Tuesday, June 6, 2017

High Altitude Balloon flight - 6/3/17 from Rochester, NY Hamfest

High Altitude Balloon flight

Having modified a camera and built a controller, I wanted to fly it to see how the camera worked.  I assembled the payload into a foam box and rigged it up with a parachute.

Photos of the launch

Are available in this album

Payload parameters

Parameter Value
Payload mass 650 g
Parachute 30" octagonal (corner to corner)
Estimated Descent Rate 5.91 m/s (using Hexaagonal)

Flight prediction input

Parameter Value
Location Set with map to the Hamfest venue
Launch Altitude Unset (0)
Launch Time 15:00
Launch Date 6/3/2017
Ascent Rate Initially set to 5.5 m/sec.  Predictor overwrote it to 6.3.
Burst Altitude Left to default at 30,000.  Predictor over wrote it
Descent Rate 6.0 m/sec
Burst Calculator
Payload Mass 650g
Balloon Kaymont 1200
Target Ascent Rate 5.5 m/sec
Gas Hydrogen

Flight prediction Output and Actuals

Parameter Predicted Value Actual Value
Burst Altitude 34315 m 33880
Time to burst 93 m 88 m
Ascent Rate 6.12 m/sec 6.388 m/sec
Launch Volume 103.8 cu ft 103.8 cu ft (est)
Descent Rate 5.91 m/sec 10.12 overall
 3.2 m/sec last 2000 feet
Descent Time 133 m 143 m

Flight path comparison

The flight followed the prediction almost exactly!  It landed 10.24 miles further SSE than predicted. The parachute performed very well.  I need to re-measure it to assure it's a 30" (and not 36") chute. That might explain the discrepancy.  NB: I measured it.  Yup, it was a 36" chute.  That explains a lot!

Temperature data

Charts in aprs.fi showed the internal tracker bottoming out about -39 degrees celsius, and the external tracker at -42c.  I didn't work really hard on taping up the box, but it seems that such a large open box with a still camera really doesn't have much thermal mass.  It'll be interesting to see how that changes when I have a video camera in there on the next flight.

Lessons Learned

  • Using a time profile for scheduling the snapshot frequency worked just fine.  It was a good way to get more picture density during interesting periods of the flight.
  • I need to work on sealing the payload box better to see if I can stabilize the temperatures some. Everything functioned, but I would have preferred the cameras stayed warmer.
  • Using the "Landscape" mode on the Kodak Pixpro yielded blurry pictures.  I got some great images, but more than 50% were blurry.  I'm going to try "Sport" mode next time.
  • The camera performed well enough that I have purchased a second one to use as a video mode camera in the payload, and run two at once.
  • I need to remeasure the parachute to see if I can sort out the discrepancy in descent rate.


  1. Just saw this - thanks for doing this at the Hamfest, great job! I was really impressed that you were able to predict the flight so well.