Almost There 1

One day I read about Weather Balloons in Nuts & Volts Magazine (, they had started a new column called "Near Space"; when I saw pictures of "Near Space" where you could see the curvature of the Earth, black sky above, and the atmosphere below I was immediately hooked. I will (in all likelihood) never get to Space myself, or even get to launch a satellite; but now I know that even on a lowly Software Engineer's salary I can build something myself that can get "near" Space (almost there.) This is something I just HAVE to do before I die, it would give me that much of a feeling of satisfaction (well, almost that much.)

It MUST be Safe & Legal.
It goes without saying that it must be safe and it must be legal (I live my life by that), but I'll say it anyway "It must be safe, and it must be legal."

The 3 Big Options.
Most groups doing this are in the Western states. Here on the East Coast, we have a real disadvantage, all our weather predominately comes from the West, meaning that anything launched will get blown out to sea. I see three basic ways to deal with that...

  1. Reusable using GPS: Drive about six hours inland (so it can land before reaching the ocean,) launch it, drive for hours to find it (using GPS), drive for hours to get home.

  2. Reusable using GPS: Make the whole thing float, and everything (that you care to use again) water proof; launch it, then use a boat to retrieve it (using GPS.)

  3. Disposable: Make it inexpensive enough that you don't mind using it once and throw it away (let it dump in to the ocean, never to be seen again.)

    (Actually, there is a fourth option of building a GPS guided glider that fly's it self back home, but that seems a little out of reach, and way too much can go wrong dumping a lot of time, effort, and money in to the ocean.)

I decided to use option 3, at least for my first flight.

Helium vs. Hydrogen
For a while, I consider generating my own hydrogen. After some research I found two ways to go...

  1. Electrolysis: I could build an Electrolyzer. It's more efficient to use Sodium Hydroxide (Red Devil Lye) as an electrolyte, but that is dangerous. You can use common Baking Soda, it's less efficient, but safer. This works, but it would take several weeks to generate enough Hydrogen.

  2. Chemical: You can mix Sodium Hydroxide (Red Devil Lye), water, and common scrap aluminum and it will generate Hydrogen rather quickly, but also rather violently. This method may work well for filling up a small 6 inch party balloon, but I had safety concerns ramping this up to produce gallons and gallons of Hydrogen.

Ultimately, it was the storage problem that made me decide not to use Hydrogen. I guestimated that I'd need as much as 60 gallons of Hydrogen, and that producing and storing that much Hydrogen at home is just too dangerous. As I read some place on the web "Anyone who produces and stores any quantity of Hydrogen at home will deserve the Darwin Award they are destined to earn." (that may not be an exact quote.)

Nope: I'm using Helium. Period. The decision is final.

I could use AX.25, which is a standard protocol for sending data over Ham transmitters, but I'm using a PicAxe as a CPU, and it's way too complicated for a PicAxe to handle (if you can prove me wrong, please do.)

I could use DTMF, but again, the PicAxe is too weak to handle that too (and again; if you can prove me wrong, please do.)

I finally decided to go back to basics and use plain old fashioned Morse Code. Yup, even a PicAxe can handle that.

The other part of communications is the transmitter; I could get a Ham license and a Jersey FireBall 40 Transmitter, or just use a License free band like CB, or FRS Radio. I decided on FRS because they are smaller and lighter than most common CB radios.

System Block Diagram
So, here's what I decided to do. Use a Pressure Sensor and a Temperature Sensor fed in to a PicAxe CPU that approximately once a minute will trigger the transmitter and bang out the raw data using simple Morse Code.

On the receiving end, I'll use a matching FRS radio to receive the data, feed it in to the Sound Card of a Laptop, and a Cassette Recorder with sound level activation to record the data for time immemorial. Run software on the Laptop that decodes the Morse Code and displays it on the screen, and saves it to a Data file. Then (finally) I'll have a conversion chart that I can look up the raw data values from the screen and manually convert them to Altitude and Temperature readings.

System Detail Block Diagram
System Block Diagram

Component Description

  1. Battery: UltraLife U9VL-BP. I was going to use a 6 volt Lithium battery and a Very Low-Dropout 5 Volt regulator (about 0.75V), but after looking up some datasheets on Lithium batteries I saw that they drop about 20% of their voltage at very low temperatures; that plus the Regulator Voltage Drop would put me below the 4.75V minimum necessary for some of the components. I finally decided to use a 9V Lithium battery because I need the extra voltage at very low temperatures. Fortunately, at exactly the same time I started looking for a 9V Lithium Battery, Radio Shack put batteries on sale! (Buy One, get One Free.) It should run the system an estimated 10 hours (even at low temperatures) which is far more than I need.

  2. Regulator: REG1117FA-5.0. I'm using a Low-Dropout Regulator that I got on E-Bay so it can run as long as possible, even if the battery voltage drops more than expected at low temperatures.

  3. Pressure Sensor: MPXA4115A. The Pressure Sensor I'm using is only good to about 45,000 feet or so. Not the best choice, but this is my first launch, I've used an MPXA4115A before, and it was easily available on E-Bay.

  4. Decided NOT to use an LM34Temperature Sensor: Dallas DS18B20X. I was going to use an LM34; it can go below 0C if you pull the output pin negative. I was going to use two diodes to raise the ground lead, and two ADC ports to measure the voltage difference, but I would have run out of I/O pins on the PicAxe 08M. I finally decided on a Dallas DS18B20X, because it interfaces easily with PicAxes, and I could get it cheaply on E-Bay.

  5. CPU: PicAxe 08M. I decided to start simple, and the PicAxe 08M seemed like the simplest, cheapest CPU that would do the job (and I could get it on E-Bay.)

  6. Transmitter: Motorola T4900. I picked up a few of these at Radio Shack on sale. They were cheap, light, and operated on a license free band. I spent a little time developing the interface from the CPU to the Transceiver, I finally found that I can pull a connection low through a resistor and the Data rides on that (by the way, this radio does NOT use a standard 1/16 stereo plug; it appears to use a custom one with several connections.)

  7. Antenna: 1/4 Wave Length with Drooping Radials. I destroyed the "Rubber-Ducky" antenna that came with the radio cutting off the rubber because it was too heavy, so I had to make another one. After some research I found a 1/4 Wave Length with Drooping Radial Antenna on the web, and built one like it from calculating the wave length and looking at the picture.

  8. Receiver: Motorola T4900. Matches the Transmitter bought at Radio Shack above.

  9. Laptop: Panasonic Tough Book. I had bought this on E-Bay for another project (, so it was handy to use.

  10. Tape Recorder: I'm using the Magnetic Mass Storage and Retrieval Unit (I mean Cassette Recorder) that came with my Radio Shack TRS-80 about 30 years ago. It has sound activation which will save hours of tape since data is only transmitted once a minute or so (it cuts off the first 1/4 second of data, but I don't need the first word anyway, it's always going to be a zero.)

  11. Decoding Software: GetCW. I poked around some Amateur Radio Web Sites and News Groups until I found this. It seems to work very well for my purposes.

  12. Conversion Chart: Almost_There_1_XL8_Chart.html (short for Translate.) This is a JavaScript driven HTML document I made up that displays an abbreviated conversion chart of the values I expect to get back (between saving the data to Tape and a Data file, I can always go back and make any other conversions I need in the future.)

The software is very simple.

  1. Increment a Counter
  2. Read raw Temperature Data
  3. Read raw Altitude data
  4. Every ten minutes Identify Self
  5. Send raw data
  6. wait about a minute
  7. loop back to 1

Getting Started
Once I had all the components, I was going to "just build it" assigning the I/O pins while building; but, then I realized that I'd run out of I/O pins (that would have been a major oops!) The PicAxe 08M has very limited I/O, and some of the pins can only be input, and/or output, and only certain ones can be used to read analog voltages. If I had just assignd pins ad-hoc, it couldn't possibly have worked. I finally had to combine the Data Out with the Debug line going back to the PC (I could no longer debug the program without interfering with the Data output.)

After that hiccup, I decided that I'd have to do a "Full System Test" with everything breadboarded exactly as I was going to use it before the final build.

2005/09/28 - Here is a picture of the "Full System Test"; only differences are, breadboard (obliviously), 120V power supply instead of batteries, and there is no tape recorder yet. Otherwise, it's a full system test from sensors, data processing, Morse Code output, send/receive RF communications, Laptop interface, decoding, and displaying.

First Full System Test 2005/09/28
(How did Man Kind ever work with Electronics before the invention of E-Z Clips and Alligator Clips???)

