The day started a little later than anticipated, we arrived at the launch field around 11 am. It was a cold windy day which did not help things out.
We setup fairly quickly and checked out the payload. Powering the unit resulted in a failure in the transmitter. This quickly proved to be the connection between the gps board and the cpu board. This connection supplied gps info to the cpu board for local logging. This being an unnecessary inclusion, I removed this cable (a mistake I caught after launch).
The payload was now up and running and transmitting, we were just waiting for lock. At this point we are filling the balloon. Due to high winds, and maybe some man-handling of the balloon, the first balloon popped. This is why I always buy things in pairs!
We are able to fill the second balloon with helium, but run into power problems. Powering up the rest of the electronics causes one of the cameras to retract and power off. The longer cables I made the night before plus the colder temperature I am guessing. The other camera works fine but I had taken off the safety diode on that a while back during testing. Luckily we brought the soldering iron so I removed and bypassed this diode which fixes this problem.
Balloon filled, transmitter still transmitting we are ready to launch. We release the balloon and it hurls off in a strong horizontal movement, but with enough vertical movement to ensure that it clears all obstacles. Ascent rate seems slower than expected meaning we probably underfilled it. We used a differential amount of wait to determine the proper lift of the balloon, needing lift equilibrium to be properly filled, but the high winds made this difficult.
Soon after launch we realized that the transmitter was repeatedly transmitting the same coordinates over and over. This was an anomoly I had not encountered before but apparently showed up the night before. Scrambling to keep track of it visually, we jumped in the cars and hit the road. It would pop in and out of our visual grasp, but it would be a while before we completely lost track of it visually.
We were still receiving transmissions, albeit with improper coordinates. To try to take advantage of this, we decided we could try a directional antenna. The chase vehicle with the gps tracking continued down the interstate in the expected path while the other went back to grab some parts. We grabbed what we could and headed back out, at this point visual tracking a long lost memory, when the transmitter started transmitting its correct coordinates (presumably). It was right above the gps chase vehicle.
By this time I had realized my failure when I removed the cable between the gps and cpu boards. My recent code on the cpu board applied no polling with serial IO, so with no serial IO coming it, the CPU board was sitting dead. The CPU board also controlled the cutdown device, which was programmed to drop the payload after 2 hours. More than 2 hours passing and no descent confirmed that the cutdown device was not going to trigger. It also meant that we underfilled the balloon as well since this should have burst 30 minutes prior to the cutdown device triggering.
We continued to track it for about 30 minutes past the prescribed cutdown time when the transmissions went dead. We drove and listened but no more transmissions came in. This being our only tracking method meant that this payload was lost. I also neglected to add contact information to the payload. This ended our adventure, the chase vehicles met up and headed home.
While the launch was a failure, we learned a lot from the failures we had so that we can plan for them in REHAB2. I am thinking the cameras experienced enough of a power dip to cause them to retract and turn off. This may have reset the PIC or GPS which caused it to start transmitting correctly. The cameras being retracted let the cold in freely which probably froze the batteries after some time.