#1  
Old 19-06-2013, 12:22 AM
Paul Haese's Avatar
Paul Haese
Registered User

Paul Haese is offline
 
Join Date: Jan 2009
Location: Adelaide
Posts: 9,937
Automation thoughts and observations so far

I have been automating my observatory for what seems like forever to me. It began with getting my PME several years ago and as many of you know I had some trouble with the RA worm being replaced and not installed properly. That was sorted some time ago.

So since then I have worked towards full automation and I can say it has been a step by step approach and for good reason. There is a lot to learn and understand. Software has its own issues and approaches to the idea of automation.

First thing I tackled was camera and guiding: result being getting guided images. Sounds pretty simple but that meant I had to learn Tpoint, AAG Tpoint Mapper and PinPoint as well as Maxim. I look at pointing first. PME's need good pointing to work well and point at near 10 arc seconds even with big scopes. To do that you need to use Tpoint, AAG Tpoint mapper and pinpoint. Each of these pieces of software work in concert with your camera software to give accurate points in the centre of the frame. Each point is recorded in a Tpoint model and those points are used to map out your mounts defects. With enough points you get great pointing. I recommend 150 points as a minimum for consistent pointing night after night. More if you can afford the time. Tpoint was a bit daunting work out at first but once I read the manual for about the 10th time is all started to make sense. AAG Tpoint mapper is easy to use and freeware. Pinpoint (the full version) costs money but is very good at plate solving. Using other programs I had trouble with plate solving, which was very frustrating. Using the above I immediate results with plate solving. Make sure you get mapping and plate solving right first as these will come in handy later.

With guiding, you need consistent reliable results. Round stars is the order of the day and that is what you should be aiming for every time you image. Take the time to working out the guiding parameters. It pays to know what is going on even if that is a rudimentary idea. Guiding will make and break your images every time. I was using PHD for a long period of time and was reluctant to move past it, but eventually I did move onward and I found that MaximDL was so much better at doing things and calibrated so easily compared to PHD. The guide parameters are much more thorough than PHD, which can be quite a mystery much of the time, but some simple experimentation can produce great results. For the most part though the default settings work well.

Once you get past these issues your should move onto automated focusing. I bought the gear for doing this months and months ago and it was only relatively recently that I got on top of sorting this out. In the end it was so easy I wondered why I had not done this earlier. For me I am using a Robofocus on my TSA and Feather touch auto focus on my RC12. Each uses Focusmax to autofocus. Focusmax take images at incremental steps and measure the diameter of the star to determine where the best focus lies. You need to take what are called V curves first but the first light wizard is very good. Make sure you go to focusmax and download the video tutorials. The settings are critical but the videos make it really easy. Having Pinpoint working here is great. I was impressed watching Acquirestar in action. Basically it takes an image of where you are imaging, then plate solves that and then sends the scope to the meridian to focus. Once focus is acquired the scope is sent back the previous plate solved image and then it places the mount back exactly in the same position. Very clever stuff and very simple to get going.

My next issue was working on meridian flips. I am using a rotator made by optec. These are simple in design but you need to get the software going to make the units work. Make sure you download the CDM file for the USB to serial. Without it the command cable will not work.

Now I am finally up to the point of command automation software. I hunted around a bit and looked up quite a few command programs and finally settled on CCD autopilot pro. Pretty much all the others do the same thing but I found I really like the way CCDautopilot works. Sorted most of the teething problems this week and now imaging a target, flipped the meridian and continues onward with data collection. There is still plenty to learn about it and I am sure will be a great asset to my imaging in the years to come.

So that's it so far. Next roof automation and cloud sensors. After that reserve power in case the power goes out and then internet control. In 5 months or so I hope to have it all sorted and imaging any time I like, no matter where I am. I will get back about each of the next steps and let you guys know the issues that I had to sort. The main thing with this is to remember to get imaging practice along the way. It teaches you more about things than just reading. You conquer each step along the way by doing this.

