Balloon Mission 1 - Overview

Mission 1 Results: FAILED

Read the report here.

Description

My goals for my first balloon mission were fairly aggressive:

1. Launch, fly, land, and recover balloon and payload safely.

2. Achieve a peak altitude of 80,000 feet or more.

3. Capture high-resolution images from altitude, both vertically (down) and horizontally.

4. Downlink low-resolution thumbnails of high-res imagery using SSTV.

5. Track flight of balloon and payload using APRS.

6. Recover all payload materials quickly and efficiently.

Dates and Frequencies

The launch date was scheduled for February 5, 2006. I used NOAA's measurements of winds aloft, combined with EOSS's Balloon Track software to generate landing predictions. The day of the flight, I expected to be able to use the most up-to-date data to predict where the balloon would go, and where the flight string would be touching down.

I planned to use 145.960 MHz as the primary simplex frequency for the flight computer. I was planning to send both APRS and SSTV traffic out across the 2.5W primary radio, so it would have been inappropriate to use the national APRS frequency (144.390). This meant that I would not have the ability to hit APRS digipeaters along the flight path. I had asked a number of stations in south-central Virginia to listen for APRS traffic on this frequency and I-gate it to the Internet if possible during the flight. I encouraged others to track the balloon's flight via FindU.

Just in case the flight computer, main batteries, or primary radio fail during the flight, I also planned to send up a Byonics Pocket Tracker fitted with the Pocket Fox firmware. While this unit would not be sending GPS data during the flight, it would send out the balloon's call sign once every minute. This would then allow the recovery team to use radio direction-finding techniques to track and locate the payload. As with most Pocket Trackers, the only two transmit frequencies available are 144.390 and 144.340 MHz... I chose the latter for the long tone-and-CW sequences.

Flight Equipment

The following materials were included in the payload (also see pictures at bottom):

In the photos below, I laid out the components in my driveway, and began tying ropes and laying out cable. Barely visible beyond the orange-and-blue descent parachute is the (30 gauge) wiring for the hot wire cutdown device. The brown cable to the left of the coolers is the J-Pole antenna.

The primary digital camera peeking out of the front of the blue cooler was to snap and occasionally SSTV photos of the horizon - the primary goal of the mission.

The "eyeball" camera just below the primary payload is intended to look downward during the flight. During the descent, the flight computer would send SSTV images from this camera to provide the recovery crew with a view of the landing zone, in case radio contact was lost after landing. The piezo beeper is also intended to aid in the ground search after landing.

Close-ups of the flight computer are shown below.

I cut and mounted thin pieces of cardboard under each circuit board to prevent accidental grounding (in case of a rough landing, etc.) - this can be seen at the bottom of the power supply board. The board mounted to the lid was an IDE-to-CompactFlash interface - the computer booted from CF, saving about two pounds of additional weight, an amp of current draw, and loads of excess heat from the flight computer (booting from flash was a novelty in 2006).

On the left above is the fully assembled flight computer, ready for external wiring. It may not look like much, but this was a fully functional Linux PC. In the end, this computer ran great, but even with its very low power consumption (by 2006 standards), the weight of the batteries to power it for a few hours doomed this design to failure.

The "eyeball" cam in the right-hand image has lousy image quality, but I had it lying around the house, so it was free. The digital camera was $25 on clearance. It had a voracious appetite for double-A batteries, so was modified to be powered externally. The radio was a terrible Chinese domestic-market unit that I picked up for a song - $42 with three-day shipping from Hong Kong (this was just before Wouxon and Baofeng swept into the US market). The radio feels cheap, and the receive audio is unpleasant, but it does have 2.5W of transmit power (more than the Yaesu VX-1 I had considered using), a speaker-mic jack (for easy interfacing to a PC or tracker), and is cheap enough to be nearly disposable. Shown here with no battery pack, this too was modified for external power.

To save the weight of a TNC on this mission, I am using Thomas Sailer's soundmodem software to generate AX.25 packet audio with the flight computer's sound card. I built this fly-weight interface to connect the computer to the radio. It uses the request-to-send (RTS) pin of the PC's serial port to close a tiny relay. The whole assembly fit neatly in a discarded blister pack, which prevented shorts and bent pins, weighed almost nothing, and was free. The entire interface (except the wire and plug for the radio) was intended to ride inside the "cookie tin" computer case, so a more robust housing was not required.

The radio and PC were powered from the same bulk battery pack, so ground loops were not a concern. I should have added bypass caps to avoid RF on the audio lines, though.