a rotator but I don't change framing too often, a rotator may introduce some tilt which at f/2.8 may be a problem, flats are more fussy, I don't have too much free backfocus space and I'm tired to deal with stars in the corners. Maybe in the future... @Space Oddities: Could you please share which rotator model are you using? Thanks a lot.

3x4 would be marvelous! You're probably right, maybe 2x3 would be just I-want-to-but-I-can't, specially as I don't even has a second rig so I don't know what my requirements will be. Thank you for your insight, really appreciate it. Please, keep this thread updates with future improvements on your obsy or simply your experience using it. That 'post-mortem' feedback is as important as the build itself!

Thank you for the kind response @AstroKriss. I also plan a 2x2 observatory but I'm 1.90m tall and I would like to be able to stand with the roof closed... It's good to know you don't feel it too narrow. I will need a taller pier but, as I plan to make it of concrete, I suppose I can make it the right size if I take careful measurements. I also live in a residential area but I'm not so constrained to size (although I need to keep it at a minimum). I'm also considering 2x3m, leaving a bit of space for a second pier for an small refractor or even a portable/moveable pier... Do you feel it possible with that dimensions, based on your experience? Oh, by the way. @Starflyer, I also use a IR cam to control my scope when in the backyard and, despite I tend to turn IR off except when actively monitoring, I didn't notice any issue with my 6.5nm SHO filters.

Well, it is really impressive if, as you said, "I had no real idea what I was doing", LOL. Really like it. I plan building a small shed as yours, different material and constructive approach but same dimensions. Can you please share some details about dimensions? You said it is a 2x2m frame for the base. What's the wall and pier height? How many horizon degrees you lose? It seems you can close the roof without parking that scope. Do you feel the remaining space too constrained? Would you go bigger now you finally build it? Too many questions, I know, but I really appreciate you answers if possible. Anyway, thanks for sharing that pics!

Thank you for pointing that issue, @powerlord. Yeah, I know stars are a mess but I couldn't get them better and I'm not sure where the problem is. I build the master S+H+O, do a ChannelCombination for SHO, apply BlurXT+NoiseXT+StarXT in linear phase. I think nebulosity processing comes just fine but I struggle to move the stars image from linear to non-linear (using Hyperbolic Generalized Stretch, with Color+Arcsin method). I also use the Bill Blanshan script for star reduction and star blending and there is a (very probable) chance I'm using it wrong. Any tip is welcome!

I had acquired some data to start building the large mosaic I mentioned a few post ago. I focused in O3 to take advantage of the past moonless nights but I also added some Ha and S2 to one particular panel. This way I can build the mosaic while having some data to fully process individual tiles and show some progress. Well, this is my first image it deserves to be shown. The Crescent Nebula region, ~8h with the Samyang 135mm @ f2.8 + ASI183MM Pro 300s + SHO 6.5nm Baader filters. This is also my first "properly processed" SHO image, this is, using PixInsight and the "correct" steps and a repeatable workflow. I know I have a loooong way to go but it is a first step. For instance, I have some issues with star processing I must deal with (check that dark halos around many of them) but I think the nebula it is reasonably accomplished (to my untrained eye). Please, feel free to C&C and point out any aspects that could be improved!

Some posts ago, after dealing and (mostly) resolve backfocus issues, I said I deserve some fun with the Samyang. Well, this is my upcoming project: The pic above is just a test-shoot to make sure I can fix anything in planning prior to the actual shooting. This is a 4x3 mosaic with just 1x300s Ha, calibrated with darks and processed with Pixinsight (GradientMergeMosaic, DBE and manual stretch). It went much smoother than I expected... which scares me as hell as it could be just good luck 😅 The initial plan is to add 12x300s per filter (SHO) per panel (4x3=12). This is 36h total to build a 'base' mosaic and, then, add more integration where needed up to 3h/filter (108h). What do you think? Any advice from those with more expertise? It seems ambitious for my skills -never did a mosaic before-, for my sky conditions -clouds and rain most of the year- and many things may go wrong but... hey! it seems fun!

