High Altitude Balloon Electronics

Last Updated  9/23/2007

Balloon (and Car) Tracking System

Ground Chase Vehicle Tracking System

This seems to change by the minute, but right now it looks like we’re going to have a few substitutions from the block diagram shown below.  The GPS is going to be a USB eTrex model.  The HT is probably going to be some sort of mobile radio.  We’ve had a really hard time getting Tracker2’s, so we hope to use a Kantronics KP 3+.  If not, we’ll try to decode packets via laptop sound card.  If not, we’ll use an iPhone to track it. 

Balloon Power Supply

The design for the power supply is to have two batteries.  The first is an 11.1V / 5000mAH Lithium Ion pack that powers the OpenTracker2 and radio.  The second is a 3.7V / 7200mAH Lithium Ion pack that powers the camera and strobe.  The idea behind this is that I don’t want the camera to kill the transmitter functionality if it takes a while to get to the site.  We want the transmitter operational when we get to the site to allow DF possibilities.

The voltage regulator below provides overcurrent protection on the batteries and provides the +12V for the OT2, +7V for the radio, +3.7V for the camera and +1.5V for the strobe.  LED1 was removed for energy savings purposes. 

Custom Cables

There are a crapload of custom cables involved in the balloon payload.  See below:

Serial GPS to Tracker2 Cable

This cable connects the Garmin GPS18LVC GPS Puck to the Serial port on the Tracker2

DB9F                                                         GPS Puck
(Tracker2 Serial)                                          (Garmin GPS18LVC)

1 – NC
2 ——————————————- TXD (White)
3 ——————————————- RCV (Green)
4 ——————————————- +Vcc (Red)
5 ——————————————- GND (Black)
6 – NC
7 – NC
8 – NC
9 – NC

This cable allows us to connect to the Garmin and program the firmware in the GPS puck:

    DB9M                                        DB9F                                    USB A
(Computer)                                    (GPS)                         (Computer – Powers GPS)
      2 ——————————– 2
      3 ——————————– 3
      4 ——————————– 4 ————————— +5V
      5 ——————————– 5 ————————— GND

Tracker2 to Radio Cable

This cable connects a Yaesu VX2-R radio to the Tracker2:
DB9M                                           Yaesu 3.5mm 4 pin                                  DC Power                                 Shutter Release Relay                     PTT Resistor
(Tracker2)                                    (Radio)                                                  (2 pin Adaptaplug)                     (2 pin jumper header)                     (Inside DB9 shell)

1 ———————————- Ring 2 ——————————————————————————————————- 1K Resistor Side 1
2 – NC
3 ———————————————————————————————————————————————— 1K Resistor Side 2
4 – NC
5 ———————————- Tip
6 ———————————- Sleeve ———————————– GND ————————— Relay Coil –
7 —————————————————————————- +Vcc
8 – NC
9 ————————————————————————————————————- Relay Coil +

Tracker2 to Camera Cable

Click Here for info about the shutter release mod we used.

Tracker2 Camera Shutter Release Hack


For our first balloon flight, we used an Open Tracker OT1 / OT+ / OT1+, which has a high current output that goes high at the time the TNC transmits.  This was pin 9 on the DB9 “Radio” connector.  We used this to power a relay used for the purpose of firing our camera. 

For the second flight, we decided to upgrade to a Tracker2 / OT2 / OT2m OpenTracker 2.  Unfortunately the shutter release option did not exist on the Tracker 2. I suspect this is because that particular output is driven by the uC on the OT1/+, and the T2 has a different uC that may not be able to source as much current as the one in the OT1/+. 

The Solution

Note:  It is totally possible to power the shutter release relay using the onboard high current output FET.  Just connect the coil of the relay to the OUT+ and GND terminals and enable the “Power Control” option in OTCFGWIN with a delay of 0 seconds.  The reason we didn’t do this is that we were already using the POWER option for something else.  We were using it to power a secondary cutdown option and strobe and buzzer designed to help us locate the capsule when it returns to Earth.

