Micro Observatory Project

originally posted by: JadeStar

originally posted by: tanka418
I learned some exciting stuff this morning.

I have bee discussing this project with a fellow ATS member "JadeStar", who is actually an Astronomy student, and of course, far or knowledgeable that this old engineer.

This morning in a message she asked IF was interested in Exoplanet research, and that my 8 inch telescope would be large enough to detect Jupiter sized planets.

Well, the reality is that I am interested, in that aspect, and, I suspected that a 8 inch telescope, whose resolution is on the order of .6arcseconds "should" be able to detect a Jupiter sized planet...though I hadn't completed the research on that as yet.

So it now appears that I have yet another feature to add to my system; "Exoplanet Research."

After reading a couple of associated papers, and learning of actual software from NASA to help with this, it has become clear that this would only be an added "method" in the Telescope Time allocation area of the system. This hasn't even bee started yet, so...this method, to allow the collection of data from a specific star on a regular basis, can easily be added to the initial specification...pretty much as I imagined.

And like I said, if you need any help with software packages built for transit detection let me know. I see you found the one from JPL/NASA which a lot of the amateur exoplanet community has used or modified

Thank you...actually if you have that link it might be very helpful.

I read the two documents you linked...very interesting, and not unlike what I thought. Though I would like to look at NASA's package if possible, and hope that it is, or can be made, Windows compatible. But, all I would really need would be the source code...

One suggestion you might want to consider is that when you have good seeing conditions and no one has put in a request in the system queue or is controlling it in real time you have a transit target list so the telescope skews to those stars to stare at when it's not in use.

That way you're always collecting data.

Ahead of ya on that...I haven't started the scheduling code yet, so it's still in the pure design phase...this functionality is already part of the design specifications.

Also are you familiar with python?

Yes, though it's not one of my favorites...the software I have to plot stars in 3D space used Python as a way to create the basic object, color, and position it. The system also uses c# to retrieve star data from SQL server and convert to X,Y,Z coords...

originally posted by: BGTM90
a reply to: tanka418

Are you going to place the camera at f2? I live down the street from Starizona and their hyperstar is quite amazing. 10 second exposures look like hours of data collection. I know Celestron makes an f2.2 astrograph but its in like the $9,000 range and Starizona gave me a $5,000 figure for a complete 8" imaging system including software (a little more or less depending on the camera.) I'm saving up should have it in a month or 2.

looked it up the Celestron Rowe-Ackermann Schmidt Astrograph is 3,500 but that just the scope you still need the tracking stand, camera, and software.

My system will employ a Celestron 8" Schmidt-Cassegrain system, and a Nightscape 10MP camera. the system costs around $3100 for everything. However I won't be using their software, and I will be adding additional automation and precision to the system. I'll be spending a wee bit more on the computer than the telescope.

You can see the Telescope here: www.celestron.com...

The computer will be a 64 bit Dell system with windows server 2012, and SQL server 2014. The computer will allow me to create software that can perform a vast array of special functions; like returning to the same star on a 24 hour period (very precise 24 hours), and "image" an object. I will also be able to implement some sort of "auto-focus" (as well as a manual one for users), by including a relatively complete computer vision library, astrometric database, and some other stuff, there are virtually no limits to what I might be able to do with this machine.

Here's where you can get the source code:
oscaar.github.io...

It some of it was originally developed for NASA's Kepler mission:

A new program will let amateur astronomers detect exoplanets – worlds outside our solar system – by observing nearby bright stars and recording faint dips in their brightness caused by transits from planets in orbit around them.

The program is called Open Source Differential Photometry Code for Amateur Astronomy Research (OSCAAR) and was developed in part with NASA funding.

From NASA's Kepler mission, we know there are potentially thousands of exoplanets or more," said Brett Morris, a research associate at NASA's Goddard Space Flight Center in Greenbelt, Md., who is lead developer of the OSCAAR program. "These planet candidates were discovered by looking at the brightness of thousands of stars over time. A certain fraction of the planets orbiting those stars are aligned such that they transit the star -- that is, they pass in front of the star as seen from earth. The transit will block out just a small amount of that star's light when we view it from Earth. If we measure that star's brightness over time, it will change by up to two or three percent, which can be measured by the commercial-grade detectors that many amateur astronomers and small observatories at academic institutions already have."

