Jump to content

Banner.jpg.b89429c566825f6ab32bcafbada449c9.jpg

discardedastro

Members
  • Posts

    895
  • Joined

  • Last visited

  • Days Won

    1

Posts posted by discardedastro

  1. 55 minutes ago, Hughsie said:

    I recently purchased a Dell Optiplex 790 short form factor (SFF - smaller and more compact) and they are very easy to upgrade. I bought one last year for c£300 via Amazon with an Intel i3 and 8GB of RAM + 500GB spinning hard disk. Now it’s got an i7 installed (£90 off eBay) 16GB Ram and a 500GB SSD (from Crucial UK) plus a 2TB Hybrid SSD hard disk for storage (£50 off eBay). You can get cheaper 790’s on eBay, but sometimes there is no warranty and if something was wrong with the item I would prefer to argue with Amazon to get my £300 back.

    Going second-hand for PC parts can work well but you do need to be very careful - there is an incredible amount of fakes, misrepresentation, and scams on PC parts on eBay. I'd definitely avoid hard disks (SSDs probably fine) on eBay as many will have had a rough life. Even things like RAM are somewhat reliant on people having proper ESD control etc to avoid damage. The scam/fakes side of things is much worse if you're buying relatively new bits, too. GPUs in particular are incredibly sought-after on the second hand market at the moment because of lack of supply (most retailers have waiting lists for the current generation of hardware), so there is a lot of scamming going on.

    Second-hand cases, motherboards, RAM, monitors etc will normally be OK, but for the quiet life I'd always buy new.

    One thing also to consider especially with smaller cases is heat dissipation - higher end parts, particularly CPUs, need large air coolers or water cooling (easily done with all-in-one units which are sealed at the factory) to work fully. They won't die prematurely unless they're really abused, but they'll thermally throttle themselves and slow down if not adequately cooled. AMD recommend a 3-fan watercooling setup as a minimum for a 3950X, for instance.

  2. Very impressive for Bortle 5/6!

    You've got a bit of background which could get cancelled out with some processing, and colours look a bit blueish - suspect this is a side effect of processing to remove sodium glow? You might want to have a play with masks to protect the brighter light sources. But very nicely done. Focus is spot on!

    Hope you don't mind but I had a little play in PI to try and bring out some colour - results attached. The data's really good - I did one which I just did a background subtraction, solved the image, corrected colour balance with PhotometricColorCorrection, and then tweaked saturation with a mask a little. The other I also desaturated and denoised the darker background areas of the image.

    Really good data and lovely framing - got a lot of the little galaxies around M51 in there!

    _510469141_M51Whirlpool_jpg_f8279eea3ca81757ffb3dc0f93043fbc_DBE.jpg

    _510469141_M51Whirlpool_jpg_f8279eea3ca81757ffb3dc0f93043fbc_DBE.jpg

  3. I'll leave the Canon discussion to the experts, but if you're thinking about imaging deep space objects then do consider a dedicated astro camera.

    Secondhand cooled CMOS/CCD cameras can be had for £500 or less sometimes depending on how old a sensor you can tolerate, and will have a much lower noise floor than any DSLR for long exposures. But obviously they're not any use in daytime, so not necessarily the best bang for buck depending on what you want to do! They do also have the advantage of an unfiltered sensor window and easier software control/integration with common software. If you want to do any narrowband imaging then this is pretty key, though you can astro-mod Canons to achieve similar outcomes.

  4. 23 hours ago, Clarkey said:

    Having checked the specifications for the system requirements for PI it would require me to buy a new PC for the processing. Although I have a reasonable PC, it does not meet the recommended requirements for PI. So not only would I have to buy the software but I would need a new computer too. My question is this. Is the improvement with PI, is it worth the cost of a new PC? I do sort of need a new PC anyway, but this would require quite a good specification and higher cost.

    For me, PI was night and day compared to a lot of the other tools out there. You will need "Inside PixInsight" as the missing user manual, but it does exist and a wide range of excellent tutorial content is available online also. It has a steep learning curve but is definitely best in class for the low-level bits (from calibration up to integration) and I would argue is best in class for a lot of the "post" processing like colour calibration, deconvolution, etc. It's definitely the tool to use if you want to make every little bit of your data work its hardest!

