Automated Observing

[higginsdj]

Anyone interested in automated observing/imaging?

Cheers

David Higgins
E14
Canberra, Australia
http://www.david-higgins.com

[ving]

sounds interesting enough david, what is it? :confused:

[rogerg]

Hi David,

I definitely am.

Last night when I was sitting on the couch doing astronomy (wirless connection to the observatory) waching it take exposures I was thinking "ok, next step - keep this going while I'm asleep".

I became quite keen about the remote & automated observing a year or more ago, started writing my own programs to do it. Ended up with a program that will scan a grid of the sky for me, moving the telescope to each grid cell, taking x expsorues etc. I hit one problem, making it automatically lock on to a guide star so my exposures could be longer than 30sec at 1x1. I wrote a few more programs, one for setting the scope to polar or land mode, another for setting slew speed, and I can't remember the other functions. THe idea was I could make it scan a grid of the sky for 3 hours, then re-scan the same grid, then slew to park position and set the scope to land mode such that it stopped tracking the sky. I never got it all working in sequence together but I have the pieces. I lost interest about there, going back to normal imaging. I'm becoming more keen about it now after having re-located my observatroy in the mean time and now finally having it all working nicely again.

I wrote a webpage discussing the ideas (for the purpose of those who don't know much about it - I know you David do, remember emailing you about something related to minor planets or comets in the past):
http://www.rogergroom.com/rogergroom/esh_rog_item.jsp?Item=29

It doesn't say much new, just discusses it a bit, as I found it hard to find information on the topic on the net, especially in Australia and how to get started.

Roger.

[iceman]

Sounds great Roger.

I know Brad Moore also does a lot of automated imaging.

[higginsdj]

OK, since there appears to be some interest.......

My observatory (Hunters Hill, Canberra) has been operating since 2002 doign Minor Planet (and the occassional comet) research (Astrometry and more recently Photometry).

I am no genius :D so my set up is off the shelf - plug and play.

What can I do with it?
1. I manually open and close the observatory (simple roll top roof)
2. I manually power up and down the telescope, computer and Cameras
3. I manually produce an observing plan
4. At the start of the night I load the plan into the software and press go.
5. In the morning I download all the images from my Observatory computer.

Now the suite of software I use does support observatory/dome automation - I just can't afford the $15K to buy such an observatory dome :)

What the automation does is:
1. AutoFocus the scope
2. Work out if the target is observable.
3. If observatble (ie above the preset horizon) then point the scope to the co-ords
4. Take a pointing image and plate solve it (ie determine where the scope is actually pointed)
5. Synch the scope and re-slew to the right co-ords if required
6. If the exposure is longer than the no guide limit take an image and find a guide star then start guiding
7. Change filter (if required)
8. Take the image
9. Repeat from 1 for the next target or do multiples of the current target (or a combination as detailed int he observing plan)

The software will handle moving targets - ie it will automatically repoint after each image to keep the target in the centre of field.

Hardware I use:
Meade 14" LX200GPS on a Wedge
Meade 4" SCT piggybacked to the main scope
Starlight Xpress MX716 as the guide camera on the 4"
SBIG ST-8E (or 9E depending on what I am doing) on the main scope with an f/3.3 FR producing f/4 on the ST-8E and f/5.6 on the ST-9E (ST-8E has a CFW-8 with BVRI filvers)
Star2000 Relay Box (for autoguider connection)

Software:
ACP4 (main control package)
MaxIm DL 4 (Core product)
Pinpoint 4 (Optional since Maxim includes Pinpoint LE)
FocusMax

Points to note:
My 14" has 120" PE (This is NOT a typo)
My imaging resolution ranges from 1.3"/px to 2.64"/px
The SBIG cameras are parallel and I do NOT use thier internal guide chips
The MX716 is USB
My scope puts up with extreme amounts of dew, frost, ice (gets to -15 c some nights at 90%+ humidity)
I only leave it open to it's own devices IF I am reasonably confident that it won't rain (most of the time these days)
From a research perspective the automation has increased my data capture rate from an average of 40hrs/month to 85hrs/month

I have produced a quick and dirty paper on the subject that is available from my web site:
http://www.david-higgins.com/Astronomy/Automated%20Observing.pdf

Welcome any questions you may have.

