discardedastro

Looking forward to using these tonight if the weather holds - not looking likely but I can hope! OVL 32mm panoramic and 7mm Nirvana eyepieces, plus my first (second hand) green and black piece of glass arrived today, along with a new adapter for the focuser.

I'll echo the HDR comment - this is a perfect use case for a long exposure to capture ISS combined with a short exposure to get the detail in the moon and bridge. It's a lovely bridge - I was involved in running new fibre optic cables across it a few years ago, the trust were very cautious and rightly so!

AI in phone cameras is 99% faking stuff well enough that you don't notice. I'm not a "just what the eye would see" purist but inventing data by asking a neural net to guess what my pixels should read based on what it thinks good stars should look like (for instance) is a bridge too far for me. The death of the camera market has already happened insofar as the "common case" went. What's left is enthusiasts and professionals. These are both huge markets, and those markets are normally after high performance sensors - see the shift to mirrorless and large format sensors in recent times. This is especially true of professional users. Industrial use is ramping up rapidly and undoubtedly a large part of why Sony et al still do mono sensors and will continue to. Autonomous driving, smart home devices, smart city stuff, all these things have volumes in the hundreds of thousands to millions. But the point is that actually manufacturing these things isn't that different from manufacturing smartphone sensors. These are all semiconductor processes done in fabs and that fab line can either churn out 10,000 big sensors or 100,000 small sensors. The price difference is largely down to how much area you take up on a wafer - big sensors are more wafer so you have lower throughput through the system, since all processes are wafer-level till the wafers are broken. Making mono vs OSC will be a more complex production change and I would wager that for at least part of the process there will be a dedicated mono line which will be your capacity limit. In fabs you don't tend to make manual alterations to individual operations, you have a different part of the fab for different process variants. You'll also start to see more fabs able to work with high-end sensor processes in the next 5-10 years. At present this is pretty limited to Sony/Samsung when it comes to volume - e.g. Sony just dumped another $1bn into a new sensor fab (which should be a clue this market's not going away - note that while this is mostly geared to smartphone sensors, those wafers can be for any size of sensor): https://www.eenewsanalog.com/news/sony-plans-multiple-wafer-fabs-address-5g As more fabs on the fringes of the market start to bring processes onboard we will undoubtedly see capacity and capability start to crop up outside of Sony/Samsung. Take a look at the mid-end semiconductor market - there's many many more players now than there used to be because the fab tech has come down in price, fab refurbs trickle down, more demand, new markets, etc. You may not find the cutting edge here but that's the sort of area where the QHYs and the ASIs of the world might well be able to get chips made with "last decade's tech" with weirder requirements accommodated. But in the meantime I don't think Sony in particular is going to stop making large-format mono sensors and these will probably persist for many, many years.

Just to add to this (perhaps too late!) so long as you have a good drawrope in the duct it doesn't matter now. You'll find nylon 2kN drawrope available at lots of places as lots of utilities use it like BT etc. I'd always provide for a local earth at the obsy especially on long runs and treat the obsy as a TT system hanging off your house (which could be TN-S/TN-CS/TT, doesn't matter) with a RCD on the supply at the house end. Local earthing will generally give you a better ground for fault currents and means you're not reliant on the cable for a safety ground. It's also cheap to do - bash a rod in, clamp to the top, bond it to your incoming earth. Put a cap on the top and reinspect every 5 years or so to make sure it hasn't corroded away. This is all notifiable work so you'll need to get a sparky to at least sign it off for insurance etc - may be best to just get one in who can advise properly anyway. I'd not worry about SWA armouring in duct, but I would make sure the ducts are continous and sealed at each end. Either use a plastic cap over wire mesh or use a rodentproof sealing mastic - there's various ones out there, we use Filoform sealing mastic at $dayjob which is pretty well anything-proof. The mastic approach also stops gas/water which can be a benefit.

I'd tend to agree - machine vision is a huge market and still quite heavily monochrome for good reasons. Full frame is a bit of a weirder niche and that may indeed become harder to get sensors for, but there's plenty of large format mono users in industry. I suspect the number of sensors and variety may come down but mono sensors geared up for HDR/low-light aren't going to go away. CCTV is another industry where mono reigns supreme at the high end (and that's a lot of market).

I actually got the telescope partly converted back to visual mode and managed to dig out an eyepiece. Not quite so good here but it was lovely to actually look through things again for the first time in a year or so! Some lovely detail along the terminator regardless of a little bit of wobble.

Bit late to this thread but this surprised me. I have G&H insuring my whole rig left under a Telegizmos 365 cover - not for weather but for eveything else - in a pretty open garden in a quite low risk area. £9 a month, which I'd view as pretty reasonable...