    Photoshop, Affinity etc are great finishing tools but I've never found I need Photoshop; I have it for work, but haven't ever reached for it in my workflow - I find I can do it all in PI just as easily.

    In terms of PC spec, PI can run on quite low-end hardware - it'll just be kinda slow to process stuff, but if you're patient that's not an issue. It's hard to advise without knowing what your current rig is.

    Storage performance and CPU performance are the two big ticket items for PI, though having enough RAM to take advantage of both is also important. NVMe solid-state PCIe disks for your working files (I still use spinning disks for bulk storage), a CPU with a few cores (AMD these days are by far the best bang-for-buck for PI and other multithreaded applications), and at least 32G of RAM are all ideal. But I did for a long while use PI on an old Intel 2600K with 16G of RAM; upgrading to NVMe made a huge difference. Going from that to my current rig (an AMD 3950X w/ 32G RAM and faster NVMe disks) delivered a pretty major performance boost because of the much greater volume of faster CPU cores, but it doesn't enable any more functionality - just takes less time when you hit "go" on a process!

    If you're looking to go for a new PC anyway, then just consider PI in your mix of requirements. You'll be looking at around £1.5-2.5k for a high-end workstation-spec machine. I would also recommend considering your monitor - most people I know haven't upgraded theirs in years even if they have upgraded their PCs, and if you're producing images for others then getting something reasonably accurate for colour (optionally with a colorimeter so you can calibrate) is in my view an important consideration.

    • Like 2
  5. That's a good starting point for sure. Something like the iOptron SkyGuiders (iPolar is nice to simplify setup) are worth looking at for a decent tracker, or the Star Adventurer from Skywatcher. Basically this lot: https://www.firstlightoptics.com/star-tracker-astronomy-mounts.html

    You may find a dew heater useful for your lenses if you're outside for any length of time - cheap and cheerful ones will do fine with a 12V supply from a power tank or any old 12V battery. Guessing you'll already have a good solid tripod but if not definitely worth getting one. A Bahtinov mask can be a really helpful and low-cost upgrade for focus accuracy, and a cheap flat field generator (try A4 sketching tablet lights on Amazon etc, perfectly good and cheap as chips) will be invaluable in doing proper calibration of your images which if you're using lenses with any vignetting will be well worth doing (and plenty of lenses have noticeable vignetting for long exposures - even a Zeiss Otus f/1.4 28mm benefited).

    Software is a non-trivial expense depending on the route you go down, Photoshop etc will get you far but if you want to combine multiple images then there are plenty of free options like APP or commercial options like PixInsight.

  6. The most important thing to answer this question is what do you want to take photos of?

    Deep space objects will require tracking, and moderate FLs, and will normally involve long exposures. Planetary requires plenty of aperture and long focal lengths. These are frequently incompatible so you're looking at different scopes for each in all likelihood!

    You can do a lot of AP without a telescope at all, just a tracking platform for your existing camera and lenses (the R5 isn't ideal for AP compared to the R6 owing to smaller pixel size, but is still a superb camera for it). This can be ideal for wide field.

    If you're not sure then I'd start with widefield as you can do this quite affordably and see if you like the hobby while buying minimal specialist kit - you might end up with a decent fast lens, a platform, dew heater and so on. What's your background with observing?

  7. 1 hour ago, Barry-W-Fenner said:

    I am not having much joy with our council. Sent two polite emails with pics of the offending lights and still no reply. To make matters worse our council hasn't even managed to join the rest of the country in the 21st century and install led lighting yet. We still have the sodium lights and the sky is a shade of orange. My folks live 2 miles away and have led street lighting as they are under a different borough and the sky is noticeably darker where they are.

    Baz

    The good news is that if you're imaging, sodium line filters are much less problematic than white LED filters, and even for visual should work better. But direct light intrusion still sucks, for sure.

    21 minutes ago, SteveNickolls said:

    In a wee while you will get chance to vote in the county council elections but I doubt any change will happen as regards street lighting. Such a shame the dimmed lights etc got put backon at night.

    Regardless of what your local councillors are, politically speaking or otherwise, it never hurts to write to them (or your MP) on occasion to raise awareness. If you have specific actionable things to discuss then all the better but making them aware that there is at least a proportion of their voters interested in the matter will certainly make more of a difference than doing nothing!

