Also, to capture a high level of what we talked about architecture-wise via email I'm adding the following.

• Identify the major elements of a ground station
• Break those elements down into kite-level phases
• Identify existing implementations of the major elements and score them (open source has some preference over closed options, but closed options are fine if there are many suitable and easily substituted ones)

Ground Station Components¶

1. Computer System

• Hardware, may be a PC, Laptop, or embedded PC like a Raspberry Pi.
• Prediction and plotting software, I like gpredict, but there are plenty of others.
• Optionally the ability to automatically operate the other two systems. This can be accomplished with hamlib which interfaces with gpredict.

2. Radio System (the receiver and transmitter are usually part of the same radio equipment. This is a "Dual band" transceiver).

• Receiver in the VHF ham range (example OSCAR-52 can be monitored at 145.9250-145.8750 MHz). Reception of signals may be done without a license.
• Transmitter in the UHF ham range (example OSCAR-52 can be contacted at 435.2250-435.2750 MHz). This requires a radio operator's license.
• A licensed radio operator. I include this in the radio system because without them, the set may legally only be used to receive.

3. Antenna system

• A high gain omnidirectional antenna set up for the expected frequency and polarization such as the "eggbeater" and "turnstile" designs.
• Optionally a high gain Yagi array or dish which are directional and must be pointed.
• Optionally a way to point the directional antenna at the satellite using azimuth and elevation data.

This should be a good start and may be expanded.

Level One Kites¶

Computer System:¶

For the computer system, a level one kite has little cost unless we want to run gpredict on a raspberry pi (which is both cool and possible). The trouble is that is is not really all that practical, with the CPU nailing 100% utilization when running with hamlib and serial port emulation to tie it into a radio and azimuth-elevation mount. It would be better to run the system on an existing laptop or PC. A possible level one kite could be predicting an ISS pass and watching for the reflection from the solar panels.

Radio System¶

A level one kite for the radio system of this project may be a challenge. Radios which are capable of doing the job and are able to be controlled by the computer system to automatically compensate for doppler shift cost about $200-300 to start and go WAY up from there. The way I'm going to recommend getting around the problem with the cost of the radio is this: Just as we assume a person has a computer to talk to Arduino for Shepherd, it should be a safe bet that anyone who attempts this project has both a computer, is a licensed amateur radio operator, and owns a dual band (VHF/UHF) radio. I recommend therefore that a level one kite for the radio system should be the receipt of a technician class ham radio license.

Antenna System¶

The antenna design for a level one kite has already been done for us, and the designs for the antenna are free to use, available at http://on6wg.pagesperso-orange.fr/Page%201.html. I think we can do a better job on the instructions though, an perhaps our focus for the level one kite should be refining the design, a construction guide, and usage instructions. This is another thing we can "kitify".While omnidirectional antennas like these lack a certain level of appeal the movement of a tracking high gain rig has, as well as some of the gain, we can save that for another kite.

With the computer running gpredict, connected to the radio and the completed antenna, the licensed user can accomplish the actions outlined in the project scope.

Satellite Tracking Software¶

Here is a spreadsheet comparing several software titles. The limits to this list is that the software must run on Windows, Mac, or Linux (no WinCE or Palm), and must be available to the general public for less than the $200.00 level one kite. Scoring is fairly straightforward, with points assigned for each capability and platform supported. Additional points are awarded for zero cost and open source.

Headings¶

Headings in the spreadsheet and their definitions include:

• Map proj -Map projection for geographic visualization. These come in various states of usability, Refer to the software's respective web page.
• Az-El track -Data to show Azimuth and elevation data during the satellite encounter
• Az-El ctrl -The ability to control an antenna rotator and point a dish or antenna
• Doppler -Data necessary to tune the transmitting and receiving radios accurately
• Radio ctrl -The ability to directly control a radio's transmitter and receiver frequency to compensate for doppler shift.
• Update -Automatic updates to the software's satellite database.
• Win -Windows application available working under XP, Vista, or Seven.
• Lin -Linux application available which is not distribution specific.
• Mac -Mac OSX application available.
• Cost -Cost of a registered, permanent, and full featured version of the software
• OS -Software is published under Open Source license.
• Score -Self explanatory.