Kepler has a clear view from its location in space in orbit around the Sun, far from the distortion produced by Earth's atmosphere. However on Earth, atmospheric variability, such as thin clouds passing overhead, make measuring the brightness changes due to planetary transits more complicated, according to Morris. "The purpose of a differential photometry code – the differential part – is to compare the changes in brightness of one star to another nearby. That way you can remove changes in stellar brightness due to the Earth's atmosphere. Our program measures the brightness change of all the stars in the telescope's field of view simultaneously, so you can pull out the change in brightness that you see from the planet-hosting star due to the transit event."

That bolded paragraph is genius. It's almost like amateur adaptive optics.

Amateurs who would like to try the program need a telescope equipped with an electronic light detector, called a charge-coupled device, or CCD, and software capable of reading the output from the CCD with a computer running one of the following operating systems: Windows 7 and up, Mac OS X 10.6 and higher, Ubuntu 12 and up or similar Linux distribution.

"We're not saying the program will give groundbreaking results or science competitive with Kepler, unless you adapt OSCAAR for that purpose," said Morris. "But the observations can be very satisfying knowing that you're watching other planets, and we hope that OSCAAR users will be inspired to take their exoplanet studies further after they get a taste for photometry."

I may have some ideas how to modify its effectiveness for use on your 'scope.

According to Morris, people will detect mainly hot-Jupiter type exoplanets around nearby stars. These are large planets, so the transit will be substantial enough to be detectable. Also, hot Jupiters move quickly since they orbit close to their parent stars, so the transit will proceed rapidly enough to be seen during an observing session of about six hours or so on a typical night.

The number of stars people could search using the program depends on the quality of their equipment and the viewing conditions at their location -- the darker the night sky, the more stars they will be able to analyze.

So... How dark are your skies?

"People should be able to make measurements of maybe about a dozen bright planet-hosting stars even from urban areas with heavily light-polluted skies," said Morris. "We've successfully used the program at the University of Maryland Observatory in College Park, which is located within the beltway of Washington D.C., and we have pretty miserable light pollution compared to some of the big professional observatories," says Morris.

Morris hopes astronomy students and educators will use the program. "With the Kepler mission, exoplanets have become a hot topic in astronomy," said Morris. "More and more students are interested in doing their own observations of them, but there's a shortage of mentors who have experience doing that because it's a newer sub-field in astronomy. Our program is meant partly to help fill that gap by putting a tool out there that has lots of tutorials so undergraduates can pick it up on their own."

Which I have. Too bad the skies in Washington state are not clear for most of the year though.

The team also encourages people to modify the program. "With some straightforward adaptations to this code, it can be applied to observations of some other astrophysical phenomena," said Morris. "Anything in the sky that changes brightness over time could be observed with OSCAAR.

For example, variable stars can become a few times brighter in just one night, a change many times greater than what is seen from a star with a transiting exoplanet. Alternatively, some asteroids are irregularly shaped and rotating, so the cross-section seen from Earth gets bigger and smaller, causing the amount of light from the asteroid to increase and decrease, which can be detected with OSCAAR. We've tested the program in these situations and it works pretty well, but there are always slight tweaks that would make it work even better, so we encourage people to make them."

"We're hosting the program on 'GitHub', a website for collaborative code development, so it's easy for people to make a copy of the code, modify it, and offer it back to the user community," adds Morris. "Also, we're issuing the program under the 'MIT license' for open-source code, one of the most liberal licenses you can put on code so that other people can use it for whatever they want. They can remix the code, adapt it into other works they've already created or create new things around it, and even use it in their own proprietary software."

The team includes Morris, Daniel Galdi, and Dharmatej Mikkilineni of the University of Maryland and Luuk Visser of Lieden University/Technical University, Delft, the Netherlands. Development of the program was funded in part by the NASA Astrophysical Data Analysis program.

Nancy Neal-Jones / Bill Steigerwald
NASA Goddard Space Flight Center, Greenbelt, Md.
301-286-0039 / 5017
[email protected] / [email protected]

originally posted by: JadeStar
Here's where you can get the source code:
oscaar.github.io...

It some of it was originally developed for NASA's Kepler mission:

"The purpose of a differential photometry code – the differential part – is to compare the changes in brightness of one star to another nearby. That way you can remove changes in stellar brightness due to the Earth's atmosphere. Our program measures the brightness change of all the stars in the telescope's field of view simultaneously, so you can pull out the change in brightness that you see from the planet-hosting star due to the transit event."

