An attempt at setting up remote observatory operation...

[AMcD]

As I have got older I am increasingly disinclined to stand in the cold for hours trying to obtain passable astro-photos.  I am still however, willing to sit in the warm and spend hours trying to obtain passable astro-photos.  I have over the past decade set up an observatory in my back garden using mostly second hand equipment.  I now have a Celestron C8 with Fastar that I bought second hand in 1999, a WO Zenithstar 66mm Apo that I purchased new in the US some ten years ago and a TS6 achromatic refractor that I purchased second hand last year.  The Zenithstar is piggybacked on the C8 and is used with a second hand Starlight Xpress SXV-M7 for guiding with PHD2. I have an original QHY8, again purchased second hand, that I use with the TS6 or the Fastar and Nebulosity.  The mount is a second hand Losmandy G11 with a late Gemini I system attached that I purchased from California some 15 years ago .  The observatory is from Alexander Observatories.

My aim is to be able to control the mount and equipment from my study in the house.  As I work away, it would also be nice to be able to have the capacity for truly remote operation, but one step at a time.  It seemed to me that the minumum requirements for remote operation from the comfort of the house were (a) the ability to reliably control the mount and cameras using a remote computer, (b) the ability to reliably see the equipment in the observatory whilst using it remotely and (c) the ability, ultimately, to remotely control the observatory roof, which control would have to have built in safety features to prevent weather and impact damage to the equipment.

For remote control of the mount and equipment, I decided to use Windows 10 Pro Remote Desktop to communicate between a computer in the observatory and a computer in my study.  Whilst I was attracted to a Windows based system that could be piggybacked on the scopes, I was not willing to pay the high price of bespoke systems like the Primaluce Eagle 3.  The solution I arrived at was a HNSUN industrial mini-computer purchased from Amazon for £290.  It has a 128G solid state drive, 8G of RAM and has Windows 10 Pro pre-installed. It is also small enough to mount on the TS6 notwithstanding that it has four serial ports, four USB3 ports and four USB2 ports, as well as two LAN ports.  Its only drawback is that it does not have the power ports for astro-equipment that solutions like the Eagle 3 Pro give. But it is over £1000 cheaper!  Using a wooden backboard, I mounted the mini-PC on top of the TS6's scope rings.  On the mini-pc is installed the software for controlling the mount and the cameras via ASCOM and Gemini.net scripts.

At present, the computer in my study and the mini-pc on the scope communicate wirelessly over our household broadband, which reaches the observatory by virtue of a wi-fi range extender. Remote Desktop replicates the mini-pc's desktop on the computer in my study, allowing me to control the mount and the cameras. I hope shortly to introduce digital focusing.  It is becoming clear however, that this wireless link is not the most reliable.  In particular, when others are using up bandwidth the link drops out relatively regularly.  As such, I have decided to install a hardwire LAN link between the two computers by running a Cat 7 LAN cable from my study to the observatory.  This will involve drilling a hole through the outside wall and running the cable out from the house to the observatory.

To view the equipment as it being used, I have installed a Reolink wireless CCTV camera in the observatory, which can be viewed wirelessly on the computer in the study. This allows me to keep an eye on things and to make sure the mount is doing what the programmes on the remote desktop are telling me it is doing. It will also in due course permit me to check that the scopes are parked before the roof shuts and that the roof has closed at the end of the session. The wireless link to the CCTV camera seems to be a lot more stable than that to the mini-pc, perhaps due to using less bandwidth.

As to the roof, I am now investigating options for automating the roof.  I think the Talon systems looks to be the gold standard but, again, the price for such a bespoke unit is commensurate with its specialist / dedicated nature.  Any cheaper ideas would be very gratefully received.

The challenge now is to point the scope, centre the target and take the pictures without leaving the house. Wish me luck!
 

[Swoop1]

I might be looking for alternative career options and I am quite cheap.

Just saying like......

[steppenwolf]

Having gone down the remote operation path myself, I can thoroughly recommend it but it will take some fine tuning to get it all working correctly. Good luck!

[AMcD]

You are not wrong Steve.  I just get one thing sorted and then something I thought was sorted becomes 'unsorted'.  Perhaps that is the defining characteristic of the hobby of astronomy, or at least of the hobby of astrophotography.  I laid the hardwire LAN connection today and, after an afternoon of head scratching regarding IP addresses, now have the computers talking to each other over an ethernet connection which, needless to say, is much more stable and much faster than the wireless.