Top Three¶

The top three selections chosen from the highest scores in this survey are the following programs:

1. Gpredict from oz9aec.net with 11 points
2. Satellite Toolkit 10 Free from AGI with 8 points
3. Orbitron from Sebastian Stoff with 8 points

Compatible Radios¶

The attached spreadsheet lists all radio receivers and transceivers that are known to work with gpredict (production or beta drivers) and other software which use hamlib. Scoring is fairly straightforward, with points assigned for each capability. Additional points are awarded for the cost of the unit being $200 or less.

Headings¶

• MFR - Manufacturer of the radio equipment.
• Model - Specific model of the equipment .
• Tx - Capability to transmit as well as receive.
• VHF - Tunable to the VHF ham band, necessary for the downlink.
• UHF - Tunable to the UHF ham band, neccessary for the uplink.
• Full Duplex - Able to receive on one band while transmitting on another.
• Portable - Powerable by batteries of vehicle power.
• HT - Handheld transciever, a "walkie talkie".
• New - New prices either represent MSRP or are the best price from discount houses.
• Used - Used prices listed are what could be found with a quick look at the end bid of auctions and want ads in online swap meets.
• Score - Self Explanatory

Top Three¶

1. Kenwood TH-G71 with 7 points. The main advantage is that this excellent radio can be purchased used for less than the $200 of a level one kite.
2. Kenwood TH-F6A with 6 points
3. Kenwood TH-F7E with 6 points

Procuring a Radio¶

Most ham operators are closet space nuts. It is possible to get a better price, particularly at a ham fest or from a private seller, if the seller is informed of who we are, what we represent, and what we plan to do with the equipment. There is also a good chance that a licensed operator will come with the equipment.

Aaron - I'm slowly working through this great information. If we were to create a block diagram based on "Ground Station Components", would this be anywhere close?

I've also attached the original SVG so that it's easy to make changes.

Very close. The antennas require no control input from the computer running hamlib, only the positioning drive does to point the high gain. Another change I made to the diagram was to clarify that the operator controls both the transmitter and receiver, though this may be a moot point since they tend to be combined into one unit. Another minor difference is that there are usually different antennas for the UHF and VHF sides, even if you only see the one device, so I made the word antenna plural.

I will have information on antenna rotators, directional arrays and omnidirectional designs shortly.

Have a look at http://www.ve3sqb.com/hamaerials/oz2oe/ and http://www.amsat.org/amsat-new/information/faqs/crow/. The larger antenna is VHF. An easy way to remember is that as frequency goes down, the parts get bigger.

The antennas require no control input from the computer running hamlib, only the positioning drive does to point the high gain.

The antennas don't have any auto-tuning functionality attached to them that might be controlled by a computer? I'm a little out of the loop on all of this. I really need to get at least a technician class Ham license.

Would it be more appropriate to show that the VHF receiver and the UHF transmitter may be grouped, or does that just clutter the diagram up and make it less clear? Notice the dashed box around the receiver and transmitter.

Well done Jeremy. I like the dashed line to show the potential of the transmitter and receiver being one unit.

Autotuners are usually not controlled, but rather automagically find the matching impedance by hunting for the lowest dissipation across the circuit. They generally are self contained and placed in line between the transmitter and antenna. The gotcha is that just like a series resistor, the addition of a tuner (automatic or manual) will drop the power applied to the antenna depending on how mismatched the antenna is to the radio. The difficulty compounds in higher frequencies to the point that autotuners are rare in VHF and just about unheard of in UHF. Experience has shown most hams that they are better off replacing a mismatched component than trying to match it.

