Before you get really hungry, let me tell you that you cannot eat a Raspberry Pi; it would not taste very good. You would probably get little chunks of plastic and metal stuck in your teeth. Actually, a Raspberry Pi is a very small inexpensive computer, which along with the Arduino, has been extremely popular with the “Maker” community for the last several years. These devices have been extremely popular as teaching tools for circuitry, basic computing, programming, and even as inexpensive computers for schools in developing countries. The Maker community has been putting them to use developing prototypes for devices that could change the world and also creating fun projects for themselves. I have personally seen Raspberry Pi’s used for a homemade arcade table, for a weather monitoring station at Mount Cuba, for a laser targeting system (science fair project), and several other interesting school related projects. I had played around with a Raspberry Pi before a few years ago, but I had never come up with a good project for one until now.
A month ago (May 13th, 2016), I received an email from Pleiades Astrophoto, the makers of PixInsight, saying that they were incorporating imaging capability into their image processing software and supporting the “INDI Protocol.” I was quite excited by this proposition because PixInsight is a fantastic piece of software for astrophotography processing. It is a little pricey, but the software does a fantastic job and greatly automates a very tedious process. I was thinking that a company who developed software as great as PixInsight would do wonders for the image-taking side of things as well. So I downloaded the “INDI Server” program they recommended and updated my PixInsight software. It did work and looks promising, but it really is not finished yet, so I thought no more about it for a week.
A week later I got to thinking that I might want to investigate INDI server a bit more to see what it really is and how it works. It turns out that INDI server is to all platforms what ASCOM is to Windows. It is a program that connects to all of your devices, like your motorized focuser, your computerized telescope, your digital cameras, and other devices then serves them up to whatever INDI client programs need to use them. That way, each one does not need to have its own drivers and support for all of the devices, nor do the programs need to prevent other programs from getting the information from the devices. The INDI protocol was primarily developed for Linux systems (but also works on Macs and Windows), which explains why I had not really heard of it before. My computer is not Linux.
Then I thought, since I was checking out the INDI Protocol, I might as well see what programs are INDI compatible and if any of them are any good. I was happy to see Cartes du Ciel, a great cross-platform free planetarium program. I was intrigued by a new low-cost Mac OS X program suite from Cloud Makers. And I saw another one that really caught my eye, KStars and Ekos. The last program is for Linux primarily, but there are versions for Macs and Windows, and they have a virtual machine image that is fully configured and free to download so that you can try it out without risking your system, installing software you might not want to keep, or buying a new Linux computer just to check it out. So I checked out KStars and Ekos.
KStars is a planetarium program and Ekos is an integrated Astrophotography tool based on the INDI protocol. For a planetarium program, KStars was fairly simple, but decent, which is of course what you want running in the field where the beauty of the software doesn’t matter. Ekos, however, blew me away. This is a fully featured program for astrophotography that handles guiding, focusing routines, imaging, image sequences, target lists, plate solving, and robotic observatory control all within the larger program KStars. And the cost is the best part. I had thought that paying $50 for the astrophotography program Equiniox Image, the software I had before, was such a good deal since it was so much cheaper than MaxIm DL and The Sky X, which were several orders of magnitude more expensive. KStars and Ekos are FREE! Not only that, they appear to be better than what I had paid for before.
So now I was determined to come up with a way to get KStars and Ekos into the field. At the time, I was just thinking of running KStars and Ekos on the virtual machine on my laptop and running cables to the scope the way I had always done it. But then I heard of some people who were running an INDI Server on a Raspberry Pi and then using their own computer to run INDI compatible software on their computer using either an Ethernet cable or Wi-Fi. This sounded like it might be worth trying out. Then Jeff Lawrence mentioned to me that the Raspberry Pi 3 was only $35 and had a 1 GHz processor and 1 GB of ram built in. That got me thinking that I could fully run the KStars and Ekos program on the Raspberry Pi and just connect to the Pi over Wi-Fi using VNC, basically using my laptop as a wireless terminal for the Raspberry Pi. So I quickly ordered a Raspberry Pi and it arrived May 28th. I showed the Raspberry Pi that night at the AP-SIG meeting along with demonstrating how the KStars and Ekos programs worked on the virtual machine (I had not yet set up the Pi).