I'll throw my hat in - ESTI have standards for temperature/humidity profiles for outdoor, indoor, and locations like street cabinets and whatnot. Most consumer kit will be fine in sheltered locations but once you start looking at sites with heat-trap you tend to go into industrial rated components. This normally gets you from 0-45c to -5 to 70c or thereabouts. In an observatory in the UK you'd be really working hard to get above 40c ambient even with pretty minimal ventilation, so for most kit you'll be absolutely fine. Humidity is much more of a concern - as David above says you'll normally use fibreglass conformal coated PCBs in electronics if you expect something to get wet, but this is expensive and complex so rarely happens (industrial/mil-spec kit, telecoms etc, yeah - but not your average computer or even most industrial PCs). And there is of course potential for direct impact on scope optics in a more drastic way than heat (hot glass is just hot glass and it isn't like any coatings mind anything but extreme temperatures - so long as everything can expand enough together it's going to be fine). So as long as you have some ventilation, maybe some insulation against the sun (painting it white is the easiest way to reduce insolation, of course, and highly effective), and a means of humidity control, your kit should last as long as it would stored in a home. My current air management plan is a thermostat-operated fan pulling air in through a box filter (exhausting through a more permissively filtered louvre if it needs it, otherwise just trying to achieve slight positive pressure to reduce dust ingress) plus a hydrostat-operated dehumidifier. That's all it'd take to keep things under 30c ambient and 50% RH 99% of the time.

Right, well, it turns out my whetstones were just rubbish. Thanks all for the comments - knowing this wasn't right was what pushed me to fix the problem! 20 minutes with the new diamond pads and I have a 3mm bevel now. I used 60 grit initially and then 200 grit. I might get a finer grit to finish with but it's already OK really, and chip-free - a pass with the 200 took all the remaining little chips out. I'll see if I can get a 400 pad tomorrow and if not may just crack on with the grinding. Edit: for documentary purposes, here's the (still wet) bevel. It's slightly rounded which is fine - the diamond pads are on a very stiff foam which allows for a tiny bit of deflection, allowing a nice soft bevel to form quite naturally by moving the tool around the edge. For the 60 grit I used a downward stroke to avoid causing chips - on the outer edge there were quite a few chips formed, so this was definitely worthwhile.

I've definitely got to come up with a better way of beveling. I just spent a full hour with bucket and coarse whetstone and I have increased my bevel by 0.3mm. So that's going to take me... 5 more hours of bevel grinding. I'm fairly sure I'm going to physically wear through my whetstone before I'm done. Either my technique is terrible or my whetstones are irredeemably clogged (I've been cleaning them after every rotation, though). Grinding away from the surface definitely is lower risk - I introduced a couple of (tiny) chips when I went back to being in contact on both strokes, so quickly swapped back though it's slower going. Edit: I've bought a few "diamond sanding block" things from Amazon, 60/120/200 grit. They look a bit more resilient than the whetstones and have seen some threads on CloudyNights recommending them or similar things. In lockdown, if I can get the materials, rather tempted to get on with upgrading my grinding stand to, uh, an actual stand. Doesn't look too hard to make something that should be able to start out life as a simple rotary table to make some things and working on the mirror easier and then get upgraded to a fixed post machine if I so desired later...

OK, so, question for those who've gone before! I've been working on widening my bevel - realised going into fine grinding it's still a bit small. It's now about 1.5mm wide which I think should be OK for a 8" - it is very slow going to widen it with the 1000 grit or 4000 grit whetstones I have, and I am now nervous about proceeding with the coarser stones as I've had some damage from this, shown below in various angles. It's a chip of approx 0.8mm dia, not very deep, on the inner surface of the bevel. What's the best way to deal with this? Should I widen the bevel another 1mm to absorb the chip, and if so, how best to do this without risking more damage?

Fair - I think it depends on the framing, too. I tend to be imaging targets small enough that the "border" I know I'll lose to dither overlap is pretty large with most of my images. Dither enough and capture enough, of course, and you'll end up with a wider usable field - sort of a small-scale mosaic!

You'll see pretty limited benefit, if any, from drizzling at 0.9px/" scale except with very good seeing. I think you'll find neither me nor wimvb were posting screenshots with a STF applied - wim used a Arcsinh stretch rather than AutoSTF. You'll probably need to use a more careful stretch to draw out detail either in the final histogram etc stretch or in your working STF. Also, do note that automatic STF can mess up white balance if not linked - worth checking to make sure you've got that set right. Dithering, I'd just set the defaults in whatever you use and do a manual dither or two before you start to tweak. You want to achieve enough motion in each dither step that you're moving the objects at least 35+ pixels from their current position (for most apps which spiral or random-walk, meaning you can use pretty large motion steps and thus get a nice big distribution of sensor positions for any given astronomical object). If your step's motion is visible between two subs, you're good. That'll probably help a lot with rejection of more of the pattern noise, too, though you'll probably need to play with PI's integration rejection settings to get that gone.