Cheers

David Higgins
E14
Canberra, Australia
http://www.david-higgins.com

[rogerg]

Great stuff david, it's that increase in observing hours that I'm after.

I'm interested in more info on how you handle 4, 5 and 6.

I'm currently using CCDSoft together with TheSky for saving images with their co-ords etc, and Astrometrica for matching up the star field.

Unfortunately it's not practical for me to switch from Software Bisque to MaximDL for $$$$ reasons, so I guess you might not be able to offer much help.

Roger.

[higginsdj]

Well for plate solve purposes, Pinpoint 4 is a VB App that can be scripted (VBscript).

Platesolve requires that the software know about the image (pixel scale, orientation, size etc) and then it needs a catalogue. Default is GSC but USNO A2, UCAC1 or 2 or USNO B1 are all handled.

Once the Image is platesolved (it returns image centre, scale, orientation ect) the RA and Dec are passed to the scope and a Synch command is issued. (Both functions can be achieved via ASCOM and Meade Drivers.

The auto guidestar selection is handled by ACP and MaxIm but basically it is a simple matter of taking an image of X seconds, pick the brightest star, scale on time to work out minimum exposure time required then set the guide star position in the imaging software to the X, Y location already derived then issue a start guiding command.

Another automation option for you is Orchistrate which relies on Sky 6 and I think CCDSoft.

Cheers

David

[rogerg]

I'm using USNOA2. I always get confused on the different ones, but I need more stars. The 8gig of USNOA2 doesn't seem to be enough, TheSky sometimes doesn't show a star in the FOV when in fact there is. More common for the guide chip.

Orchistrate - not even worth a 2 second look from what I can see. I have it, came with TheSky6 professional edition. From the time I've spent trying to work it out, it appears extremely limited and to have low capability together with clumsy usability - it took me quite a while to create a relatively simple script. It definitely doesn't support any concept of sync/matching a star field, every image must be individual specified and delays specified seperate to the images, etc.

I'm thinking about CCD Commander, that looks very capable. It relies on TheSky & CCDSoft also, but that's good for me.

Roger.

[higginsdj]

How small is your FOV? USNO B1 has far more detailed coverage (and far more accurate astrometry) but it's 80gb.

I haven't encountered a field yet that USNO A2 couldn't solve on - but you might have problems around the large/bright DSO's since they are generally too bright to have any stars offered nearby.

Cheers

David

[rogerg]

I'm not great with these FOV calculations and the like, but, using Ron Wodaski's calculator:

My scope's details:
focal length=2160mm
aperture=300mm
f-stop=7.2

resulting in:
0.88 arc sec/pixel
7.3' x 10.9' FOV

I stick away from the bright DSO's, hunting down the fainter ones.

Roger.

[higginsdj]

Thats a small FOV and a high resolution but probably what you want/need for pretty pictures :-) I don't image under 1.3"/pixel (ST-8 Bin 1) and my FOV is about 3x yours. My seeing is typically 3-4" (or the wind makes the seeing appear 3-4") but I can sometimes get to 2" - but thats pretty rare. With the ST-9 I image at f/5.6 (2"/pixel).

My guidescope FOV is reasonably large as well producing a scale of around 3.0"/px

Cheers

David

[rogerg]

Yeah, I would prefer a slightly larger FOV (1.5 arcsec/pixel would be good) but this is with the F/6.3 reducer, using the F/3.3 reducer I have never had success: I always have problems with extremely un-flat FOV, distorted stars, etc. I think because of the long image train but not really sure.

So I've stuck with the F/6.3 which gives me nice flat images and nice round stars, trade-off being 0.88 arcsec/pixel.

I'm often surprised by the image quality too (how good it ends up considering all), so aren't too unhappy with the situation. But yes, small FOV, which means I need lots of stars to have anything to match against.

Roger.