Feel free to ask me any questions and I hope I can answer those for you.
Reply With Quote
  #2  
Old 19-06-2013, 02:06 AM
beren
Registered User

beren is offline
 
Join Date: Oct 2004
Posts: 2,810
Quote:
getting my PME several years ago
geez time has flown by.
Coincidentally I've being going through some of the software mentioned tonight and trawling through the manuals, learning curve but worth it....well more so if I could get that dream backyard ob
Reply With Quote
  #3  
Old 19-06-2013, 02:19 AM
Paul Haese's Avatar
Paul Haese
Registered User

Paul Haese is offline
 
Join Date: Jan 2009
Location: Adelaide
Posts: 9,937
Yeah the time does go by very quickly. How is the EM400? Having and observatory does make a huge difference Stuart. In time I am sure you will have that obs.
Reply With Quote
  #4  
Old 19-06-2013, 02:58 AM
beren
Registered User

beren is offline
 
Join Date: Oct 2004
Posts: 2,810
It's going well Paul Even though I haven't got a Ob {my wife's veggie patch/cashflow will mothball that project for a while} I have all the gear on a mobile platform that can be moved from my shed and set up quickly but still it's hard to fully utilize the software. I can easily imagine how satisfying it would be to fully automated
Reply With Quote
  #5  
Old 19-06-2013, 06:40 AM
DJT (David)
Registered User

DJT is offline
 
Join Date: Nov 2011
Location: Sydney
Posts: 1,452
Nice write up,Paul. Plenty of pointers there.

Good luck with the imaging
Reply With Quote
  #6  
Old 21-06-2013, 08:10 PM
Shiraz's Avatar
Shiraz (Ray)
Registered User

Shiraz is offline
 
Join Date: Apr 2010
Location: ardrossan south australia
Posts: 4,918
what are you planning for the power control Paul? I haven't been able to find much that I would trust for remote switching of multiple mains circuits, although I assume there must be gear out there that does it - maybe I could just modify a Samsung internet fridge.
Reply With Quote
  #7  
Old 21-06-2013, 09:10 PM
Paul Haese's Avatar
Paul Haese
Registered User

Paul Haese is offline
 
Join Date: Jan 2009
Location: Adelaide
Posts: 9,937
For power control I am looking at 1500VA-3000VA UPS. That will be last in the scheme of things but on the investigation list.
Reply With Quote
  #8  
Old 22-06-2013, 05:51 AM
frolinmod's Avatar
frolinmod
Registered User

frolinmod is offline
 
Join Date: Dec 2010
Posts: 573
Now all you need to do is upgrade to TheSkyX 10.2.0 latest daily build, Tpoint Add On and Camera Add On.
Reply With Quote
  #9  
Old 22-06-2013, 10:40 AM
RobF's Avatar
RobF (Rob)
Mostly harmless...

RobF is offline
 
Join Date: Jul 2008
Location: Brisbane, Australia
Posts: 5,716
Sounds like exciting times Paul. Fred was one of the first to crow on the benefits of mega-data here that I came across. I think all imagers can gain from some well thought out and tested software automation. Those of us without observatories setting up in the backyard, or stringing a few nights together in a dark sky site rely on getting as much good data as we can without staying awake beside the scope all night long, and a little automation can go a long way there as well of reducing time setting up an imaging run.

Probably the important thing you've emphasised is the need to build the house of cards one step at a time. No use trying to do it all at once - you have to have the basics in place re tracking and guide star reacquisition or the rest is pointless
Reply With Quote
  #10  
Old 28-06-2013, 09:00 PM
jase (Jason)
Registered User

jase is offline
 
Join Date: Sep 2006
Location: Melbourne, Victoria
Posts: 3,916
Good stuff Paul. Fun journey fraught with danger. Being methodical will get the results you are expecting...and you should be expecting the worst. Build in as many fail safes as possible from the obvious roof closure through to watchdog timers on pc's so that when they freeze or crash the bios automatically reboots them.

Sounds like you've got all the right gear to make it happen, and the important glue to make it all work together harmoniously with CCDautopilot. Great software, shame there's still no web interface. If you are looking to move beyond automated and get into robotic where you upload plans and out spits data (everything in between is handled such as when to shoot the target, close and reopen the roof as bad weather has passed), ACP with Scheduler is your answer. Not cheap, but its depends on what you want to get out of your set up. Includes PinPoint so no need to purchase it separately.

Talking PinPoint and platesolving, playing with ACP 7.0 the other evening refining polar alignment. The ACP script SyncLostScope is an absolute cracker. Within TheSky ASCOM settings, turn off inhibit sync to protect tpoint model, run SyncLostScope script which platesolves using Astrometry.net i.e. it will platesolve literally any part of the sky even if polar alignment and pointing are way off. Once Astrometry.net returns a result, ACP cross reference this with a local catalogue i.e. GSC or USNO-A2.0 then sync's the scope. Enable inhibit sync to protect model again and you're off ready to do another tpoint run with AAG Tpoint mapper. No more trying to find a bright star to first sync on which can be an absolutely pain in the backside at long focal lengths with no finder scope. 4hrs later, 305 points mapped.