Let me add this picture to complement my previous post. Top row shows the stacked SHO channels, bottom row shows the single SHO subs. I think it clearly shows I still have room to improve, specially in the O3 filter and that right side (tilt?). However, the final edited version seems just right. IMHO, starts in SHO color are not great so I'll probably use the ones in the Ha sub despite being white. I'll deal with stars shapes in the future when I buy RGB filters and a bigger filter wheel. For now, as I say before, I think this is 'well enough' and I think I deserve some fun just imaging without dealing with technicalities. But feel free to contradict me PS: those subs are just random selected subs for illustration purposes, some AF issues may happened there, specially in the O3 channel!

Well, I had some issues recently with my Samyang, my Baader SHO filters, backfocus, guiding… I think I sorted out most of them (or at least minimize them to my standards) to the point it worth to explain them a little so I can help others with my achievements. Let’s go! My current OTA is: Samyang 135mm stopped down to f/2.8 Astro Essentials Samyang Lens to M48 Adapter ZWO 16.50mm M48>M42 adapter (the one provide with the camera) ZWO EFWmini 5 position filter wheel ZWO ASI183MM Pro These elements add, respectively, 0+16.50+20+6.5 = 43mm. I need 44mm total backfocus as I’m using the Canon EF lens version. I also use the SHO Narrowband CMOS optimized Baader 6.5nm filters, 2mm thick. It is well known you should add 1/3 to backfocus, this is, 0.67mm. So, in total, I need to add 1.67mm using spacers. I bough the ‘Artesky Fine Tuning Ring Set T2”, a cheapo kit of 10 metal spacers from 0.1-2.0mm. Spoiler: bad purchase; they are not labelled and use poor quality inspection techniques, their thickness is not consistent and some are very slightly bended but enough to introduce tilt (to my eyes). However, I add 1.0+0.5+0.15=1.65mm spacers and made some tests. I don’t want to bore you with all my troubles but, although my Ha shoots seemed just ‘right – but not good’, when I purchased S2/O3 filters and tried to mix all channels, that was a mess. Just one representative sample: As a whole, I like the image (this version is edited to taste but not carefully) but those corners are just awful even with some star reduction. This is going to get me into troubles in the future for sure (mosaics or when adding RGB stars, for instance) so I need to address them. This is all close I could get to the supposed ‘right’ backfocus after many nights of AF tuning parameters, manual focusing, mess up with 0.1mm rings, etc. At that fast ratios, a 0.1mm spacer makes a BIG difference but I tried to set the infinity mark ‘just in the base of the L mark of the lens barrel’, as many suggest. None worked. Discarding the user error or AF misconfiguration, I opted for a more drastic approach: forget about the ‘right’ backfocus and adjust the Samyang mechanically. My first step was reinstall the Astro Essentials M48 Adapter and make sure it was not overtightened and check for evident tilt. First lesson: the Astro Essentials M48 Adapter has the same size, height and thickness than the Canon mount so any measurement regarding backfocus can me done from the face of the adapter, just as you should do with the original mount. This is a common question in this forum. My second step was to check if the infinity position of the Samyang was accurate, at least to take it as a reference point. Spoiler: it was not. This video from Nico Carver (Nebula Fotos) is very helpful so I don’t will repeat this info. The procedure is quite simple, it takes 10 minutes and has no troubles, anything can be undone, but I take these variations: 1. Instead removing the infinity hard stop, to have room when past the correct infinity focus to avoid damaging the motor of the EAF, I just move one position using the same screws. This way I still keep a hard stop to prevent (possible, don’t know for sure) damaging the lens if I go waaaay out towards infinity and all lens parts keep stored inside 2. I checked that the external lens barrel, the one with the ‘L’ mark, is just placed with three screws *with no precision at all*. That mark is essentially useless if the lens was not assembly with care. If your stars are not right, you just cannot trust it. Left: before / Right: after. Pay attention to how I’ve displaced the infinity stop just one screw position, so it still works. The bottom row shows how hard stop position shifts, leaving