That bolded paragraph is genius. It's almost like amateur adaptive optics.

Actually, that is brilliant...And, takes care o an issue I was thinking about once in a while... Atmospheric distortion can be responsible for more "noise" than I would want to try to deal with...this offers a good solution

I may have some ideas how to modify its effectiveness for use on your 'scope.

I'd love to hear your ideas...

According to Morris, people will detect mainly hot-Jupiter type exoplanets around nearby stars. These are large planets, so the transit will be substantial enough to be detectable. Also, hot Jupiters move quickly since they orbit close to their parent stars, so the transit will proceed rapidly enough to be seen during an observing session of about six hours or so on a typical night.

So... How dark are your skies?

actually, the larger, faster moving stars world probably e a little better or "business", easier to find, quickr gratification for the user...


It is reasonable dark here. I live near Kaufman, Tx...a rural agricultural county about 35 miles south east of Dallas. Kaufman is a small town of only a few thousand, and is about 2 - 3 miles from home. We have 10 acres...

It is very dark in all directions except West and a bit north. the glow from Kaufman is about the same as the glow from Dallas, and isn't very significant. In the near field, our neighbors yard light (100 yds away) is as bright as either city. The only other light source of any significance is the lumber yard a quarter mile away.

The eastern, southern, much of the Western, and most of the northern sky are clear of significant light and other obstructions. I even have a virtual parade of UFO's here...looking East at night I can see aircraft, probably landing in Dallas, for 15 - 20 minutes before they pass over head. When I first saw it, I had to stand and watch...didn't take long to figure out what they were...even get some from the south.

Which I have. Too bad the skies in Washington state are not clear for most of the year though.

I moved from Spokane to here in 2003.

Actually Texas has more "rain" days than I would have thought, but 'm looking at something over 50% of available time being clear.

I've found a Weather Station that incudes an option to join a weather network, I'm kind of hoping that this will allow me some small fore-warning on the weather and perhaps a way to gain a small advantage in scheduling...I definitely want it to help gain an advantage over serious weather...this is also tornado country.

a reply to: tanka418

Thanks for your reply that is the same scope and I think the same mount I've been looking at. I was going to save up to get the 11 inch but the guys at the shop said I could trade it in for the bigger scope when I get the money. They said 8 inchs is enough for imaging most objects. But I was wondering what focal point and the focal ratio you were going to set up. I've seen some amazing things at f2 I would really suggest looking into it if you haven't already. Also move to Tucson. Mt. Wightson and Mt. Lemmon are both higher than 9,000 feet and have some of the darkest and clearest skies in the world. Plus we have like 300 clear night a year here.

originally posted by: BGTM90
a reply to: tanka418

Thanks for your reply that is the same scope and I think the same mount I've been looking at. I was going to save up to get the 11 inch but the guys at the shop said I could trade it in for the bigger scope when I get the money. They said 8 inchs is enough for imaging most objects. But I was wondering what focal point and the focal ratio you were going to set up. I've seen some amazing things at f2 I would really suggest looking into it if you haven't already. Also move to Tucson. Mt. Wightson and Mt. Lemmon are both higher than 9,000 feet and have some of the darkest and clearest skies in the world. Plus we have like 300 clear night a year here.

I'm sorry...I was going on about the stuff I wan, and forgot your original question...my bad.

The telescope I'm thinking about is "f10"...

Unfortunately, moving to Arizona isn't an option...We already have 10 acres of Texas, mostly paid for. As for days/nights clear enough for observation...Statistics tell me that there should 135 clear days, 97 mostly clear. That's 63% of the time, and something I'll just have to live with. Though, perhaps I can "teach" the system to recognize when the sky is clear enough to "see"...