[AMcD]

I have managed to get everything connected.  All the equipment appears to to talking to the mini-pc on the mount and the mini-pc to indoor computer LAN connection remain fast and reliable over a 50M Cat 7 Ethernet cable.  The wireless connection to the CCTV camera has remained stable.  Things have been made much easier with the newer versions of ASCOM and Gemini.net which seem, to me at least, to allow multiple programmes (PHD2, Cartes du Ceil and SGP trial version) to connect concurrently to the Losmandy Gemini much more reliably than they used to.  It occurs to me that when the observatory and indoor computers are connected it should be possible then to connect to the indoor computer from a third computer, via MS Remote Desktop, for truly remote operation.  However, as I say, walking before running!

[DaveS]

I run my obsy remotely. In addition to the computer I have an AAG Cloudwatcher for weather monitoring and a Talon roof automation system. All the IT stuff is connected to a big UPS. There is also a POE CCTV camera for monitoring.

The computer is mounted on the obsy wall, not on the telescope. There is, however a Pegasus APB for power routing and auto dew control.

[AMcD]

The Talon and a Cloudwatcher are on my wish list, as is a Pegasus APB or Hitec Mounthub Compact as I have a lot of power cable spaghetti at the moment.

[AMcD]

My experience of amateur astronomy over the years has taught me that the question should never be "what went wrong?" but rather, "what have I learnt?".   I accept that this is sometimes easier said than done.

I tried running my first remote sequence this evening.  I was, as planned, able to execute most of this from the warmth of my study (having had to go outside to roll the roof off, turn on the Gemini and, in the current absence of an electronic focuser, check the focus on the TS6!) 

The demonstration version of SGP worked very well in allowing me to create an ordered sequence of actions from slewing to the target, plate solving to centre the target, commencing autoguiding and taking the image frames.  SGP slewed the scope to M31 (the specified target) and commenced autoguiding with PHD2.  The problem came with the plate solver.  I am using ASTAP via SGP, which continually failed to solve a plate with M31 partially out of the frame, the Gemini being fairly accurate but not perfect (I could have simply centred M31 manually but I was keen to see the 'centre' feature of SGP in action).  The online version of Astrometry.net did eventually manage to solve the plate but only when I ran the 'centre' function independent of my saved sequence.  I need to look into the failure of ASTAP to solve the plate but I suspect it may have had something to do with the less than ideal seeing conditions (with quite a lot of high cloud) and/or the next problem I encountered.

I am running a classic QHY8 on my TS6 achromatic refractor.  Whilst the QHY8 has a desiccant tube, a heated ring and is kept in a sealed box with a silicone gel heater it (you guessed it) iced up very quickly.  I thought I had solved this problem with careful storage, but it is back.  Needless to say, this eventually produces all sorts of interesting shapes on the image which will definitely defeat a plate solver.  I however, wonder whether this was what was happening even before they became apparent on the image  and impacting on ASTAP's ability to solve the plate.

At this point the clouds rolled in and I was denied any further experimentation, let alone any good imaging data of M31. 

I have learnt that the setup was relatively easy to control using SGP via the MS Windows 10 Remote Desktop/LAN connection between the house and the observatory.   I have learnt that this will be even more the case once I am able to invest in an electronic focuser, filter wheel , field rotator, TALON roof opener etc.   I have also learnt that I need to work on understanding ASTAP and that I will have to find another solution to the icing problem on the QHY8.  I know there are a number of options, including the use of a drill and some rubber tube!  I did however, for the first time, steer my scope to the broad vicinity of its intended target from the comfort of the house. 

All in all, I intend to call this evening a score draw...

[MakeItSo]

Based on what you’ve discussed I’d recommend the Pegasus UPB as the next purchase; this will allow you to recycle USB ports and keep dew under control. USB issues can completely ruin your remote experience whereas the opening and closing of the roof is only needed once. The next thing would be the Talon. It’s well worth it. I used a remote gate opener to begin with but it was near impossible to integrate with everything on the mount. So as is often the case, I bought twice and got the Talon. This is tailor made for ASCOM integration and works really well. The CloudWatcher is last. I say this because I have it but my observatory is in my back garden and I never take the chance of imaging for an hour or two between rain - others may differ on this. In a truly remote set up ( I have one in Spain) the CloudWatcher is essential. My roof doesn’t clear the mount and scopes if they are imaging so I’m a little squeamish about trusting the CloudWatcher over a full night unless the forecast is clear. 

Hope this helps. 
 

(also, it would be good to shut off the IR on the security camera if you have it on. Lastly, have you tried an oversized dew heater on the QHY?)

Edited December 31, 2020 by MakeItSo

[wimvb]

On 27/12/2020 at 11:17, AMcD said:

 It occurs to me that when the observatory and indoor computers are connected it should be possible then to connect to the indoor computer from a third computer, via MS Remote Desktop, for truly remote operation.  

Do you need to use the indoor computer if you also use the 3rd computer for remote access?

If the imaging and control software (drivers, ASCOM, SGP) is installed on the mini-pc, and this has a wired connection to your home network (router), you only need RDP on a computer for remote access. To control your gear while away from home, you need a public ip address and open a port on your router. You can then, from any device that has internet access, control your obsy, without the need for an intermediate, indoor computer.

I've done this installation for a friend who has his gear in my obsy. Some time ago, he initiated a sequence while travelling on a bus. He used his mobile phone as a hotspot, and his Mac laptop with RDP to control his telescope. ASCOM and SGP are installed on a mini-pc in the obsy. The advantage of having SGP on the obsy computer is, that if you lose network access, any sequence will continue to run. Whereas, if you have SGP on the remote computer, a lost connection will crash a sequence. This has happened to me on more than one occasion in the past (INDI and Ekos/Kstars).

Just a thought.

[AMcD]

On 31/12/2020 at 19:16, MakeItSo said:

Based on what you’ve discussed I’d recommend the Pegasus UPB as the next purchase; this will allow you to recycle USB ports and keep dew under control. USB issues can completely ruin your remote experience whereas the opening and closing of the roof is only needed once. The next thing would be the Talon. It’s well worth it. I used a remote gate opener to begin with but it was near impossible to integrate with everything on the mount. So as is often the case, I bought twice and got the Talon. This is tailor made for ASCOM integration and works really well. The CloudWatcher is last. I say this because I have it but my observatory is in my back garden and I never take the chance of imaging for an hour or two between rain - others may differ on this. In a truly remote set up ( I have one in Spain) the CloudWatcher is essential. My roof doesn’t clear the mount and scopes if they are imaging so I’m a little squeamish about trusting the CloudWatcher over a full night unless the forecast is clear. 

Hope this helps. 
 

(also, it would be good to shut off the IR on the security camera if you have it on. Lastly, have you tried an oversized dew heater on the QHY?)

Many thanks for this excellent advice.  Happily, I have a birthday coming up so I am hoping that Pegasus will be the beneficiary of some spending by those near and dear to me!

With respect to the Talon, my roll-off observatory was built by Alexander Observatories some years ago now.  The roof runners are very solid and have lasted those years well but they comprise plastic castors running in wide aluminum channels rather than metal v-shaped castors on metal rails.  In the circumstances, the roof can exhibit some lateral 'play'  when it rolls off and back on.  Am I likely to need a v-shaped wheel / rail roll off mechanism in order to use the Talon reliably?

[AMcD]

4 hours ago, wimvb said:

Do you need to use the indoor computer if you also use the 3rd computer for remote access?

If the imaging and control software (drivers, ASCOM, SGP) is installed on the mini-pc, and this has a wired connection to your home network (router), you only need RDP on a computer for remote access. To control your gear while away from home, you need a public ip address and open a port on your router. You can then, from any device that has internet access, control your obsy, without the need for an intermediate, indoor computer.

I've done this installation for a friend who has his gear in my obsy. Some time ago, he initiated a sequence while travelling on a bus. He used his mobile phone as a hotspot, and his Mac laptop with RDP to control his telescope. ASCOM and SGP are installed on a mini-pc in the obsy. The advantage of having SGP on the obsy computer is, that if you lose network access, any sequence will continue to run. Whereas, if you have SGP on the remote computer, a lost connection will crash a sequence. This has happened to me on more than one occasion in the past (INDI and Ekos/Kstars).

Just a thought.