I think we have our block diagram. We can add other bells ans whistles later.

What do you think for the next step? I'm thinking maybe a roadmap of the level one kites you've outlined with a timeline, and metrics on each component of the roadmap for current completeness and whether it's open vs closed source/hardware. In my mind the timeline would only show the relative number of days/months to have each component completed, not when during the year they would be ready. Once we have that roadmap, I'm thinking the architecture study item that we have on the Mach 30 strategic planning outline would be complete, or at least ready for review to see whether it was complete or not.

I think we still need some data on antenna systems, both omnidirectional and directional, as well as a small, lightweight antenna rotator. The rotator will be a much more interesting project since none of the hobby designs are quite up to the task, and the commercial offerings are expensive and primitive. Once this documentation is in place, we can start on the level one kites for the antenna subsystem.

There are several antenna projects I see as level one kites:
1. Construction of an omnidirectional antenna (turnstile or eggbeater designs)
2. Construction of a yagi directional antenna
3. Lightweight antenna rotator using a microcontroller, stepper motors, 3D printed planetary gear train, and enclosure/mount.

Once the level one kites are somewhat solidified, let's come up with a roadmap. It should be noted that a minimum ground station will consist of at least five kites, making the cost around $1000 to go from nothing to functional when everything is said and done.

At this point, I would say the planning is complete, and we can submit it for review. If approved, I would like to create separate sub-projects for each level one kite (most are already in place in ODE) so that development can go parallel.

Guys, this is some great work. I am especially excited to see the block diagram coming together. I think we are off to a great start, but in reviewing this, I realized we skipped the questions from Amanda Wozniak's process (though the conversation has been leaning towards several of them). If you all will indulge me for a few more posts, I would like to run through them and make sure we have covered everything we need to in the requirements phase.

I will post my best summaries/approximations of answers to the questions and we can then discuss them.

Q1. Why are we making this?

Tracking, communication, and command & control of spacecraft are essential capabilities for a spacefaring civilization. Ground stations are the principle links between the spacecraft and Earth used to facilitate these capabilities. The series of ground station related projects will develop the necessary open source hardware to perform these functions.

Q2. Who is this for?

Mach 30 volunteers, Makerspaces and their members, CubeSat designers and operators (including universities).

Q3. How will this be used?

Early ground stations will be used to facilitate voice communications with orbiting satellites (including the ISS, HAM satellites, and CubeSats). Note, not all satellites will support voice communications, in which case the early ground stations will simply provide a means of listening to the satellites' other radio signals.

Q4. What features does it need to have (now)?

Ability to predict opportunities for contact, and the ability to receive and send voice traffic.

Q5. What features does it need to have (later)?

In the future, our ground stations should be able to support sending and receiving command and control data from orbiting spacecraft. Additionally, it would be interesting to investigate the possibility of linking multiple ground stations together to provide larger coverage area (imagine if makerspaces across the country, or world, had ground stations linked over the internet so smaller satellite operators could have greater coverage and contact to their satellites).

Q6. What are the legacy requirements?

Transmitting will require HAM radio licenses (technician level or greater).

Q7. Who's going to build this?

This is a point of discussion, since the current 2013 Mach 30 Annual Plan does nto yet have hardware expenses budgeted for this project.

Q8. How many do we want to make?

This is a great opportunity to develop a kit which individuals, makerspaces, universities, etc could build and set up.

Q9. What is the budget?

Individual level 1 kites have a budget of approximately $200. However, this project will clearly have multiple kites, so the over all budget will go much higher. Current estimates are about 5 kites or about $1000.

Q10. What is the timeline?

TBD - again the current 2013 Mach 30 Annual Plan did not budget for this to go into production so quickly (we did not anticipate increased levels of participation coming so quickly)

Q11. What waste products will be produced by the manufacture and/or operation of this?

TBD - are there any elements that cannot be reused in later kites that would need intermediate disposal? Would any of the prototypes be kept as museum pieces?

