inFINNity Deck

Hi John, thanks for those example images. I do not mention chip-size simply because it is irrelevant. If you do not want that black space around the object, simply crop it in the post-processing. None of the planetary images I show on my website are shown at the actual image size they were imaged at, most of them have a larger background than original, while almost all have been resize by that 200% I mentioned before in order to make them a decent size for viewing. Going back to your images: number three was taken without an ADC (or with a poor aligned camera to eye-piece set-up), so no point discussing that one. The first image was not properly exposed. The histogram looks fine, but the central part of the planet seems overexposed (or flattened in post-processing), so again no use discussing that one. The second image is fine. You wrote that it was taken at f/17.5, where the optimum would be between f/12.9 and f/15.9, so we may expect it to be oversampled if we analyse it. So I ran it through ImageJ to get the frequency spectrum of it: If it were properly sampled the result would have filled most of the image, now it shows that the original image was oversampled by at least a factor 3. That this is more than based on pixel-size and focal length alone is due to seeing (I explain that in my second article). This means that if we resize the original image to 33% and then resize it back to original we should see no significant loss in detail. So, I took your image, resized it to 33%, then resized it by 300%. I then took another copy, resized it by 50%, followed by a 200% resize: The one at the left is the 33% resize. The process has made the smallest details a bit square, but there is hardly any significant loss of detail. But as I did not like the square details, I did the second resize with 50%. I leave it up to you to decide which of the two remaining images is your original and which is the resized one. The point I try to make is that, when using an optimal focal ratio, the exposure times go down significantly. Had the image been made at about f/9, the exposure time would have been 4 times as low, resulting in a higher frame-rate and thus more data in the same time-span (and thus a better signal-to-noise ratio had the same percentage of frames been stacked). Nicolàs PS: the file-name reveals the answer

Hi John, good point, sure there is a difference, so let's have a look at the maths. Pankaj plans to use a 10" f/5 Newton with 2x Barlow and his Canon with 4.3 micron pixel-size. That combination yields a resolution of 4.3/(25.4 x 10 x 5 x 2) x 206.3 = 0.349"/pixel. Currently Jupiter is about 44.36" in apparent diameter, so that is 44.36 / 0.349 = 127 pixels on his camera. Now he should be imaging between f/12.9 and f/15.9, so should be using at least a 2.5x Barlow (f/12.5). Using that Barlow he would get a planetary diameter of 127 x (2.5/2) = 159 pixels (3 x Barlow would make this even 190px). In my set-up this would become about 204 pixels, so indeed significantly larger (about 28% more than with with the 2.5x Barlow), but not a whole lot, and even insignificant if he would use a 3x Barlow. The diameter of Saturn with rings is approximately equal to Jupiter at the moment (44.14" vs 44.36"), so that planet would give the same results. The planet itself (i.e. the globe of Saturn) is of course quite a bit smaller, that would become 67 pixels would Pankaj use a 2.5 Barlow (81px when using a 3x Barlow). Above images of Mars were taken on 26 December 2022 while it was only 15.35" in apparent diameter, so a third of current Jupiter and Saturn and thus only about 75 pixels in my 3 x [pixel-size] set-up, so approximately the same size as the Jupiter that Pankaj can image with his set-up. Nicolàs

Hi John, Please note that there is a difference between "a decent sized image" and an image with maximum detail. The formula 3 x [pixel-size] (or 3.7 x [pixel-size]) gives the focal length at which the maximum detail is recorded. Going above this means that no additional detail is recorded while exposure times go up drastically (quadratic with the amount of oversampling, so going from a factor of 3 to a factor 6 increases exposure times by a factor of 4!). It is for that reason that I stick to the formula and create a decent size after processing by using a bicubic resize of 200%. I have been testing this with a ASI290MM and found no advantage detail-wise when compared to the ASI174MM that I normally use. The pixel-size of the ASI174MM is twice that of the ASI290MM (5.9 vs 2.9 micron), so I resized the ASI174MM images by 200% after stacking to make them the same size in below image (all imaged at f/20 with a C11 EdgeHD). The image at the far right is the same as the one at the centre, but with additional sharpening (bit too much to my liking, but it was to see what happens in the image analysis): The bottom row images show the frequency spectra of the recordings made with ImageJ. I ran the three images through ImageJ to see if there were any differences in resolution. The three FFTJ transformations show that all images have approximately a 3-fold oversampling (if the rectangles are completely filled with data then there is no oversampling, if they are half filled then the oversampling is 2x, with a third it is 3x). For the two ASI174MM images this means that the originals have 1.5x oversampling, as they were scaled 200% in post-processing to make them the same size as the ASI290MM image (and a 200% scaling produces 2x oversampling). So far I have had no issues with my alignment points at this lower scale, I usually use a combination of 24 and 48 pixel APs in AutoStakkert!3.. Nicolàs