Right now, it is enough that I fond a way to begin to compensate for some atmospheric distortions...

a reply to: tanka418

Yeah at f10, wich doesn't sound right for a shmit-cassegrain, you are going to get 1/32 the amount of light and it's going to take you 32 times longer than if you put the CCD at f2. I would really look into hyperstar or an astrogaph with the camera mount in the f2 range. With hyperstar you remove the secondary mirror from the scope put the hyperstar lense in its place and attach the CCD to that. So a 10 second exposure is the same as a 320 second exposure at f10. I think I did the math right I may have not but moral of my story shorter focal ration the faster you get your image. I've seen them do short exposures from the suburbs here and it looks like a full nights worth of data.

originally posted by: BGTM90
a reply to: tanka418

Yeah at f10, wich doesn't sound right for a shmit-cassegrain, you are going to get 1/32 the amount of light and it's going to take you 32 times longer than if you put the CCD at f2. I would really look into hyperstar or an astrogaph with the camera mount in the f2 range. With hyperstar you remove the secondary mirror from the scope put the hyperstar lense in its place and attach the CCD to that. So a 10 second exposure is the same as a 320 second exposure at f10. I think I did the math right I may have not but moral of my story shorter focal ration the faster you get your image. I've seen them do short exposures from the suburbs here and it looks like a full nights worth of data.

Okay...focal length of this telescope is 2032mm its aperture is 203.2 mm; that is 10:1.

Light gathering properties of a Celestron Rowe-Ackermann Schmidt Astrograph telescope isn't even 2X better (actually 1.8) , but, that should mean that it will take 1.8 times longer...in theory.

And, the dis-assembly of the telescope, and installation of "parts" (doesn't matter which ones) is not something that can be automated. Remember, this instrument must be able to stand alone, and be operable remotely over a typical Internet connection...Hell I can't even add Barlow lenses yet since there is no automation I've found that will change the lens, filters I can do, but not lenses.

What I'm trying to do here is build and deploy an inexpensive, yet very useful instrument. Small changes to the system have to be seriously evaluated, and the gain must be very large to justify the additional expense. A small increase in "production" quality image latency is, for now, acceptable.

Once the system is operating and established I will be able to add new equipment that can out perform even your recommendations. This is actually part of the log range plan.

Plus, for now, I wish to demonstrate that inexpensive instruments like what I've chosen, are in fact, reasonable instruments for serious scientific query.

Exoplanet searches...




We will be able to conduct exoplanet searches using a "transit" method...planets the size of Jupiter, perhaps some smaller will be possible using a combination of software technique.

Firstly we will employ a system of differential photometry. What this does is help to remove distortions caused by the atmosphere by "monitoring" the "magnitude" of non-target stars, and using that data as a sort of error correction method. By using the difference between one image and the next, kind of like the difference from one frame to the next, we can correct the magnitudes of many of the stars in our field of view. More importantly, we can correct our target star. Using this method we can improve the performance of our magnitude detection, and thus refine the planets we can find.

Next, we will develop an "adaptive" search algorithm to allow us to find a wider range of planet orbital periods. Currently planets have been found that have orbital periods as short as a few hours, or as long as months and longer.

By creating this search method we can optimize the use of the telescope.

For any of y'all that are interested...

A link to a database table containing all confirmed exoplanets...

dev.wolfmagick.com...

You may also click on the "Telescope" link...then "instrmentalt" to see what will become the Telescope User Interface.

This interface contains a search engine; if you enter a bayer/flamsteed name, or most "common" names of stars, or a Hipparcos identifier, into the text box adjacent to the "magnifying glass" and then click the mag. glass, the system will search the database, and IF found, will display some of the star's properties on the screen. The system makes good use of wildcards, so, you may not have to spell out a name completely; you only need enough characters to make it unique.

There are still some issues with the search, so only about 43,000 stars are available.

an update...I have sort f settled on the combination f technologies going into my Robot Telescope. These technologies will allow this system to behave much like a significantly larger system, while keeping costs down, and the performance up.

Observatory Instruments

The following instruments comprise the observatory's sensor instrumentation.
•8 inch Schmidt-Cassegrain Optical Telescope w/multi-filter remotely operated filter wheel
•Kodak KAI-10100 Color Sensor. Mounted to Telescope (2.5 - 10 MP)
•2 MP All Sky Camera and 180 degree lens
•Spectrometer - Software, diffraction based (diffraction grating in filter wheel)
•Weather Station

Software Instruments / systems
•Differential photometry
•Light-curve compilation
•Spectrometer
•Velocity-curve compilation (Doppler based)
•Field of View Object Recognition
•Main sensor Object tracking
•Field of View Object processing...differential photometry / Light-curve production in "normal" mode. Spectral analysis / velocity-curve production when in Spectrometer mode.
•Background Exoplanet Search System

Just completed our first database update...

Tues. May 12, 2015; Open Exoplanet Catalog update...there are now some 2015 confirmed exoplanets.