Sorry, I think I misunderstood what the scope of our current discussion is. I was shooting for the content of an architecture study, but that wouldn't include the initial questions of the Mach 30 SEP would it?

I had not heard of Wozniak's process, but it fits the way I evaluate projects, which is probably why the questions were pretty much already answered.

Here are my thoughts on each point:

Q1. Why are we making this?

Tracking, communication, and command & control of spacecraft are essential capabilities for a spacefaring civilization. Ground stations are the principle links between the spacecraft and Earth used to facilitate these capabilities. The series of ground station related projects will develop the necessary open source hardware to perform these functions.

Spot on, though for C&C you are going to need a tracking system since those channels tend to be in a higher frequency range and thus will need a dish This capability would probably be a level 2 kite.

Q2. Who is this for?

Mach 30 volunteers, Makerspaces and their members, CubeSat designers and operators (including universities).

With some minor modifications, most of them software, we can add radio astronomy to the mix.

Q3. How will this be used?

Early ground stations will be used to facilitate voice communications with orbiting satellites (including the ISS, HAM satellites, and CubeSats). Note, not all satellites will support voice communications, in which case the early ground stations will simply provide a means of listening to the satellites' other radio signals.

Data capabilities can be added easily to the system, either by demodulating the audio stream (think computer modem), or by using an Software Defined Receiver (SDR) such as the AMSAT Funcube Dongle or USRP and running the demodulation through a software module (GNU radio) using a PC or Laptop.

Q4. What features does it need to have (now)?

Ability to predict opportunities for contact, and the ability to receive and send voice traffic.

Precisely. This means that the directional antennas and antenna rotator should both be reserved for Version 2.0, which would keep the costs to a minimum.

Q5. What features does it need to have (later)?

In the future, our ground stations should be able to support sending and receiving command and control data from orbiting spacecraft. Additionally, it would be interesting to investigate the possibility of linking multiple ground stations together to provide larger coverage area (imagine if makerspaces across the country, or world, had ground stations linked over the internet so smaller satellite operators could have greater coverage and contact to their satellites).

Linking would require standardized equipment with the audio stream being synched by a precision GPS timebase. This will probably be necessary for radio science and video applications as well. For something similar in many ways, have a look at http://www.irlp.net/

Q6. What are the legacy requirements?

Transmitting will require HAM radio licenses (technician level or greater).

Different countries may have other requirements.

Q7. Who's going to build this?

This is a point of discussion, since the current 2013 Mach 30 Annual Plan does not yet have hardware expenses budgeted for this project.

I will be building myself one at the very least.

Q8. How many do we want to make?

This is a great opportunity to develop a kit which individuals, makerspaces, universities, etc could build and set up.

We can kitify the omni antenna, for the current project and the yagi antennas and rotator for Version 2.0

Q9. What is the budget?

Individual level 1 kites have a budget of approximately $200. However, this project will clearly have multiple kites, so the over all budget will go much higher. Current estimates are about 5 kites or about $1000.

...if you add the radio(s) and extra toys that price can go WAAAAY up. We need to keep the Version 1.0 ground station as simple as possible and still operable. I was even thinking about removing the integration with the software. We'd loose doppler compensation (manual works), but it would allow the use of dual band radios not on the list. These tend to be simpler and cheaper, with some available for under $100. The gotcha is that these radios cannot be controlled by the software, so the upgrade path is then blocked. Thoughts?

Q10. What is the timeline?

TBD - again the current 2013 Mach 30 Annual Plan did not budget for this to go into production so quickly (we did not anticipate increased levels of participation coming so quickly)

My timeline for the unit I intend to build and use is within 2013 and in time to do so outdoors.

Q11. What waste products will be produced by the manufacture and/or operation of this?

TBD - are there any elements that cannot be reused in later kites that would need intermediate disposal? Would any of the prototypes be kept as museum pieces?