I'm looking forward to seeing the productivity levels rise as your adventure continues.
Reply With Quote
  #11  
Old 28-06-2013, 09:40 PM
Paul Haese's Avatar
Paul Haese
Registered User

Paul Haese is offline
 
Join Date: Jan 2009
Location: Adelaide
Posts: 9,937
Hi Jase,

I did look into ACP and found it hard to work with myself. Maybe I had spent too much time with CCDautopilot. Just wondering though doesn't CCD Navigator do the robotic stuff?

I suppose in the end I will pick ACP. For now though I am happy working on more data.

Jase already noticed my output going up. On my third night now on a target with 9.5 hours of Ha and now doing colour with more Ha once the moon rises. Love being able to meridian flip too.

NO matter which program I end up using in the end, this is way better than imaging manually and having to check focus every 45 minutes.

Thanks for taking the time with your comments. Appreciated.
Reply With Quote
  #12  
Old 13-09-2013, 02:57 PM
Paul Haese's Avatar
Paul Haese
Registered User

Paul Haese is offline
 
Join Date: Jan 2009
Location: Adelaide
Posts: 9,937
Update

Bit of an update here. It's been another few months since the last update. So what has been happening?

Command software - CCDautopilot5 has been glitch most of the time. In the 30 odd runs I have done with it, it has only ever completed a run 1 and closed down properly. However that is not to say it has not closed the roof as required, it just tries to start again and again and again.

One thing I know about this is that it is most likely a lot of other little things that are causing the command program to vaulter. So don't take my meaning to be that the program is the fault.

There has also been issues to sort with roof automation and USB's dropping out. I had thought it was to do with the roof motor and RF or signal being a problem. In the end it looks like it was an incompatibility with the software and the MHP. I removed all the USBs going into it and put those cables through the mount. It gave me greater reliability. I still need to use the robo focus unit instead of the MHP to control focus as I think using the MHP is causing problems in CCDAP.
One thing I am going to do is try the trouble shooting ideas on CCDware website.

So at least the roof opens and closes properly.

Next I addressed the Cloud and rain sensor. I went with the AAG and that has been rock solid. It provides reliable data and has shut the roof every time cloud cover has occurred. It will not let AP open the roof if there is cloud about.

The next thing I decided to sort once I had internet on site was being able to turn things off and on via an IP Switch. Being able to connect via the router is important and if you are using Wifi getting the signal sorted with the settings can take a bit but I have a thread here detailing how to do that. Once you get a connection you can control the unit via the main computer or via webpage. I am yet to get the webpage settings sorted but for now I can turn off the mount, weather station, roof and camera. That way if need to reboot anything I can. With the wifi of this unit just remember to put the SSID of the modem router into the IP Switches SSID software. I made the mistake of thinking I had to put the IP switches SSID into that field when in fact it has to be the modem router that brings in the internet.

The latest thing I have done is get a UPS. There are several lines of thought in this regard. You can use a small ups to power the computer and one to power the roof or you can get one to power the lot. I went with the later. The UPS I bought is very expensive. My thinking was to extend the life of the imaging session, so I would not lose the chance to gather more data. Really in the scheme of things this is less important than protecting the gear. So for now I will go with protecting the gear and think about buying the extension packs later it at all.

So at present I am imaging from home, but tentatively going to bed on nights in which the sky is totally clear. I don't yet trust the system and there is still some more wrinkles to iron out before fully trusting the system is real.
Reply With Quote
  #13  
Old 13-09-2013, 04:14 PM
Merlin66's Avatar
Merlin66 (Ken)
Registered User

Merlin66 is offline
 
Join Date: Oct 2005
Location: Junortoun Vic
Posts: 8,904
Paul,
Based on your experience to date....what do you think the chances are of being able to remotely operate a slit spectroscope?
Similar issues to astrophotography BUT
1. We need to identify and find the correct target star
2. The guide optics only see about 10 x 10arc min
3. The target star must be placed (and held) on a 25micron slit gap.
I take it that the plate solving techinque is an important part of your success.
Any comments?
Reply With Quote
  #14  
Old 13-09-2013, 05:01 PM
Paul Haese's Avatar
Paul Haese
Registered User

Paul Haese is offline
 
Join Date: Jan 2009
Location: Adelaide
Posts: 9,937
Wow Ken, I cannot really say one way or the other as I don't have that baseline of experience.