Hi Pankaj, Important is to know when you achieve the maximum resolution, which you can find in my white papers on the topic: https://www.dehilster.info/astronomy/optimal_focal_ratio_part_1.php https://www.dehilster.info/astronomy/optimal_focal_ratio_part_2.php In short it comes down to that the focal ratio can be about 3 x the pixel-size of the camera. If I understand well the Canon 1200D has a pixel-size of 4.3 micron, so optimal sampling is done at a focal ratio of about 3 x 4.3 = f/12.9. Theoretically you could go as far as 3.7 x 4.3 = f/15.9, but seeing will limit this rule to about a factor 3 (at least at my location, not sure about yours in India). For your scope this means you could use a 2.5x or 3x Barlow. There is no need to guide, I have seen plenty examples of great planetary images using 'manual driven' dobsons. Please note that exposure times go up with larger Barlows. When using a dedicated camera like the ZWO ASI174MM and filters, you may achieve frame-rates up to about 200fps in FireCapture. These 'videos' can be stored in uncompressed SER-files, which then can be read and processed using AutoStakkert!3. On my website I usually explain the settings I use for this in the planetary section (you will also find the lengths of the separate recordings there): https://www.dehilster.info/astronomy/mars.php https://www.dehilster.info/astronomy/jupiter.php https://www.dehilster.info/astronomy/saturn.php Post-processing I usually do in PaintShop Pro, but I believe that the majority does that in PhotoShop. If you want to use the Canon, I suggest you capture in highest resolution and maximum frame-rate. If that means you have to image separate files instead of AVI or other movie-format, you can use PIPP to create a SER-file from those separate images, which then can be processed in AutoStakkert!3. Please note that collimation and focus are key in planetary imaging. I always check my collimation on a nearby star and focus using a Bahtinov-mask on either the moons (in case of Jupiter) of a nearby star. There is no need for a focus-motor. HTH! Nicolàs

At a first-time serious collimation one should also check whether the focuser is pointing towards the centre of the tube. Best way is to mark a point (or line) opposite of the focuser and temporarily remove the secondary (finding that spot kan be done using a strip of paper wrapped against the inside of the tube with a mark halfway the circumference of the tube. If you wrap it first clockwsie, then counter clockwise, and take over the mark on the inside of the tube, the oppositie location is halfway between those two marks). Alternative method is using a hand-held mirror to see if the laser-spot of a laser-collimator falls in the centre of the secondary (but for that you need to test whether that is already properly centred, but that can be checked with a tape measure). First time I properly tested my 300PDS I found the focuser to be seriously off, here is an image after adjustment (at the left it is higher than at the right): Luckily that focuser has adjustment-screws for this lateral tilt, which is not always the case in other Newtons: HTH Nicolàs

I was not aware that there is an upgrade-set, so no. But for me that is no issue, I use it mainly as a collimator for other scopes and once a year for reach-out where I collimate it at about 45-60° altitude. Still I would like to learn more about those springs, is there a set off-the-shelf available? Nicolàs

Well, my SkyWatcher Explorer 300PDS does show slight deformation when tilted from horizontal to vertical. It is not much, and perhaps even insignificant, but it sure does happen. Best is indeed to collimate it somewhere around 60 degrees or, even better, at the altitude where it is being used. Nicolàs

I have two set-ups (SkyWatcher Esprit 80ED and 150ED), both with ZWO 36mm unmounted LRGB+NB sets (manufactured 2020 and 2023) in front of ZWO ASI1600MM Pro Cool. NGC281 in narrow-band, Esprit 150ED: NGC7822 in narrow-band, Esprit 80ED: M37 in LRGB Esprit, 150ED: NGC884+NGC869 in LRGB, Esprit 80ED: I have no reasons to complain. 🙂 Only the ZWO filters manufactured prior to June 2018 have terrible reflections in H-alpha and S-ii, see second image in this post (open in Chrome for a translation). Nicolàs

Even better: the RASA initially had the same mirror suspension and mirror-locks, but that has been completely redesigned (by a user! (but I cannot find the thread in which that was discussed)) a few years ago, so, yes, they are very well aware that the mirror-mounting is not adequate. The new RASA are equipped with "NEW Ultra-Stable Focus System - six precision sealed ball bearings virtually eliminate focus shift". And that was only to mitigate mirror-shift. The mirror-flop is caused by the silicon filler between tube and mirror not properly holding the mirror. Still, once properly collimated these are very nice scopes (but please do use it with a third-party focuser)! In below image (taken on 9 November 2022 around 20:53UTC with a C11 EdgeHD @f/20 and ASI174MM) we can even see correct detail on Ganymede (far left): Nicolàs