Hmmm. If you upgrade to directional antennas, you don't need the omnis, though they would still be a fine backup. If a cheap (not on the supported list) radio is used, it would eliminate the ability to later perform doppler compensation, and this would prevent data and telemetry work. Parts were chosen to give the user a clear upgrade path with a minimum of orphans on the way.

Omnidirectional Antennas¶

The attached spreadsheet lists all omnidirectional circularly polarized radio antennas that can be constructed in the average home workshop.
Scoring is fairly straightforward, with points assigned for each dB of gain, for the ability to pick up a signal from zenith, and for simplicity of construction.

Headings and Scales¶

• Antenna type - Construction and form of antenna.
• Typical Gain - Gain of a simple antenna of the particular design in the UHF ham band.
• Top Blind - severity of blind spot at zenith (coverage area looks like a torus instead of sphere).
• Difficulty - The relative difficulty of construction for the average person with regular hand tools.
• Score - Self explanatory

Top Blind Scale
0 - Severe, no signal.
1 - High, -6dB at zenith.
2 - Medium, -3dB at zenith.
3 - Low, Some loss less than 3dB at zenith.
4 - None

Difficulty Scale
0 - Very, nearly impossible to align with hand tools.
1 - High, requires precision cuts and curves.
2 - Medium, requires tuned lengths.
3 - Low, 4 hour job and moderate effort required.
4 - Simple, 5 minute job with the materials, nothing critical.

Top Three¶

1. Eggbeater with 12 points
2. Lindenblad with 11.5 points
3. Parasitic Lindenblad with 11 points

Roadmap¶

Given the details in the discussion, our road map should be comprised of the following steps. Note that not all the steps should be considered kites as they benefit the individual more than the individual, and/or cost much less (possibly free) than the $200 limit of the level one kite.

Version 1.0¶

Goal: To enable contact with space-based assets.
Timeline: One year from start.
Budget/Member Commitment: $425.00
ROI Potential: Eggbeater antenna kits bundled with a satellite communications how-to guide.

1. A member of the Mach30 board becomes a licensed ham radio operator. This will be in an effort to verify that the project criteria are met.
2. A Mach30 board member and another member procure radios from the supported list
3. Construct eggbeater antennas. Level 1 kite.
4. Attempt to make contact with the other member directly between radios. This should fail to prove that without the satellite communication is impossible.
5. Install and configure gpredict software with doppler correction on radio.
6. Make contact with an amateur satellite. This is the first project criteria to be satisfied.
7. Make contact with the other member through the satellite. This is the second project criteria to be satisfied.
8. Make contact with ISS, verified by QSO card. This is the third project criteria to be satisfied.

Version 2.0¶

Goal: To enable better, horizon to horizon communication.
Timeline: One year from start.
Budget/Member Commitment: $400.00
ROI Potential: Inexpensive rotator and kit, X-yagi antenna kits, updated how-to guide.

1. Expand the reach of the ground station by constructing VHF and UHF Yagi antennas Level 1 kite.
2. Construct rotator assembly Level 1 kite.
3. Make gpredict communicate with the rotator control
4. Make contact with space based asset.

Version 3.0¶

Goal: To enable satellite command and control.
Timeline: One year from start.
Budget/Member Commitment: $400.00
ROI Potential: L-band C&C upgrade kit and manual, radio astronomy kit, satellite C&C as a service.

1. Expand the receive capability by using a separate receive antenna and an AMSAT Funcube Dongle and install GNUradio SDR software. Level 1 kite
2. Add L-band transmitter and dish to rotator array possible level 1 kite.
3. Test L-band data capacity on terrestrial assets.

Version 4.0¶

Goal: To enable satellite command and control and audio contact via internet connection anywhere in the world.
Timeline: Two years from start.
Budget/Member Commitment: $2200.00
ROI Potential: GPS disciplined receiver upgrade kit and manual, global satellite C&C as a service, spacecraft global monitoring as a service.