Good spot - yes, I got that too. But doesn't appear to have a drastic difference on the processing.

Interesting - i think there's some issues with pattern noise which might be throwing off the automatic stuff. Just doing PCC (all default settings, using metadata from image) and then whacking colour sat up 3x to have a look at it gives the below: You can also see lots of spread in the PCC charts - this isn't good, you should usually see a pretty tightly grouped set of points. There's not much in it in intensity between the fainter nebulosity and the background. How many frames stacked was this? Are you dithering the scope during guiding?

I'm still using a couple of rulers as a sanity check! 🙂 never rely entirely on the high-tech...

Finally got around to widening the spherometer centre port a little to accommodate this new, slightly-higher-accuracy indicator I wrangled from eBay; this has a data port so if I get bored I might hook up something to do direct ROC readout and datalogging. ROC is a bit shallow now, so need to do a bit more tool-on-top work to get back into spec, but not a million miles off. 0.42 in the middle, 0.44 at the edge, 0.41 is the target value. Bevel is a bit small, but increasing it is very slow going. Suspect I'm struggling to get all the glass out of the coarse whetstone that came with the Stathis kit; got a fresh 400 grit whetstone in the post yesterday which is a bit fine really, so might try and pick up another coarse one and see if that helps. I'm definitely not at 3mm, though, more like 1.5mm as it is, maybe 2. Will spend some more time on the bevel tomorrow to bring it to at least 2.5mm and then do one final pass of 180 grit - checking with a laser and manual inspection I'm fairly sure I've gotten all the 80 grit done. Rather surprised how quickly I've gotten through 180 grit if I have in fact done so - I've only spent 10-20 minutes grinding I'd guess. Will reinspect carefully before I get to cleaning up for the next grit size. The beeswax is definitely catching some of the 180 grit (visibly) around the edges of things; I've got a hot air gun on the way and will try the "simply melt it under the surface" approach for starters, but have also got another half kilo of beeswax just in case (along with some disposable alu trays to use for heating it up - I'm still scraping it off the pyrex jug I used!)

That kinda is what the Black 2.0/3.0/Vantablacks of this world do. Flocking is still always going to be more effective because of the surface area the material provides for absorbtion.

Tilt would still be my guess, looking at those subs. If you're seeing it at short exposures then if you can introduce a tilt adjustment element into your optical train you could quickly adjust and verify this - they're not terribly expensive adaptors for T2, e.g. https://www.firstlightoptics.com/zwo-accessories/zwo_t2_tilter.html If you've got any slop or slack at all in the optical train - and if you've got spacers in it, there will be some, then that may simply change by geometry of loading forces over the night.

Back to the grind! Got everything good and clean and did about 15 minutes of 180 grit (normal stroke, mixed mirror-on-top and tool-on-top). Tool took a couple of minutes to bed in but looked to be making good contact pretty quickly. Tool after some grinding: And a moderately unremarkable mirror, now sporting a 180 grit finish. It looks pretty consistent, will probably do another 15 mins or so just to ensure it's all covered after checking with the spherometer.

The glow on the right is a very well characterised pattern on the ASI183 models.

There is a good section on this in the Inside PixInsight book but basically integrate up the RGB masters, and then immediately integrate the masters - without pixel rejection and with zero weighting. I think that's all, from memory. Then do all your processing work on SL, R, G and B masters (plus L if you captured it, of course). I've used L, R, G, B => SL before to good effect.

I'd always match darks to subs, since amp glow and other effects can grow nonlinearly with exposure time. You need the darks anyway for calibration so why not match them? Using the master dark is always a good idea as calibration won't necessarily remove hot pixels but the master dark will act as an excellent reference for where to focus CC's attention. The CC algorithm can then make the required corrections with knowledge of the sensor's hot/cold pixels.

Yeah, tricky. I pulled it into PI, built a quick mask to protect the bottom of the image and just used CurvesTransformation (separate G/B channels) and ColorSaturation. I did have a play with dynamic background extraction but not to much avail. I think it's hard to aim for a really neutral sky colour without a good reference. It's a fantastic image though! Here's my attempt at tweaking to remove the turquoise cast: Pushing much more aggressively with DBE and more "creative" curve correction can kill off most of the background but I probably screwed the colour in the process:

iOptron CEM25 with encoders can achieve <.3 RMS unguided for ~10kg but is a good £250 or so more than the EQ6-R. CEM40 can do ~20kg but no encoders and a bunch more expensive. Have to go to the CEM60EC to get something that's quite drastically better than the EQ6-R in my view (RA encoder so ~.3 RMS assuming no DEC motion, on paper, and 25kg payload). But £1k more. EQ6-R is hard to beat at the price point.