enough room to past infinity during the AF procedure but still maintaining that hard stop. Detail of the focus ring fastening. There are 3 screws like this which hold the whole focus sing. Pay attention to the amount of adjustment space available, which directly impacts on the precision of the ‘L’ mark position (the whole focus ring piece is market in blue). Those screws may loosen up for many reasons (temperature shift along the years, excessive stress over the focus ring, lens aging, poor quality assembly…) so you can’t really trust that mark. I suppose that’s the reason why many in this thread disagree on where to set infinity respect the ‘L’ mark. The ’correct’ procedure should set infinity focus based purely on a test image and then disassembly the lens and move the focus barrel according to that, so the ‘L’ mark is in the correct position, ie., calibrate the focus barrel. I think this effort is not worth it. But, if you feel you are too close to the infinity stop, displace it a bit to have more room, it is an easy task with minimal risk and reversible. I have a 77-49mm step down ring to step down the lens to ~f/2.9 (I’m not sure this value is correct, I need to check it) instead using the aperture ring (so I can avoid spikes in brighter stars). I suspect this element may deform the stars in the corners or introduce tilt, somehow, so the next attempts were done stopping down the lens to f/2.8. I prefer to ensure backfocus first and try this later. Finally, I made some tests with different spacers, again. However, this time I did some DIY to the metal spacers to ensure flatness using a sandpaper, rasp. As I said before, bad purchase but it was all I have for now. I use NINA for autofocusing using both internal NINA algorithms and Hocus Focus plugin. The later uses multiple measurements in all corners and centre so you can find more accurately how your optical train behaves, including backfocus issues and tilt. You can also re-run the focusing process offline (during the day) with different parameters, so it is very helpful to address issues. HOWEVER, I could not get consistent results between their indications and my visual inspection. I don’t know why, it may be user erro or wrong plugin configuration. After some test, disassembling the whole setup each time, I added the only 2mm spacer ring, adding a total backfocus of 45mm, which is not 'the correct' backfocus. Is this right? I don't know and I don't care: although not perfect, it is way better than my previous attempts and purely based on visual inspection, it is good to me (at least for now). 7.5h total exposure (2.5h per channel). Data stacked and calibrated with darks/flats/darkflats and WBPP in PixInsight. No noise reduction nor blurxterminator-like apps, just Automatic Background Extraction, RGB channel combination and curve adjustment to taste. This one below is the original full resolution picture (if the attachment worked fine 😅) if anyone is interested in viewing more details. To me, bottom right corner shows some tilt (it is more visible in the subs) BUT given the effort that take me to reach this point, I’ll try imaging for a while and have some fun. I will evaluate the 77>49mm ring in the future and, maybe, a rotator/tilter, but the later will be too problematic given the large the FOV, the difficulties doing flats and the possible tilt at this fast f-ratios, so I'll probably pass. This Samyang mega-thread was VERY helpful so I expect this may help some of you who are having troubles with this lens and give something back to the community, not everything should be ask for advice. Feel free to ask/correct anything!

I've checked my settings and I confirm my previous values (50 step size, 150 overshoot, 4 displacement steps). However, I experiment with different step sizes (25-50-75) as I read that the recommended difference between minimum and maximum HFR values should differ x3-4 times for a good and consistent AF. Your graph (and mine!) shows ~5 / ~11 HFR, this is, x2 difference. This also happens to me and I can't make this different bigger without messing *a lot* that step size. However, AF seems to work well enough so I finally set 75 as default step size when stopping down the lens at f/2.8 using the ZWO EAF (12V version). I suppose working at this fast focal rations things may be different compared to more conventional scopes, I don't know but hope this helps. PD: That 'V' shape seems right but I think you still have a 'bug' in your AF procedure, @Petrol 😂