upahill

Can't take credit for the video but yeah its 60 satellites in a row, the ones at the end of the chain are getting further apart as they climb higher (I think, at least thats my understanding of it). I managed to catch a glimpse of all 60 as they got brighter in the east. The ones at the front are still very close together and look more like this:

Well touch wood it has been working so far tonight. I redid the coupler on the focuser which is going to have to be replaced but seems to be the cause of the majority of issues with that. Its a flexible coupler made of some sort of aluminium putty. Will replace with a solid one I think and some better quality grub screws. Set the binning on guide camera in the driver rather than through EKOS and that has enabled guiding images to come through much better. So im guiding now. It's at 1.78" RA but that could be due to lack of dark subtraction and a combined star/hotpixel. It hasn't crashed yet, and the platesolving server is working perfectly and solves blind in about 30 seconds with no load on the raspberry pi. As far as proof of concept goes im getting happier with it, and I think a more powerful mini pc running ubuntu would be the next logical step. Want to future proof a bit and get something with plenty of USB3 throughput that will run on 12v

Starlink Satellites -

Last bit of Starlink perhaps -

Definately going to get in the way for astrophotography, not that it will in anyway deter any of us I suspect. Just annoy us. It was nice to not be the only one in the garden saying "Wow" for a change though, something to get people looking up 🧐, 12000 of them could be a different story though. I noticed the last few have already climbed enough to not be that visible at all, and trailing quite far behind.

Just watched the chain go past again, as impressed as the first time - and so was the rest of the family who made the effort to come out and see it 😀

Enjoyed watching these earlier, will stay up til 2.30 to try and catch another glimpse I think. No idea what the long term ramifications will be if any but an impressive project and sight either way.

Oooh just caught starlink going past. Made my night since I was busy packing away gear in a huff. Silver lining I guess

About 60% cloud cover here, taking the opportunity to practice some focusing in between clouds but pretty much given up on the idea of any imaging happening anytime soon. Forecasts seem to be pretty much random guesses at the moment.

Unfortunately very common as smaller retails who have to compete on price a lot more reduce capital outlay/risk by dropshipping or using just in time fulfilment from suppliers. 😞 Thinking about it, its actually a part of many businesses now by design, like companies who use Amazon for their warehousing and fulfilment but do own the stock. It's still technically coming from a 3rd party. Have bought parts on website before and had Amazon drivers turn up with it.

To add to what Buzzard75 said I second the bahtinov mask approach - was a game changer for me. You can make your own too so you dont have to buy a ready made one straight away. Theres a few tutorials online for it, and a generator if I recall. I have used card ones and currently using a 3d printed one and they are sooo simple.

I for one would be very interested in hearing more about capture/processing, and the radio astronomy would be interesting I agree.

Going from memory here but I had a similar issue and it was down to the scopes position being so far off from where the software thought it was that it wouldn't allow a sync. My method for getting around it was to reset everything, put the mount manually to the home position, then power off and restart EQMod. Do my first sync close to home position and then go from there. Was a pain but in fairness only happened when I hadn't parked properly the previous night. Heres another thread that describes the process a little better perhaps -

Well tonight was one of those "Right its all going on eBay" nights.... I managed to eventually setup a VM on my server running Ubuntu and after a lot of tweaking got the Astrometry-api-lite setup installed giving me a fully working astrometry clone on the network. Setup the gear, tinkered to get the solving working for quite some time but it did work, so Pi running INDI, Server VM running astrometry and my Windows machine running Kstars/EKOS... Guiding was better tonight, didn't need to bin either which was weird and good. Got focuser to operate, still need to get to grips with the autofocus system (I think this is more a problem with backlash on my setup) - got the mount aligned after a few attempts and a few hard resets. Went for an imaging run as a test, and after the first sub came in Kstars/EKOS just dissapeared. 🤬 It's done this before, but always waits until everything is working as it should. So since its 1am and the clouds have rolled in I decided packing up was the best thing to do until I cool down a little.

Welcome to SGL!! Plenty of likeminded peeps here Clear Skies.