1. Upgrade system to include a precision GPS timebase receiver to synchronize packet reception. Level 1 kite.
2. Interface SDR and gpredict to secure web interface (secure to prevent unlicensed transmission).
3. Create logging server to assemble all packets in a large database, indexed by timestamp and location code. Organizational Level 2 kite.

Project Recommendation¶

Even though a similar project was found http://mygroundstations.com/ which seems to be at a more advanced stage of maturity, it is my recommendation to go ahead with version 1.0 due to it's more universal appeal and portability. The simplified block diagram for version 1.0 is attached.

Aaron - Here are a couple of links to Amanda's presentations on the open hardware engineering process.

• Video
• Slides

I'll comment on the great work you've done when I get a chance to fully absorb it. This study has gone much faster than I ever expected and I haven't been able to keep up.

I am just skimming through things now, but in general, I like where this is going. I would like to do the following before moving forward:

1. Get Greg's input as he is one of the originators of the project and I want to make sure we have not missed something he was seeing
2. Hold a hangout to wrap up the discussion from the forum on the scope/requirements (during Shepard, I found that locking in decisions was much easier working live on a hangout)
3. Move overall conclusions to the wiki to capture the decisions made so far

Thoughts on the process? Am I missing anything in that list?

And Jeremy, no worries on not including the Wozniak questions. I didn't think about it either until I posted something about it. In hindsight, I think we should probably start most hardware work (architecture or specific projects) with those questions to get us off to a good start.

1. I agree. It would be great to have Greg's input too.
2. I think I feel most comfortable with the engineering process when we work things out in the forum and then hold a Hangout as we are doing here. Doing the bulk in the forum gives everyone a chance to participate when they have the opportunity, and gives people a chance to think about things for a day or so to see if any other ideas come to mind. I agree though that the final push to finalize things like this tends to go better in a Hangout.
3. It's a nice setup because I think we've got steps 1 (Initial Questions) and 3 (Block Diagram) of the Mach 30 SEP pretty much done, and step 1 should lead to step 2 (Requirements) without too much trouble. Not to mention the roadmap and other information.

I know this is a tough question to ask until we get the Export Controls Taskforce spun up, but how far can we take this project before we get into export controls land?

I think the point you would cross that line is where you have the ability to fish for, identify, and control an unknown space resource. What this would require is a%H## <CARRIER LOST>

I've set the wiki up for this project, and have tried to combine the initial questions into a document here

Comments or corrections are welcomed.

There was a question asked in Aaron's initial questions above about whether to add the radios into the level 1 kites. My thoughts are that while not ideal, for a level 1 ($200) kite we should stick with the dual band radios that are simpler and less than $100. The goal is to get an entry level system, and the lack of software interface will just have to be something that's rectified in a later version. I would think that would also make the level 1 kite a better place to start for people new to satellite tracking (like myself). If I learn to do things manually with a lower-end dual band radio, the software radio control is not such a black box later on when the computer's handling things for me. That should also put someone in a better position to troubleshoot problems when the software radio control isn't doing its job properly.

Thoughts? Agree, disagree?

I hadn't thought of the learning experiences manual tuning would yield. I believe that you are spot-on in your observations.

Couldn't agree more with the idea that one should learn a process manually before letting automation take over. It really clarifies what is going on in the automated system. My only concern is setting people up to purchase multiple radios... It is not a "blocking" level concern, I would just like to chat about it a little more to make sure I feel like we have done due diligence with regards to stepping up between levels of capability.

I should go look at the radio chart again and get my bearings on cost vs capability since I think my concern would be far less of an issue if the intro radios were low enough cost...

Will post more after I look at radios...

We've filled in steps 1 and 3 of the Mach 30 SEP, so we've got the initial questions and block diagram. Do we have enough information to fill in the requirements document as well? That's SEP step #2. I thought maybe we should look at filling that out before completing the architecture study doc, which is a new addition to the SEP for this project.