  8. Are you doing CosmeticCorrection before integration? Best to remove hot pixels before you do integration in PI, at the very least the obvious ones. With a master dark plus/or a 3-6 sigma automatic removal you'll clear up almost all the hot pixels before you hit the integration stage.

    Dithering doesn't actually get rid of hot pixels in subframes without some logic. The reason being that you're moving the observing field. The camera stays static to "itself", if that makes sense, so the logic has to be that if you see the same "hot pixel" in different positions in the sky it's probably wrong so the statistical rejection says it's not actual signal.

    However, if you reject the hot pixels before you even get to ImageIntegration then there's no signal to act upon and the error is rejected entirely.

    • Thanks 1
  9. Thought I'd give this a go, but a slightly different approach - a positive pressure ventilated enclosure.

    My initial bodge job is below. The entire system runs at 5V rather than 12V since it's native for the Pi. Not anticipating any local storage - the Pi is configured to keep everything in RAM so the SD card won't see many if any writes.

    There's a 5V PWM fan which is PWM-controlled from the Pi and a simple I2C-controlled switch which toggles the circuit with a 25W power resistor on and off. I'm definitely going to affix a much larger heatsink to the resistor, since right now it does a poor job of passing heat to the rest of the enclosure.

    The BME280 sensor tucked next to the camera does temperature and relative humidity so can calculate dew point.

    The fan is filtered, as is the exit port, with metal mesh filters to keep out any bugs. The board is a Pi4 because it's what I had handy and the DC stage is a 5V MeanWell encapsulated supply.

    20210317_231515.jpg.dc87253211328b96c1ad284f1e16d758.jpg

    Not going to win any beauty contests, for sure. The big chunk of balsa wood is again a "close enough" fit - I'll 3D print a proper mount at some point! The top surface is covered in aluminium foil to reject the worst of solar radiation in daytime.

    DSC_0027.thumb.jpg.e82c7c48781cb914e0ba012ea88b129f.jpg

    So far it's working OK; I've been working on my own software since it's a ZWO camera rather than the Pi camera (an ASI120MC I had spare) but tests with KStars have gone well.

    On the dew front, I'm actually struggling so far with dew on the outside of the dome. The ambient temperature is around 3-5c at the moment with the heater running around 20s per 3 minutes since if it's left on all the time it rapidly gets very hot (I clocked it at ~180c with a thermal camera after a few minutes). Definitely need a better heater solution but with 5V only the common 12V "ring" heater available isn't going to do the trick.

  10. On 18/03/2021 at 10:38, Nicola Hannah Butterfield said:

    @discardedastro I finally managed to get out the other night with no moon in the sky, Rear of the scope covered with a dark towel, camera covered with a lightproof bag, home-made light shield (though this can be improved on) This is the full frame from my Canon 6D , a quick stack in DSS and a little stretching and a little editing in pixinsight/photoshop. Thank you for your help.

     

    Glad that you've figured out the cause of your woes! The image looks great!

    The mod I made to mine - an acrylic cover for the rear - can be had quite cheaply and the SVG file required for laser cutting via RazorLabs/others is available at

     

  11. I've been using Close Brothers for insurance of my (stored outside, not within an observatory) telescope for a couple of years now - about £9/mo I think. You never know how well these things work till you have to claim, of course, but they seemed quite competent and used to dealing with AP requirements. They do insure against weather damage within an observatory, but can't without some kind of structure.

    Definitely better to get it on the household insurance if possible from a cost perspective, just need to be sure they'll actually cover it in the stored conditions etc.

    • Like 1
  12. 1 hour ago, osbourne one-nil said:

    Update - I was ready to cancel and find a different use for the money when I heard it's now been dispatched from Germany. Whether this is dispatched again, or for the first time, I've given up trying to fathom, but I think it really is now going to get to me!

    Here's hoping it makes it through the UK border/customs/logistics handling. Took me 7 weeks to get an item from a warehouse in Essex to Oxfordshire as it got stuck in an endless loop of paperwork!

    Ironically, after week 5 I gave up and asked for a refund and parcel recall, bought the same thing from a UK supplier (fortunately the other parcel with the hard-to-get bits had arrived by then) and 2 weeks later the thing turned up. 3 weeks later and I'm still trying to get it back to them for a refund. Of course we've now dropped out of all the EU consumer protection stuff - so things like refunds within 15 days of asking, statutory protections on returns are somewhat more jumbled and confused now. Worth being aware of and thinking about!