On a similar thread I had thought about changing the camera format to JPEG instead of Native (CR2) or FITS, which does speed up download time from camera, but I can only do that in the backend - and it applies to all settings, so my main subs then become JPEG Im pretty sure its passing a downsampling command through to its built in Astrometry but will need to confirm.

It wont let me bin the main camera. Only the guide camera, and only when guiding. In fact anything other than 4x4 on the guide camera and I get no picture at all.

The exact phrasing of the question has been asked before. On here and googles Think its part of an online astronomy course?

I looked at the Rock64 and the USB3 is an attractive feature. My other consideration was getting the Pi 3B+ which has a slightly faster CPU - and then using something like this hub hat to give me a powered USB hub with 9 extra ports Now if that hat was USB 3 and could go on the Rock64 it would be a winner! Its taking about 15 seconds to transfer a captured sub over the network to the client machine (thats from the moment it finishes on screen to appearing in the folder, so assume that includes download from camera time too) I dont mind that, but it does slow down focusing and guiding. The best solution is just going to be a beefier mini PC, more ram, USB3 all round and then have just the USB3 hub strapped to the scope with one lead down to the PC I think. Il give the onboard plate solving another go. I will need to redownload the indexes I think as I cleared some of them last night to test and then gave up and went to bed 🤣 Connecting remotely to INDI probably removes the need to have a full blown astroberry install on the Pi and I could go for a slicker headless install.

Had another go last night, this time using Kstars/EKOS on my indoors PC, connected to the Pi on the scope over ethernet. Worked fine once I got everything setup. I did have Kstars crash on the windows machine once and not sure what caused that but I got to the end of an imaging run whilst guiding without issue after that. Platesolving online still only worked intermittently, fits header appears to be ok, im considering adding a mirror of astrometry.net to my home server and using that instead - its got more grunt than then Pi and uploading to it will be faster. I got astrometry.net installed on a Virtual Machine, but not the API/website side of things yet. Guiding worked but still having to bin 4x4 to get anything out of the camera which is weird. Until plate solving is figured out I cant really build an alignment model for the mount, so havent bothered trying to sort out parking properly etc. If its clear this evening I might have another go.

As the plonker who only used gorilla glue on 3 blocks and a paving slab (all above ground), amd a 3d printed pier adapter I can confirm this lol. I recently reversed my mower tractor into mine and it dented the mower! In a year of use my polar alignment went from Great to Good in sharpcap and I have little to no issues with wind during imaging. My logic was 70-90kg of support above ground was still 100x sturdier than the tripod it came with. The paving slab helps to spread that leaning load I think should it happen. It looks good painted by the way - I might have to do that as the S.O isn't best pleased with my building supplies statue.

Thanks for this, for testing im going to keep astroberry running on the pi so im guessing I can install this alongside The aim was for something small like the Pi, I had a laptop before but really wanted everything mounted semi-permanent to the OTA. If the Pi can't handle it then I might need to look at a mini-pc or beefier SBC. If im restricted to using a full pc then I may as well stick with what I know Thanks for the tip on setting up Park position, and the FITS header info - neither of which I have configured so could well be part of the problem.

That doesn't look like its going anywhere in a hurry 👍 It wouldn't be DIY without blood! 🤕 Guessing that was a threaded bar injury? Lethal stuff when it wants to be.

This might be the way to go. As you said I didn't think the power draw on the board was that bad, might try and find a better power block for the Rpi though just in case. Network speed isn't an issue as I have cat6 running out to the pier. I have been launching INDI/Ekos using Kstars and assume that it will now work the other way, eg I find some way to start INDI on its own, and then Run EKOS/Kstars on my indoors machine setting the connection to remote. Downloading from Camera to Pi has been a big bottleneck, especially if set to raw. If I lower it to jpeg to speed up focusing then I lose the option to shoot raw for the routines. Lots more tweaking I think but at least the forecast is clear for tonight again. Thanks.

Another night of testing and I think im getting used to it a bit more, it is struggling to run with limited RAM though, issues converting CR2 I think for processing. The main issue is still Ekos crashing when it loses connection to the INDI server - I guess this is likely due to lack of resources too. Might have to wait till I have a more powerful test system than a raspberry pi.