Turns out that the red “Transmit” LED lights up every time the Tracker2 lights.  I used this output (marked PTC1 on Scott’s T2 schematic) to also drive the gate of a 2N7000 MOSFET that I added.  I connected the Drain of the new FET to Ground.  This forms a “Low Side” switch that can sink ~200mA, so it can drive most relays in the 5-12V range.  To implement, I soldered the Gate to the side of R31 farthest from the TX LED.  I scraped off the solder mask above the ground plane just to the left of R31 and soldered the Source of the FET to Ground.  I connected a jumper wire from the Drain of the FET to Pin 9 of the Radio connector.

I liked having a quick-connect on the camera shutter release, so I connected the Source of the FET to pin 9 on the DB9 connector (I looked on the schematic and didn’t see it connected to anything).   On the Radio connector side, I connected a wire from one side of the relay coil to Pin 9 of the Radio connector, and a wire from the other side of the relay coil to the Power input to the T2 (pin 7 of the Radio connector).

For the shutter release, I took apart my Nikon Coolpix P4 camera and soldered wires to the shutter release switch.  Typically there are three wires required to fire the camera.  Shorting the common to a “prefocus” lead simulates pressing the button part-way.  It causes the camera to focus.  At this point, shorting the “Fire” lead to the other two causes the camera to fire.  In practice, just shorting all three at the same time causes the camera to focus then fire.    Why connect three wires from the camera shutter release switch to the pin header connector, instead of just shorting the preflash and fire wires inside the camera, then connecting two wires?  I tried this and learned the hard way that the camera focuses immediately then fires when you press the button half-way down.  To allow the camera to function normally when not connected to the relay, all three are required.  I placed pin header connectors between the relay and camera to allow the camera to be quickly disconnected.  I make the leads out of wire wrap wire so that they can be tucked inside the cam when not in use.  This lets me use the camera as it was intended – if I get it back… 

How the Tracking Works

 The balloon (or “Envelope”, as they like to be called) is a 1200 gram Latex weather sounding balloon from Kaymont.  It’ll rise at about 1000ft/minute.  As the outside air pressure decreases, the balloon gets larger until it bursts at an altitude of 33.2 Kilometers = 20.63 miles = 108,923 ft.  Technically this isn’t space, it’s the “Stratosphere”.  Space itself starts at about 100km in the “Thermosphere”.  The camera should be able to see the curvature of the Earth and bright blue glow of the atmosphere. 
The balloon is connected to an FAA-required radar reflector.  The radar reflector is connected to an inline parachute or “Recovery System” purchased from Rocketman Enterprises.  It’ll slow the descent to about 15MPH by the time the capsule reenters the atmosphere. 
The parachute is connected to a payload, or “Capsule” via a custom-made harness crafted by Jennifer Song.  The capsule is a styrofoam hot/cold shipping container about 11x9x7” in size made by ULine.  It looks like a small Styrofoam cooler, but the walls are about 1.5” thick.  This is to insulate the tracking electronics inside from the approximately -70C outside temperatures.
The Styrofoam cooler contains a Garmin GPS18 GPS puck that has been tested and is known to work at altitudes above 60,000ft.  It provides an update once per second via an RS232 serial port to a modem called an “OpenTracker+”.  The OpenTracker+ decodes the information in the serial data stream about time, position, altitude, course, speed and more from the GPS and creates an audio signal encoded in “AX25” packet format.  AX25 is a 1200 baud encoding scheme based on Bell 202, the same thing as CallerID uses to transmit data about the caller to a CallerID receiver/phone.  From there the audio stream is transmitted to the world via a Yaesu VX-2R amateur HAM radio at a power of about 3W.  The whole thing is powered by a 11V 7A Lithium Ion battery.  Also on board is a Canon SD40 4MPix camera capturing a side-looking image once per minute.  The camera is fired every time the OpenTracker+ transmits a data packet, so the images will be tied to a specific place and time.  The images are too large to transmit through the 1200 baud datalink, so they’re stored on a flash card that will be retrieved when the mission ends.  The whole payload capsule weighs a little under 5lbs.
On the ground, we have many ways to track the flight.  The transmitted packets are a part of a position tracking system set up by amateur radio folks.  It’s called “APRS”, or Amateur Position Reporting System.  The packets are received by a radio repeater tower and retransmitted on the same frequency (to increase the range of reception).  At the same time, it’s sent to the internet via a device called an “iGate”.  It’s possible to see this data through sites like the ones listed below.  In addition to the Internet, we have the capability of receiving the data packets directly using another amateur radio set to the same frequency.  They are decoded by an OpenTracker 2 hardware radio modem and sent to the laptop serial port.  Ryan Lantzer wrote some custom scripts that query the internet periodically, listen to the serial port for direct-received packet data, monitor a GPS connected to the laptop for the current receiver position.  The scripts will plot all data in real time in GoogleEarth and create driving directions from the tracking vehicle to the balloon.
In the air, we’ll have Nathan N’s aircraft monitoring the transmissions and relaying that data to the ground.  The aircraft will be in a much better position to receive the packet data after the capsule is back on the ground.  Nathan N has written some software to decode the received audio stream packet data using the soundcard in his laptop, and to speak navigation directions from his location to the balloon.  As of last night there were technical difficulties with the audio card in his laptop and we don’t really know what the status of the aircraft will be.