  13. Last week I spent 30 minutes trying to work out why I couldn't get scope to focus (remotely, sat in the warm) before realising I hadn't actually taken the dust cap off...

    Bahtinov mask on the scope all night I've definitely done before (one of the main reasons I moved away from Bahtinovs for imaging - hard to spot on some targets!)

  14. I've been running the smaller version (which is a linear PSU rather than switched-mode - linear wastes a lot more heat) for a while now in a similar Dribox.

    I prop one end open (so it's got rain cover should the worst happen, but can "breathe") while using it, with the other end clamped down. It gets really quite hot during operation in a near-sealed box, so caution is advisable. You'll probably be fine with the SMPS version you've listed above as it'll dump a lot less waste heat.

    Edit: Since I have it, here's a "photo" of my power supply box about 10 minutes after putting my telescope to bed for the evening taken with a thermal imaging camera - you can see it gets a little warmer than ambient!

    FLIR0189.jpg.46c8a22dfbac53e4976b4c0be20d8759.jpg

    • Like 1
  15. Well, bit of a mixed bag with acquisition the last few weeks, but I got enough data together to feel I had the start of something.

    Acquisition was my usual setup: Sky-Watcher 200PDS on an EQ6-R Pro mount, OAG guided with an ASI174MM Mini, Baader LRGB filters, Steeltrak focuser and Sesto Senso focus motor, Tele-Vue Paracorr coma corrector, ASI183MM-Pro at -10c as imaging camera, KStars/Ekos and INDI software. Postprocessed in PixInsight, with a pretty "standard" processing workflow - I used local normalisation and as an experiment processed entirely as 64-bit floating point data. No deconvolution, just DBEs and a bit of denoising with masks etc, followed by some masked colour saturation and curves to really draw out some of the finer detail buried in the core without oversaturating (I hope!).

    M_1_Crab_Nebula_2021-03_final.thumb.jpg.6845e0f0202b0fbf3cae0a3aafcf20b5.jpg

    Definitely much more pleased with this than the previous attempt using only a night's data.

    • Like 2
  16. 8 minutes ago, Louis D said:

    The retailer may not let you cancel, claiming it has already been shipped and is tied up in customs, continuing the runaround.  That's why I suggested going through your CC company.  Tell their representative what you've told us, and that you'd like to cancel the order if the retailer cannot provide proof of shipment to the CC company.  The retailer should be able to provide the name of their shipping company, tracking number, and the name of their customs broker.

    Most retailers are accepting refunds - one of the audio firms I purchased from on the 3rd Jan delivered the end of Feb (after I'd requested a refund and they accepted it) due to UPS not understanding the recall request, and are now trying to sort out a recall through UPS.

    So yeah, 1-2 month delays certainly "normal" at the moment. We're seeing similar things with work, despite having an entire team of logistics and customs people prepared for it all. General consensus is that at least for 2021/22/23 this isn't liable to improve dramatically; a lot of import checks were unilaterally waived by the UK govt to lessen the impact of Brexit and will come into force soon, which is widely expected to make things dramatically worse.

    • Like 1
  17. 1 hour ago, Quetzalcoatl72 said:

    My main issue is tracking, I can only get 10 seconds with PA, my PA isn't 100% because I'm not guiding, recently purchased an adaptor to use my 120asi to get into guiding as a first timer. I just mentioned this topic as the warm weather is around the corner and the garden is due to be done up and the patio is the main feature because I have limited space to walk around at the moment. Just worried that the patio wont be stable enough.

    I'd 100% spend money on guiding before you start mucking around with concrete.

    You can probably do better than that on your PA, too, depending on your imaging scale. PHD2's drift alignment is pretty much the gold standard in PA but a Polemaster will get you close enough for rock and roll, especially guided. You'll get a huge amount more value out of guiding and accurate polate alignment than making a super thick patio - guiding corrects for a lot more than movement of the ground, such as mount errors and mechanical inconsistencies more or less anywhere in your system. Polar alignment done with tools like a Polemaster is infinitely more accurate in a short period of time than doing it by eye and requires a lot less crawling under your mount, which is always good!