That is what I thought and the reason why I swapped the C11 Carbon for the C11 EdgeHD. The optical bench measurements of C11 EdgeHD in the animation were done directly upon arrival and with the mirror-locks engaged, so the locks do not have the expected result (in my C11 at least). The issue appears to be that the main mirror is not properly attached to the outer focusing-tube. The Italian company Blue Atelier, a technical division of the astronomical park La Torre del Sole near Milan can correct this, but they insist that the telescope is collected in person as they fear that transport may undo the repair. During the pandemic I could not travel there myself, so they explained all the steps required to do the repair, which I should now be able to do so. The repair is, however, quite an undertaking and requires a lot of additional (and at times costly) tools currently not present in my workshop. Apart from the collection costs, it is cheaper to have it repaired in Italy, than to do it myself, so maybe I will send it off somewhere in near future and then have a nice few days in Italy... 😉 Nicolàs

Interesting quote from Carlin that is, thanks. That is only when the SCT is properly constructed. I use a Celestron C11 EdgeHD and had a C11 XLT Carbon, both of which suffer from mirror-flop to an extend that it affects collimation. The current one has been back to Celestron a few years ago and was 'repaired', but upon arrival back here it showed exactly the same issue as it had before. Inquiries prior to its return learned that Celestron does specify mirror-shift (the tilting of the main mirror due to focusing), but not mirror-flop (the tilting due to change in gravity). I have an optical bench with a collimator and can measure it, so I asked for these figures so that I could check if the return journey had affected their repair. They simply refused to specify mirror-flop ("...we would not be able to divulge that information."), only telling me that "We will to make sure your telescope is returned to factory specifications." Upon arrival back here the test on the optical bench showed it was still affected by mirror-flop (the animation shows the collimation in two 180 degrees rotated positions of the scope before and after repair): Needless to say that I was not amused (the shipment alone was more than a grand). Ever since I collimate the C11 each time when I want to image the planets or the Sun in detail, even if that is on two consecutive nights and especially after a meridian flip. Nicolàs

Most important line from that article is "Anyway, if no Airy pattern can be discerned, no high resolution result can be expected (except in big telescopes for whom the Airy pattern is rarely or even never visible)." Here in the Netherlands seeing the Airy disc is a rare treat, most of the time the seeing too poor to see it. That seeing has a disastrous effect on your imaging. As I explain in my second white paper on the optimal focal ratio, the oversampling rapidly increases with seeing and thus any miscollimation will hardly be noticeable under such conditions: In certain cases a little bit of miscollimation can even be beneficial to planetary imaging. In his article "The Mars Edge-Rind Artefact", Martin R Lewis shows that a slight miscollimation can reduce the edge-rind. Nicolàs

Indeed the ZWO cameras work the other way around, here a test with a ZWO ASI1600MM Pro Cool at 3200 x 3200 pixels ROI (same imaging-chip as the QHY163 mono): Also the figures differ: minimum is 40, maximum 100 (vs. min 0 and max 60 for the QHY). Nicolàs

Well, I actually meant lowering the USB traffic to 0 or 1: As can be seen a 0 setting results in a 23FPS for my QHY163 mono at 3500 x 3500 pixels (I use it with my LS80THA single stack). It is not much faster than Dave is getting, but much better than 1.7FPS when USB traffic is set to max. But perhaps this differs for the ZWO camera? Nicolàs

Hi Dave, these are nice captures. You may be able to increase speed by enabling the HighSpeed option and/or by lowering the USB Traffic setting. Also capturing in 8bit mode (instead of 16bit) may be beneficial. I presume you use a square ROI? Which scope did you use (and is that a double stack?)? Nicolàs

All suggested scopes are way too fast, my favourite scope is f/50 (or rather f/49.67) at 1490mm focal length (30mm aperture): https://www.dehilster.info/astronomy/gtt60.php I use it on a daily basis for counting sunspots, the results of which are used for scientific research: https://www.aavso.org/solar-observing-project Nicolàs

Thanks for the likes, much appreciated! I have added yet another paper in the Hardware section, one I wrote with my friend Paul Volman on the construction of a Solar Scintillation Seeing Monitor (SSSM). Nicolàs