Here’s a pic of the capsule interior before it went up.
Here’s a pic of the composite “Double Bazooka” antenna.  It’s a piece of coax between two wooden dowel rods, and covered in heat shrink tubing.

We’ll have amateur radio operators (Mark and Nathan Bookout) on the ground and in the air to assist with communications. 

The camera hack.  The wire and jumper header tucks up into the battery compartment, so it still functions as a normal camera.

Cross-band Radio Repeater Experiment for High-Altitude Balloon Flights


The GPF Balloon Brigade will use an airplane equipped with a trasponding radio in conjunction with ground-based  radios as a means for repeating the ground-based radio signals over greater distances.  The goal is to greatly expand the operational range  of ground-based radios  to aid us in the search for our high-altitude balloons after they fall back to earth.  A cross-band repeater will be placed in the airplane and will send any signals it receives back down to the ground-based radio operators.  The airborne input frequency will be 146.450 MHz, FM; the output will be 446.450 MHz, FM, 25w into a vertical J-pole <or whatever it is>.  By cruising at an altitude of approximately 11,000 feet, the airplane will essentially serve as an extremely tall radio antenna, boosting the signals it receives and sending those signals out to a distance of several hundred miles.

Date:  Saturday, March 8, 2008 (weather permitting)
Time: 10:30 a.m. – 5 p.m. (or thereabouts)
Location:  Rolla, MO and environs


  • Determine the approximate size of the “radio footprint” established by equipping an airplane with a cross-band radio repeater.
  • Determine if the orientation of the plane used in this experiment impacts the size and orientation of the footprint in any meaningful way.


  • Nathan Neulinger will fly his airplane (tail number N1365U) at an altitude of approximately 11,500 feet.
  • His plane will carry the cross-band radio repeater. Nathan Bookout (callsign KC0VNP) will serve as the on-board radio repeater operator.
  • Ground-based radio operators will periodically send out signals during the ascent and descent of the flight.
  • Signals will also be sent out during the cruising phase of the flight at altitude.
  • Ground-based transmitters will emit their signals at a frequency of 446.450 MHz.
  • Ground-based receivers will pick up the repeater signal at 146.450 MHz
  • A “net controller” (Joe Counsil, K0OG) will coordinate the radio traffic between ground-based operators during this experiment. 
  • Participants in the radio transmission/reception portion of the experiment are asked to keep radio chatter to a minimum – what we are looking for is the maximum distance between two radio operators when accessing a cross-band repeater stationed at 11,500 feet.
  • We will attempt to record the latitude, longitude, and altitude for each radio contact established between participants.