Many thanks wimvb - I had not thought of this but it is an excellent solution.   Given the weather we seem to be having at the moment it would be very good to be able to take advantage of good conditions immediately they arise from wherever I happen to be.

[AMcD]

In circumstances where lockdown requires me now to work mainly from home, with things therefore more easily to hand during the week, I have decided to spend my evenings now trying to improve the mounting of my mini-PC and the new Pegasus Power Box Advance that arrived this week.  I have never been very happy with having the mini-PC mounted on a wooden backboard.  It is not very stiff and, to be frank, it is not very attractive.  Against this, I could not find a metal plate that was the right size and the plates that are available are quite expensive.  So I have decided to try my hand at fabricating one myself.  I have ordered a small slab of aluminum (150mm x 400mm x 8mm and £23 on eBay) and will now attempt to drill it with all the holes I need plus sufficient for future expansion / re-organisation of the gear.  The first issue that I have run into is that the slab is pretty heavy.  In the off the shelf examples this is addressed by liberal drilling and milling to remove large amounts of surplus metal whilst maintaining structural integrity in the plate.  Given that I have a power drill and a small drill press to put it in, I suspect I will be limited to drilling holes!   Now the Pegasus PBA has arrived I am also going to try using a dew heater to address the frosting issue on the QHY8.

 

 

[AMcD]

I have spent the last two days working on the metal plate for mounting the mini-PC and Pegasus PBA.  I am pretty happy with the results, although still concerned about the weight of the finished product.  I may try and do some more drilling / milling in the future to try and get rid of some more of the weight but am at present limited by my basic tools (and absence of metal working experience!).  In the meantime, you could drive a tank over it and it would probably still not flex...

 

 

 

[AMcD]

On 31/12/2020 at 19:16, MakeItSo said:

Based on what you’ve discussed I’d recommend the Pegasus UPB as the next purchase; this will allow you to recycle USB ports and keep dew under control. USB issues can completely ruin your remote experience whereas the opening and closing of the roof is only needed once. The next thing would be the Talon. It’s well worth it. I used a remote gate opener to begin with but it was near impossible to integrate with everything on the mount. So as is often the case, I bought twice and got the Talon. This is tailor made for ASCOM integration and works really well. The CloudWatcher is last. I say this because I have it but my observatory is in my back garden and I never take the chance of imaging for an hour or two between rain - others may differ on this. In a truly remote set up ( I have one in Spain) the CloudWatcher is essential. My roof doesn’t clear the mount and scopes if they are imaging so I’m a little squeamish about trusting the CloudWatcher over a full night unless the forecast is clear. 

Hope this helps. 
 

(also, it would be good to shut off the IR on the security camera if you have it on. Lastly, have you tried an oversized dew heater on the QHY?)

Thanks for the tip re a dew heater on the QHY8 MakeItSo.  It has solved the icing problem very nicely.

[AMcD]

Appalling picture of the center of the Rosette Nebula, dragged kicking and screaming from amidst the light pollution, BUT it represents the first time I have been able to successfully operate the observatory on a remote basis (apart, of course, from having to go outside and roll the roof off and take off the caps!)

[AMcD]

On 02/01/2021 at 10:43, wimvb said:

Do you need to use the indoor computer if you also use the 3rd computer for remote access?

If the imaging and control software (drivers, ASCOM, SGP) is installed on the mini-pc, and this has a wired connection to your home network (router), you only need RDP on a computer for remote access. To control your gear while away from home, you need a public ip address and open a port on your router. You can then, from any device that has internet access, control your obsy, without the need for an intermediate, indoor computer.

I've done this installation for a friend who has his gear in my obsy. Some time ago, he initiated a sequence while travelling on a bus. He used his mobile phone as a hotspot, and his Mac laptop with RDP to control his telescope. ASCOM and SGP are installed on a mini-pc in the obsy. The advantage of having SGP on the obsy computer is, that if you lose network access, any sequence will continue to run. Whereas, if you have SGP on the remote computer, a lost connection will crash a sequence. This has happened to me on more than one occasion in the past (INDI and Ekos/Kstars).

Just a thought.

With thanks to Wim I have now set up the ability to remotely access the observatory over the web with Remote Desktop Pro from anywhere in the world on any device that has Internet access.  It was not too hard to set up.  I used the following stages:

1. I ran a Cat 7 ethernet cable from our BT internet router to the second LAN connector on the mini PC in the observatory (the first LAN connector is used for the peer to peer connection between the observatory computer and the computer in the house).

