Analyzing Launch Data

While we were not able to recover the payload from our launch last Friday, we did learn a lot about what to expect for future launches and where we can improve.  The good news is that the tracking system worked impeccably, even after experiencing much more extreme conditions than expected.   Below you can see the flight path of the payload.  The balloon traveled over three times the distance we had hoped it would travel due to under-filling the balloon with helium. 

The following graph shows the elevation profile.  Our average ascension rate was about 48 meters per minute.  At the next launch, we hope to achieve an ascension rate in excess of 200 meters per minute.  You can probably notice some strange data in the graph below.  Specifically, elevation decreases and then increases again.  This could be a result of bad GPS data.  If it isn't, it's interesting. Perhaps there was an intense pressure gradient that caused the balloon to actually decrease its altitude. 

Below you can see precisely where the anomaly occurs in 3D view.  Also, a little while after the reported decrease in altitude, there is a half hour where the altitude essentially doesn't change at all.  It could be that we had less than 4 satellites locked during this time leading to latitude and longitude data, but no altitude data. 

More photos and video of the balloon launch will follow.

Status Update

Here you can see the latest data on where the balloon is.

Launch 1

This afternoon we launched our first balloon. It is in the air at the time of this post.  You can track its progress here:

http://aprs.fi/#!mt=roadmap&z=11&call=a%2FKC8DRG&others=1&timerange=3600

Assembly for Flight 1

Here you can see the flight module for test flight #1.

Thermal Test

To ensure my the reliability of my flight hardware, I conducted a test where the conditions at 100,000 feet were simulated.  To do this, I used a cooler, a 12 pound block of dry ice, a mounting system, and a temperature sensor.

Here you can see the big block of dry ice I used to get down to the desired temperature.

Here you can see the flight hardware packed inside the polystyrene foam container.  The Geiger Counter, DAQ system, and camera were all placed inside and turned on.  I placed a couple pieces of Uranium glass inside so the Geiger Counter had something to record during the test.

This is the setup inside of the cooler.  The best way to get really low temperatures around the box was to place the block of dry ice above it.  I used some more pieces of polystyrene foam to mount the dry ice above the box.  You can also see the temperature sensor that I used to derive the temperature inside the cooler.

 This is a picture of the box before placing the dry ice on top.

Now the piece of dry ice is placed and ready to go.

I put the cooler lid back on with the temperature sensor wires connected.  The equilibrium temperature inside of the cooler and outside of the flight hardware box was found to be -36 degrees Celsius, right around what I expect the temperature at 100,000 feet to be.  After two hours, a thermal equilibrium was established with a temperature of -5 degrees Celsius inside the flight module.  The heat generated from the electronics and excellent insulation properties of polystyrene foam aided in establishing this equilibrium.  At the conclusion of the test, all of the electronics were still functioning perfectly.

Overall I was pleased with the performance of the electronics under the conditions.  The next task will be to perform the same experiment with hand-warmers placed inside the flight module.  In doing so I hope to establish a factor of safety by running closer to the middle of the operating temperatures of all the electronics.  Further, temperatures will actually be lower than -40 degrees Celsius on the journey up to 100,000 feet, so having some active heat source other than the electronics themselves will be critical.

Geiger Counter

Here you can see me testing out the Geiger Counter we just got in.  It shipped with some naturally radioactive material that you can test with, namely uranium glass.  It's a little hard to see, but when a beta particle is detected, a little red LED lights up on the probe assembly.  This sensor is capable of detecting both beta and gamma radiation.

Raspberry Pi

Today I received my Raspberry Pi, which is basically a computer that will go up with the real flight hardware. Personal computers are now the size of credit cards- it's pretty amazing!

Progress Update

All of the internals for the first launch of the balloon have been ordered.  When they arrive, I will assemble them and find out the final weight of the flight hardware.  This is important for a number of reasons:

1. If the final weight is less than 4 pounds, some regulations seize to apply- to see more information about the regulations that apply to unmanned balloons see  http://ecfr.gpoaccess.gov/cgi/t/text/text-idx?c=ecfr&rgn=div5&view=text&node=14:2.0.1.3.15&idno=14
2. The weight will be an important factor in determining the size of the balloon and parachute needed to reach the desired altitude and land safely
3. The lighter the flight module, the more potential I have to add ballast that will stabilize the module during flight

There are some complications here- the whole point of doing a test flight is to learn something for the real launch where there will be hardware used to measure gamma radiation.  So I may end up adding weight just to better simulate what the next launch will be like. 

Stay tuned for impact and thermal testing this week.

Machining the Dummy Balloon Housing

In the picture above you can see how the dummy balloon housing is manufactured.  I first cut at piece of high density Styrofoam to a shape that was close to my model.  I then machined it down to specifications by hand.  It's not perfect, but it should function adequately.  This version will be used first to do a thermal test and then an impact test.  Posts detailing the procedures of those tests will follow.

Flight Module for Dummy Balloon

Here is an image from CAD of the flight module coming together- a work in progress.  I plan on toying with the idea of making the casing in the shape of a raindrop.  If I add some ballast (can't afford to add too much), this should make the module always point in the direction the wind is coming from and minimize the extent to which the module gets knocked around.  Just a Styrofoam box for now though- it will be much easier to manufacture for testing purposes.  The rain drop design would probably require some time with a lathe to really get a good shape going.  Stay tuned.

 

Stage 1

The first step is to prove that we can successfully launch and retrieve a balloon that has achieved our desired altitude.  Therefore the first launch will be a "dummy balloon," where only a camera and GPS recovery hardware will be sent up, and perhaps a couple other sensors to get a better idea of the conditions the inside of our module will see.  On the journey to the stratosphere, we anticipate seeing temperatures as low as -70 degrees Fahrenheit and atmospheric pressure about 1% of what you see at sea level (almost a vacuum).  The location transmitting devices then, must be able to survive these conditions in order for us to retrieve the module.

Once we prove that we can successfully launch and retrieve a module, the experimental flight hardware will be launched.  Manufacturing of the dummy balloon has already begun, stay tuned!

So it begins!

My name is Matt Pleatman, and I'm a senior undergraduate student at Duke University.  This blog will detail my journey to send a weather balloon to "near space" that will hopefully capture scientifically valuable information about high energy particle radiation during thunderstorms.  This project is being completed as an independent study.

None of this would be possible without my adviser, Dr. Steven Cummer, the Jeffrey N. Vinik Associate Professor in the Electrical and Computer Engineering Department at the Pratt School of Engineering.

Let's get started!