Flight Track:

One possible flight track suggested by Nathan Neulinger (since its his plane and he’s flying it, I guess we’ll let him be the one to chart a course) is as follows:

Start out in Rolla (K07) and head to Cuba (UBX) for gas.

Fly from Cuba to Sullivan (UUV).
— We begin our tracking experiment in Sullivan.

Fly from Sullivan to Grand Glaize Osage (K15)
— DISTANCE: 72 nm
— we will reach our lower target altitude of 10,500 feet during this leg

Fly from Grand Glaize Osage to Ava (AOV)
— DISTANCE: 68 nm

Fly from Ava to Bollinger-Crass (MO5)
— DISTANCE: 80 nm — we will reach our upper target altitude of 11,500 feet during this leg

Fly from Bollinger-Crass back to Sullivan
— DISTANCE: 73 nm

(click image for a larger view) Map from SkyVector.com

TOTAL DISTANCE: 329 nm (or 378 statute miles)

We will be transmitting APRS information during the flight.

How to Participate:

Contact Joe Counsil via radio (K0OG) or email (k0og@arrl.net).  If you are interested in participating as a radio operator, please be sure to give us your call sign when you contact us.  We will set up a list of participants.

Please give us your transmitter power and antenna type and height above ground.

You will transmit on 146.450 MHz, FM, and will receive on 446.450 MHz, FM.  If you have a radio station capable of working satellites, it will work well for this experiment.  You can also use a dual-band mobile or HT radio, and quickly switch between bands between your transmit and receive.  Another possibility is to use two separate radios, one for transmit on 2m, and the other for receive on 70cm, possibly connected to two separate antennas, or connected to a dual-band antenna through a diplexer.  Do not attempt to connect two radios to the same antenna without a diplexer installed since the transmit power from one radio will feed into the receiver input of the second radio thus causing damage.

What to Expect:

  1. Joe the net controller will contact two participants in turn.
  2. One participant will respond and will be given clearance by the net controller to attempt to contact the other participant via the airborne cross-band repeater.
  3. The two participants will establish contact (or not) with each other.
  4. Joe will then repeat this process with another pair of participants.
  5. Each contact session should only take a couple of minutes (at most) to complete.
  6. We will continue contacting participants until the airplane has landed.

Potential Footprint:

Based on our calculations, we estimate a “radio footprint” of approximately 300 miles or so in diameter.  As the first balloon attempt only traveled approximately 75 miles from Rolla (its origin), our radio footprint should be more than adequate to track the next balloon attempt, scheduled for March 29, 2008. This will also be more than enough coverage to enable the balloon chasers to communicate with each other via radio during the flight. Chasing a high-altitude balloon in flight is similar to chasing a tornado, except that you don’t have a nice black funnel cloud to chase. However, considering the fact the next balloon flight is on March 29, which is during tornado season here in Missouri, we may end up running away from tornados even as we are chasing our balloon. In any case, it should be interesting.

Initial Test: Sunday, March 2, 2008

On Sunday, March 2, 2008, the GPF Balloon Brigade met at Rolla Downtown Airport (K07) in order to test the cross-band repeater. We wanted to make sure the repeater worked at all before we unleashed it upon the world at large.

Nathan Neulinger offered the services of his airplane. Bryan Thompson, Ryan Lantzer, and Eric Sigler performed their techno-geekery to make sure they could receive data from the payload, which was placed upon Nathan’s airplane.

Mark Bookout (KC0VNQ) and Nathan Bookout (KC0VNP) served as the radio operators for the initial test.

Malcolm Hays took a few photos (see below) and provided some much-needed moral support. Jennifer Song also provided much-needed moral support.

Mr. Hays also found out what it is like to go up in a small aircraft with very strong crosswinds.

It looked (and felt) a lot like this:

OVERALL RESULTS: The initial experiment worked. We got some good data from the payload. The unit switched from its low altitude profile (below 2500 feet) to its higher altitude profile (above 2500 feet) as intended.

%d bloggers like this: