A weblog of my most recent coding disasters... and a lot of other stuff.

High Altitude Ballooning: A practical approach. (Part 2)

Humble Beginnings

After the team was assembled it fell upon us to think of something to do. We were nerds, never content to have done what someone else has done before; we needed to do something cool. Something innovative. We still embraced the giddy-science-nerdy goal of taking pictures from the edge of space, but no way would we stop there. Perhaps if we had set our eyes a little closer to the ground we would have actually accomplished all of our goals.  As it stands, almost 2 years later, our team has finally accomplished what we intended to do in 3 months.  We’ll need some divine intervention to accomplish what we had our sights on for 6 months. Oh lordy.

So maybe a more apt title for this blog would have been “Lofty Goals, Humbling Experience” as our group quickly discovered the limit of productivity for people working outside of their comfort zone and on an incredibly limiting timeline.


The brainstorming process was convoluted, but from the beginning we agreed on a couple things that would help us break the mold.

  1. Stabilization.  We weren’t exactly sure how it was useful besides maybe enabling us to take prettier pictures, but we did know that we were going spin some motors crazy fast and stabilize the sh*t out our package. Or so we hoped.
  2. Sensors sensors sensors. How were we supposed to know if the package was stable without sensoring everything!?
  3. Safe payload recovery. Everybody was doing it. It was necessary to get our pictures back. Duh.

It was with these ideas firmly in mind that we began making small purchases towards our project. Hilariously huge military surplus weather balloons. Check. Styrofoam minnow bucket. Check. Crazy fast motors. Check. The receipts were starting to pile up when Mark and I had a serious epiphany:

Since we were doing something experimental we could apply for a Creative Work and Research Committee grant from Valpo! We could make our side project—our scheme-hatchery—into undergraduate research.  We’d bump up the scientific rigor, write a nice abstract and badabing, badaboom! we’d have a nice resume line item, and part of the cost of the project would be defrayed. Much better than the $100 to $150 dollar price tag each that we had planned on. Mark even had attended a lecture by a Valpo professor about his work with high altitude balloons…

Enter Professor Gary Morris.  Fulbright Scholar. Dean of Physics. Leader of more national projects than you can shake a stick at. He was the man that Mark had heard a presentation from, and when we told him that we wanted to work with high altitude balloons, and that we wanted him to be the advisor for our group, he was ecstatic. He was so into us, it was scary awesome. And it was to our benefit, of course.

He helped submit our proposal to the CWRC and was excited and involved without being controlling.  He let us do our own thing. And for a long while we were damn proud of it.  I suppose I still am.  No one told us to do the things that we did.  We made the decisions we needed to, purchased what we needed to, and pulled long hours to try to make everything work. We were an entirely self-guided group, at the risk of, well, exactly what happened happening.

It’s true. It was ultimately the self-guidedness that we were so proud of that led to our largest snafus. We had many ideas.  And none of them worked. At least not really. I’ll cover the actual build processes in Part 3, but here are a few of our ideas that ultimately failed in one way or another.

First, our device for midair stabilization.  The idea was simple. Two spinning discs, each spinning in a different direction to create the greatest amount of stabilization via gyrostat possible, and lock the payload vertical. Problem is, the design went something like this:


Which, merp, anyone who really understands physics and is paying attention would have divined the implausibility of such a device… two completely opposite inertia in a closed system will cancel. There will be no stabilization. Only heartbreak. We later reworked the design in an equally heartbreaking way, though that time through no foreseeable fault of our own. You’ll hear about it in Part 3.

Second, we spent a significant amount of time documenting the use of a service called instamapper to track balloon packages with cell phones. This proved to be particularly fruitless, because A) if your flight takes any significant amount of time (>3hrs) the phone will die of loneliness/constantly searching for service and B) as it turns out, using your cell phone in an air vehicle is illegal. Oops.  Our bad. It might not seem like a big deal, since the chances of the law enforcement hearing about such a mistake are slim to none… but for the research institutions that are sending up thousands of dollars of equipment, obeying the law is paramount. Out of necessity an alternative was born, but not until much later.

Third, we tried to reinvent the wheel when it came to a sensor package. We purchased a PIC18f4550 microprocessor and 4 sensor chips, none of which had C libraries available.  Only 1/4 of the sensors worked when I was done with them, despite over 72 hours of testing, wiring, rewiring, testing, facedesking and yelling.  I can only imagine that it was my amateurish testing methods that made the debugging difficult by frying the sensitive ICs like twinkies at the state fair. Science man, it’s messy. 

Even with Murphy breathing down our neck, we learned a lot. Mark went to Houston for 5 days and apprenticed with some serious weather balloon launchers, I learned more than I ever could have desired to have learned about microprocessors— things that are benefiting our research more than a year later! And we all learned a little bit about the real side of science, and had a fantastic time building things. It brought us together. Literally. We started 500 miles apart and it reunited us for a portion of the summer. 

Eventually, we managed to actually get a balloon off the ground.

That and more in Part 3 (Build Season Odyssey)