Tuesday, 22 September 2015

The TT7-40 Recovery

An update on the TT7-40 situation. In the course of the past two months I kept visiting the landing spot every week to search the tree tops that, as I suspected, held the payload. In total, I spent about 7 hours there walking around with a pair of binoculars.

I even rode to the opposite hill tops to try and search the tree tops from above. However, without any success.

A turnaround came with September. With autumn approaching, the winds strengthened and rains became more frequent.
On my last visit on Thursday, September 17, almost two months after the launch, my luck had changed. After initially searching the tree tops again, I moved my attention to the ground and to my amazement I saw a little white box lying underneath one of the trees.
Upon closer examination, the string holding the payload and the balloon remnants joined together must have eventually loosened enough to rip apart. I assume the strong winds and rain showers helped that a lot.
A close-up on the payload box revealed that the electronic devices had not spent their time in wilderness exactly waterproofed.
My regular payload recovery equipment. There and back, it was a 34km bike ride with quite a climb in the middle.
But at least I could have enjoyed a couple of nice views. I wonder where that military radar was when I was chasing my payload. :-)
Back at home, I began examining the payload. The antenna made only of thin wires had suffered a lot. Whether the ground plane wires broke off on the initial impact or broke off later while dangling among the branches I'll never find out. However, the signal remained quite strong back when I was still trying to locate it and the battery hadn't yet lost all it's power.

Everything looked at it's original place.
The most important part at this point.
None of the electronic parts showed any signs of water related damage despite spending two months exposed to rain.
The transmitter without it's in the end useless insulation.
I then put the SD card, the camera and the LiPol battery inside a bowl of rice to suck out any hidden moisture for a day or so. I also checked the voltages on the batteries. The tracker's Energizer measured 0.53V. The camera's LiPol showed 3.74V which suggested the camera had spent some time filming. At that time, I still had no idea whether there was anything on the SD card or not.

The next day, I tested every device. The tracker got locked and transmitted normally, the camera powered up ok and the LiPol battery charged as well. All that was left, was the SD card that to my delight held about 33GB of a 5 hour video recording. That meant the camera had managed to record as long as I had expected based on the freezer tests.

Here is a short edited video from the flight. Despite the 5 hour video time, the camera died about 20-30 minutes before the balloon burst. It's a shame really, because I would like to see where exactly the balloon landed. The footage also revealed that my estimate of the altered trajectory from the previous article was more or less right.
Examining the footage and comparing it to satellite images from that day, I was able to point out a few interesting cloud formations and spots on the ground.

Here is an extended video footage from the flight. The beginning is spinning a lot, but it gets better as the balloon ascends.

So to finally summarize the TT7-40 HAB flight. I guess the main lesson for the next time is to not rely too much on (possibly wrong) modeling, to not try and push things into extreme and to rather go with experience than theoretical numbers and equations. I mustn't forget to use better and more reliable transmitter next time as well. I didn't get to know how high the balloon went which had been my goal. However, I got all the devices back in the end and I have some nice footage from what probably is +-40km. All in all, I guess all's well that ends well.


  1. Hi,

    very impressive! Finding the payload without a clear clue where it might be is a real achievement!

    I saw that you used one of those chinese keycams - these are notoriously known to disturb the GPS chips. We also lost one flight in 2013 due to this (however it was found by a lady in the Czech Republic close to the border to Austria). Furthermore, the RFM22b are also known to hang after a certain time, so many people just reset it after some time (e.g. after every 20th transmission).

    Best regards,

    Uli from team PYSY (www.stratosphaere.net)

  2. Hi Uli,
    I would say that the winds for that day had a lot to do with me being able to find it in the end. I could see the balloon with a pair of binoculars in the late stages of flight. I also had a rough estimate of how long after I lost sight of it the balloon burst. But I was very fortunate as well.

    With respect to the Mobius camera, it did fine lying right next to Ublox MAX-7 module. I haven't encountered problems there. Concerning the RFM22B, I have had trouble with it before, but I thought I had solved the problem. As described in older articles. The MCU was programmed to reset the radio every 15 telemetry strings as well. However, the issue I had could not be resolved by soft reset.

    Good luck with your future flights,