Over the past four years I have been writing some 14 white-papers on a variety of topics, but all in some way related to astro-imaging. These were, however, published in Dutch on the Dutch astrophotography forum www.Starry-Night.nl. I regularly refer to them on SGL and recommend to open them in Chrome, so that they get automatically translated into English. Although this works reasonably well, the translations are far from perfect, for which I have been asked to publish them somewhere in English. The past few weeks I have been working on that and now these papers are available through my private website. I have categorized these papers into Hardware and Imaging, where the former category deals with subjects as a Balancing System for side-by-side set-ups, Camera Centring Errors, the Construction of a Collimator, Filter Focus-Offsets, the Collimation of Ritchey-Chrétien Telescopes, Removing Lens-cell Induced Artefacts, Avoiding Newton-Rings in Solar Imaging, and Stability Measurements of an Astro-Imaging set-up. The Imaging section deals with Mitigating Imaging Artefacts (30 of them!), Calculating the Optimal Focal Ratio (in 2 parts) and Solar Imaging (with part 2 and part 3). Hopefully these papers are to use of the members here. Being not a native-English speaker I hope my English (and that of Chrome's translator) is not too much of a challenge (corrections are welcome!). Enjoy reading! Nicolàs

Hi Robin, thanks, will have a look at that, it sounds like the kind of solution I am looking for. Nicolàs

Dear spectroscopists, I recently started spectroscopy and have a question concerning guiding on the slit using PHD2. Equipment is a LHires III, C11 EdgeHD, GM3000HPS and ZWO ASI174MM guide camera. The slit is 19 micron. When guiding on a bright star like Altair or Tarazed PHD2 looses the star regularly due to "low mass". Only when choosing very short exposure times (e.g. 0.2s at 0 gain) it keeps those bright objects locked. The problem then, however, is that PHD2 starts guiding on seeing. So to avoid this, I prefer exposure times of several seconds, but then PHD2 fails. I have to say that I am even newer to guiding than I am to spectroscopy, in deep-sky imaging I never need to guide (I do that at 1050mm and 400mm focal lengths). I have the mass detection disabled. Any suggestions are most welcome, so thanks in advance. Nicolàs

Hi Neil, I follow the same procedure for all my scopes (Refractors, SCT, Newtons) and which is very well explained in the following video, with only difference that I use the Baader recommended colourless and odourless Kleenex tissues (search for Kleenex in that document): With my SkyWatcher Esprits (150ED and 80ED) I take extra care that no fluid gets in the lens cell by having one tissue sitting along the lower edge of it during cleaning. In this way any down-running water, especially the soapy water, gets absorbed by that lower tissue. Main importance is to waste as much as possible Kleenex on the job. Just a few dabs and no more than a single drag with one tissue. For a refractor one litre of distilled water and one family pack of Kleenex should do. For my C11 I use double that amount. HTH, Nicolàs

The polarizing filters usually come as a set of two: one is fixated in the diagonal or wedge, the other fixated to the eyepiece. By rotating the eyepiece along the optical axis, the combined action of the two filters darkens/lightens the view. I use a set like that when observing the Sun with my C11 EdgeHD in combination with ND5.0 foil as I still find the resulting view somewhat on the bright side. Nicolàs

A Baader foil filter is one of the cheapest ways to start observing, but please mind you that you use the right type of foil. These foils can be bought at varying Neutral Densities, most common are ND5.0 and ND3.8. The figure tells you how much light is blocked ND5.0 means 1/10^5 = 0.00001 = 0.001% of the incoming light is being transmitted to your eye. At ND3.8 this is 1/10^3.8 = 1/6310 = 0.00016 = 0.016%, so 16 times as much as the ND5.0 foil. The ND3.8 foil therefore is not safe for visual observation, only for imaging. A Herschel Wedge has a glass wedge and internal filters that take care of the transmission. Wedges too come at varying transmission types: there are photographic and visual wedges. The Starfield Wedge appears to combine the two. They state that it is safe for observing (we need to presume that that is visual observing) and it can be used for imaging. The amount of light can be regulated by its internal double polarizing filter, which (partially) blocks incoming light. Baader too sells a very nice wedge. By far the cheapest solution to observe the Sun is by projection on a white surface or (even safer) by using a Solar funnel (but the view will be less detailed): https://eclipse2017.nasa.gov/make-sun-funnel Using above methods you should be able to see sunspots, pores and granulation. If you want to see prominences and surface detail you would need a H-alpha filter like a DayStar Quark or a dedicated Solar Scope like those from Lunt or Coronado. Nicolàs

Indeed most cameras have their highest QE at around green light, or even in the blue: Starlight Xpress Trius: QHYCCD QHY533, QHY268, QHY600: ZWO ASI2600, ASI6200: ZWO ASI1600: Only red-sensitive camera I found so far is this ZWO ASI990/991: Nicolàs

Is that with a Daystar Quark? Nicolàs