However, with the right mount, with great pointing and with appropriate guide sensitivity I think it would work well in a remote situation.
Reply With Quote
  #15  
Old 14-09-2013, 12:50 PM
ericwbenson (Eric)
Registered User

ericwbenson is offline
 
Join Date: Sep 2009
Location: Adelaide, Australia
Posts: 209
Quote:
Originally Posted by Merlin66 View Post
Paul,
Based on your experience to date....what do you think the chances are of being able to remotely operate a slit spectroscope?
Similar issues to astrophotography BUT
1. We need to identify and find the correct target star
2. The guide optics only see about 10 x 10arc min
3. The target star must be placed (and held) on a 25micron slit gap.
I take it that the plate solving techinque is an important part of your success.
Any comments?
Hi Ken,
You must define remotely operated. If you mean the operator uses the remote system via Remote Desktop (or TeamViewer or etc), then your biggest problems are actually spectrometer adjustments (collimator focus, grating changes) and calibration frames (i.e. swinging the neon lamp into the light path or acquiring diffuse illuminated flats with 150W on axis illumination). This is because number 1 and 2 can be easily solved with a quality mount (AP, PME, A200, Mathis fork, etc) that can be controlled via computer. For example my guide sensor is 6.4'x4.8' and ACP hits it every time (the scope a CDK20 = heavy!, but the mount, A200HR, is heavier!). Obviously a fiber fed spectrometer solves these issues, and IMHO is the preferred way to go for a remote setup.

If you mean remote and computer operated (i.e. ACP script does everything while you sleep) then number 3 is the problem since AFAIK it must be done manually. None of the automation packages can automatically put the guide star on a particular guider pixel (i.e. the slit). Plate solve + pointing updates will get close, but not < 3" reliably which about the width of a slit. However ACP+Maxim is quite customizable and by writing your own vb/java script I imagine it could be done.

I have thought about putting my LHIRES on a OTA in Ark, but as it is not fiber fed, and getting just an imaging rig going is all the bandwidth I've got for the time being, ...

Best,
EB
Reply With Quote
  #16  
Old 14-09-2013, 03:59 PM
Merlin66's Avatar
Merlin66 (Ken)
Registered User

Merlin66 is offline
 
Join Date: Oct 2005
Location: Junortoun Vic
Posts: 8,904
Eric,
Thanks for the feedback.
Many spectroscopists are currently using medium to high end mounts and use remote control ( I'm only using a C11 on a NEQ6pro with Teamviewer) - these can successfully GOTO a target star and place it in the guide field - CdC/Eqmod etc.....using PDH/AstroArt5/ SPCAudela we can easily guide to within a pixel.
Motorising the various elements of the spectroscope (other than grating change-overs - still a manual intervention) has been done by numerous observers.
With observer "control" it is straight forward to positively identify the target star, position it on the slit gap and start guiding.
At this stage is doesn't appear feasible to get full robotic control of a system to work for us.
The "minimal" observer interaction may be as good as it gets - albeit remotely.

Re fibre optics connection - the available sizes of the fibres v's the star image are an issue.
Any spectroscope, with a bit of thought can be connected with a fibre optic. This does reduce the mechanical issues in mounting the spectroscope to the telescope, but doesn't solve the fundamental issues of target aquisition.

Last edited by Merlin66; 14-09-2013 at 04:03 PM. Reason: note of fibres added
Reply With Quote
  #17  
Old 15-09-2013, 01:05 AM
frolinmod's Avatar
frolinmod
Registered User

frolinmod is offline
 
Join Date: Dec 2010
Posts: 573
Quote:
Originally Posted by jase View Post
it will platesolve literally any part of the sky even if polar alignment and pointing are way off.
Note that TheSkyX now has all-sky image link (plate solve) which does exactly that. It does it locally without any need to access astrometry.net. You can take a photo, all-sky image link it, sync on it and then optionally follow up with a recalibrate permanently mounted telescope or recalibrate portable telescope using either more all-sky image links or regular image links.
Reply With Quote
Reply

Bookmarks

Thread Tools
Rate This Thread
Rate This Thread:

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT +10. The time is now 09:28 AM.

Powered by vBulletin Version 3.8.7 | Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.
Advertisement
Testar
Advertisement
Bintel
Advertisement