Once I had the Raspberry Pi, I was able to work on installing the operating system, configuring it, and getting the programs all set up. This gave me a fun project to work on at home as the school year was wrapping up. I will not go into all of the details of the configuration here, but if you are interested, I have kept a log of how I configured it to work, problems I ran into, and how I solved them in a Word document. Also, once I finish working it all out, I could certainly copy the image of my MicroSD card for whoever is interested. I will say here that I installed a flavor of the Linux Operating System called Ubuntu-MATE specifically configured for the Raspberry Pi. I installed vino-server for VNC purposes. I installed the latest bleeding-edge version of KStars. I am running an INDI server on the Pi so I have the option to either run KStars on the PI or to run an INDI client on my laptop connecting to the server. And finally, I have configured the Raspberry Pi to connect directly with an Ethernet cable, to work with a wifi network, or as a third option, create its own Wi-Fi network for when no network is available in the field and I want to go wireless.
The final step was getting the setup into the field so that I could try it all out, fix any bugs, and hopefully get some great images. Before I ordered the Raspberry Pi, I had tested all of my equipment that I planned to use in the field with the virtual machine to make sure that it would work, but now it had to work with the Raspberry Pi. On May 31st I was ready for a trial run at Mount Cuba. I set up the telescope and then Jeff Lawrence helped me out with the initial trials and testing. We were set up in the Sawin and my telescope was out on the pad. We found out several problems that I needed to fix including configuring the focuser incorrectly. I straightened out the issues and did some other configurations over the next week or so and was ready for another test on June 7th. This time Bill Hanagan and Fred DeLucia helped me out and I imaged several things including the Pinwheel Galaxy, the Black Eye Galaxy, and Markarian’s Chain of Galaxies. Notably, this time we were imaging remotely from the couch inside the observatory. I had so much success that night that I felt confident enough to take the system to Cheslen Nature Preserve with a small group who was out there doing visual observing. In the short summer night, I didn’t have much time, but I did manage to get decent images of the Dumbbell Nebula and the Trifid Nebula before the sun came up. I sent these photos out to the yahoo group after I processed them. The biggest problem that I noted on both of these nights was that I was having some serious difficulty aligning my telescope without a telrad (which I didn’t have yet for this scope). Happily it arrived on Thursday and I was able to take my telescope out to my driveway again on Friday to do some more imaging. I elected to do narrowband imaging due to the moon’s presence along with the light pollution from Philadelphia/Wilmington. That would have worked out really wonderfully if it weren’t for the clouds that showed up just as I finished the Ha exposures, which prevented me from getting my OIII filter data (other than one single frame). So I have a pretty decent Ha monochrome image. During each of these nights I identified issues that I needed to solve and worked on them afterward.
I believe that my experiment is turning out pretty well. I have not yet worked out all of the bugs. One is a problem with plate solving only working with the online solver so far and then only about half the time. Another is a couple of program crashes that I think I experienced because I haven’t got the memory allocated properly or I possibly need a separately powered USB hub. I also need to work out my wiring situation. But, I believe this setup is working extremely well for a $35 computer running a free operating system and free software. And my setup is now wireless as well as independent from my laptop. So now I can set up my imaging routine, disconnect my laptop, and monitor it wirelessly every now and then to make sure nothing went wrong. Once I work out the bugs and get it moving smoothly, I think this will be a great improvement over what I was doing before. Which means there will hopefully be many great astrophotos in our future.
For more information about INDI see: http://www.indilib.org
For more information about KStars see: https://edu.kde.org/kstars/
For more information about Ekos see: http://www.indilib.org/about/ekos.html
For more information about Raspberry Pi see: https://www.raspberrypi.org