    • Like 1
    • Thanks 1
  18. You don't have to go mad on concrete to make a pier - a hundred quid or two and some shovelling, or another hundred to pay someone to shovel, and you can make a fairly shallow but functional pier foundation. Concrete chemical anchors or pre-set rebar and you're set for a pier mount (which can literally be a steel tube, optionally filled with sand/concrete, right up to fancy anti-vibration models).

    Alternatively, use a wide pad with moderate thickness (say 10cm) and plonk a free-standing pier on it. This will generally perform better than a tripod and be a substantial improvement. Antivibration pads will only help decouple from local noise, generally, like people walking around on softer ground etc etc.

    The other thing to consider is - is your "wobble" measureable right now? Is it an issue? Is it actually due to vibration in the ground or is it your mount? Are you guiding? There may be other areas to improve before ground is a limiting factor. I am still using the steel tripod my EQ6-R Pro came with on soft grass, albeit with no local noise sources, and guiding gets me sharp images down to 0.4"/px resolution...

    • Like 1
    • Thanks 1
  19. Alright, so a very limited dataset (bit of a theme of mine lately as I go hunting for new targets - I think this one's going to get some more hours) - total about 59x120s for a total of about 2 hours imaging time on my 200P/183MM-Pro/EQ6-R setup, mostly shot over warm rooves and relatively low down.

    Processed with PI; the usual manual processing steps I do (cal, cosmetic fix, subframe selection, alignment, stacking, crop, DBE, linear fitting, channel combination), but I used R/B/L as a superluminance given the small dataset (green was unfortunately badly focused so ignored it). DBE'd a second time, masked using a range selection to denoise and selectively saturate the nebula (ideally wouldn't have oversaturated the stars but here we are) before LRGBCombination and final PCC/denoising.

    Making the best of poor data! I think it came out pretty well all things considered.

    M1_Crab_Nebula.thumb.jpg.91b069a4f1f03da29f11c5e11d51db84.jpg

    • Like 6
  20. 17 hours ago, discardedastro said:

    Hm, I'll try manually calibrating. I'm running with the Ekos/KStars scheduler most nights, and for some reason this seems to recalibrate before each imaging sequence (even with "Auto restore calibration" checked in PHD2) - which naturally means I'm not usually calibrating near the equator.

    Well, did a manual calibration bang on the equator/meridian intersection and got similar wonkiness alerts. But only the second time. Backlash as measured by the guiding assistant appears fine (450ms correction pulse). So thinking some mechanical still. I'm not sure what a wobbly OAG/guide camera might look like but that's one possibility (though the revised ZWO one is pretty solid, with grub screws cinched down) and beyond that it's got to be something more fundamental.

    Maybe this is the year I get the courage to do a rebuild and replace bearings etc. It has sat out in wet and unpleasant conditions (under cover, but still) for best part of 3 years now.

  21. My optical train goes: Baader Clicklock 2" focuser, Paracorr, T2 adapter, ZWO tilt adapter, ZWO OAG, ZWO Mini EFW, straight onto the camera - flange distance is different on the 183MM. I can't remember the spacers/threads off the top of my head but I do have a shim or two on the back of the EFW.

    Going Paracorr to M48 rather than to T2 would likely help on vignetting, no doubt there - more relevant for the 071MC which has a slightly bigger sensor than the 183MM.

    The OAG could also go further away from the sensor's light path, which is another consideration to avoid vignetting. I do want to stick to OAG as I've found it much more robust against differential flexure than the guidescope route even with guidescope rings machined as a friction fit from solid blocks of alu!

    I'd ideally like to end up with a rotator in the mix somewhere, but I don't think I can achieve that without replacing the focuser end with a rotator directly and I'm not sure what on the market (if anything) would support that; there's not enough backfocus on the back of the Paracorr to support it even without an OAG.

    • Like 1
  22. 1 hour ago, Merlin66 said:

    I’ve had this occasionally when I’ve calibrated PHD too far from the equator.

    Hm, I'll try manually calibrating. I'm running with the Ekos/KStars scheduler most nights, and for some reason this seems to recalibrate before each imaging sequence (even with "Auto restore calibration" checked in PHD2) - which naturally means I'm not usually calibrating near the equator.

×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue. By using this site, you agree to our Terms of Use.