2. I then established the IPV4 address for the mini-PC in the observatory on that second LAN connection.

3. I then logged onto the BT router and set up port forwarding so that whenever the RDP port (3389) is accessed on the router over the web the router automatically forwards that port to the IPV4 address, and only that address, on the on the observatory computer via the second LAN connection.  The effect is that any RDP connection made to the router over the web will be an RDP connection to the observatory computer.  It  is possible to set up further connections to other computers on the 'home' network.

4. Finally, because the IP address needed to access the BT router via the web is not static and can change from time to time, I then set up a free DNS domain name for the router which will always be the same whatever IP address BT assigns to our router in the future.

Now all I have to do is enter the DNS domain name in the RDP dialogue box on any device connected to the Internet and I am connected via RDP to the mini PC in the observatory.  I think I am going to retain the peer to peer LAN connection between the computer in the house and the first LAN connector on the observatory  mini-PC as I can use  it when I am at home without relying on the Internet.  I hope however I am one step closer to being able to use the observatory when I am away (I still have to sort out an automated roof, Bahtinov masks, scope covers etc.)

Although at home this evening, I am now going to try an imaging session using the web based connection to test it out.

 

[wimvb]

1 hour ago, AMcD said:

3. I then logged onto the BT router and set up port forwarding so that whenever the RDP port (3389)

Port 3389 is the most common RDP port, so any intruder would try that first. You can have RDP through any port, but you have to specify it when connecting. For example, if you open port 54321 on your router for RDP, and your public IP address is 111.222.333.444, you connect with:

111.222.333.444:54321

Your router will route any incoming conection to your RDP assigned address: 192.168.1.999:3389

The advantage of also changing the port is that any intruder will need to know both your IP/domain name AND port to gain access.

Edited April 25, 2021 by wimvb

[AMcD]

Having watched clear nights at home come and go whilst working away, I have decided to go the whole hog and attempt to move the observatory to full remote operation.  Having set up LAN and WAN connections as above, I have purchased a second hand 19" rack on eBay to hold all of the electronics needed.  Rupert at Astrograph is proving himself full of extremely helpful and very prompt advice ahead of my purchasing from him the bits I will need to automate the roof, scope cover and to operate the observatory from afar.   I also intend to upgrade my roll-off roof to make sure that it works well with a Talon system by changing the rolling gear and the frame onto which the roof rolls off.   Having spent weeks looking into this development I am in awe of people who have done it.  I am doing it in my back garden but setting up such a system 100s or 1000s miles from home is a real leap of faith.  Any "top tips" gratefully received...

[sloz1664]

23 hours ago, AMcD said:

Having watched clear nights at home come and go whilst working away, I have decided to go the whole hog and attempt to move the observatory to full remote operation.  Having set up LAN and WAN connections as above, I have purchased a second hand 19" rack on eBay to hold all of the electronics needed.  Rupert at Astrograph is proving himself full of extremely helpful and very prompt advice ahead of my purchasing from him the bits I will need to automate the roof, scope cover and to operate the observatory from afar.   I also intend to upgrade my roll-off roof to make sure that it works well with a Talon system by changing the rolling gear and the frame onto which the roof rolls off.   Having spent weeks looking into this development I am in awe of people who have done it.  I am doing it in my back garden but setting up such a system 100s or 1000s miles from home is a real leap of faith.  Any "top tips" gratefully received...

Which automation software are you using, as your hardware is only as good as the software driving it. Personally I use Voyager and have automated it to my specifications via dragscript and it works exceptionally well. I know there are other software packages on the market but trust in the software you are using is paramount with regards to automation.

Steve

[AMcD]

I am still looking at this.  At present I operate the observatory remotely from the house via Windows Pro Remote Desktop and use SGP to control most things, although I am told that when one starts to operate remotely at far greater distances from the observatory a server based remote desktop is less data hungry and more responsive.  I am going for remote operation rather than full automation, although I do want to automate the end of the sequences to the extent that the scope parks itself and the roof shuts at the end of a session.   I will take a look at Voyager when I am weighing up the way to go.  Many thanks for the recommendation.

Alistair