2005/10/08 - Here is the completed motherboard (next to a postage stamp for scale.)

Front of Motherboard 2005/10/08

Back of Motherboard 2005/10/08
(Sorry the Stamp is washed out, but what do you want - the Flash was only about four inches away!)

2005/10/22 - Here the Motherboard is interfaced with the Transmitter, the whole thing is only about two inches by four inches. That's the Temperature Sensor (Dallas DS18B20X) on the upper right; it will hang outside the Thermal Enclosure.

Motherboard Interfaced with the Transmitter 2005/10/22

2005/10/22 - The entire Electronic Flight System (not including the Thermal Enclosure) only weighs about 2.55 ounces! (According to this dime store scale; I don't know how long I've had it, but the internal mechanism is rusty and I'd have to take several reading and questimate the average; sometimes it would stick a little high, sometimes stick a little low.) The antenna alone was so light I couldn't accurately measure it, the (cheap) scale barely even moved (maybe about 3 grams.) I also gave this thing the best "Cold Test" I could conveniently do; I left it unprotected overnight in the freezer set on maximum. It was at -13C for 24 hours and still worked fine under those conditions.

The entire Electronic Flight System (not including the Thermal Enclosure) only weighs about 2.55 ounces! (about 72 grams)

2005/10/26 - Thermal Enclosure: That may look like a miniature Gemini Spacecraft made out of exotic and expensive space-age materials, but it's really just a discarded Styrofoam coffee cup. This thing will experience about -60F on its way up, and it is painted black to absorb solar radiation helping keep the contents warm.

The Thermal Enclosure is painted black to absorb solar radiation.

2005/11/02 - I went out and bought a better scale (on E-Bay of course) because I just didn't trust that cheap, old, dime store scale. As of right now, the entire Payload System (including the Thermal Enclosure) only weighs 93.3 grams! (3.29 ounces). That may change a little by flight time as I may be able to lighten it up some (but then again, I also may add some insulation.) The antenna alone is 7.2 grams.

The entire Payload System (including the Thermal Enclosure) only weighs 93.3 grams! (3.29 ounces) (I trust this scale.)

2005/11/06 - I worked on making it even lighter, about the only significant thing I could do was to remove the bulky, heavy shaft for the volume control on the radio. That plus a few other petty things got the weight down to 90.8 grams. Here it is, ready to fly!

Final Assembly (Rube Goldberg would be proud): The Payload String (A), get connected to the Balloon (B), then runs through the top of Parachute (C) and becomes one of the shroud lines at point (D) and connects with the rest at point (E), it then passes through the top of the Thermal Enclosure (F), gets tied to the Motherboard and Radio (G), and passes through the removable access hatch (H); finally, the transmitter output connector (I) goes through the hole in the access cover at point (J), the cover is put on and the antenna (K) is pressed on to the transmitter connector.

That's it; there is no more building or modifications to do. I have the balloon on order and the last step is to set a date and just "do it!" Here is a simulated launch using EOSS's Balloon Track program (which I can't thank them enough for putting this in the public domain; thank you, thank you, thank you!) and Forecasted Sounding data from NOAA. It will leave Virginia Beach, VA and go to an estimated height of 76,000 feet where the balloon will rupture, then float down and land about 144 miles out to sea; landing well past the Continental Shelf and in to deep water (ain't nobody ever gonna' find it!)

A simulated Launch using Forcasted Sounding Data from NOAA

2006/02/20 - I wasted a lot of time because I was concerned about the legality of using FRS Radio in this manor. I had checked on-line Ham Forums and News Groups and still wasn't sure; I even called the FCC directly to get clarification, but who ever I talked to couldn't/wouldn't answer my questions. Finally, I visited a local Electroncs Flea Market and talked with some Hams up there; they assured me that there was no problem (guess I was just being paranoid.)

Weight Breakdown

Sub-System Current Weight Grams Weight Ounces
Motherboard 10mA (max) 9.2 g 0.32 oz
Transmitter (with wiring) 26mA (receive)
280mA (transmit)
28.7 g 1.12 oz
Antenna 7.2 g 0.25 oz
Battery 36.9 g 1.30 oz
Thermal Enclosure 8.8 g 0.31 oz
Extra Thermal Insulation 1.9 g 0.07 oz
Payload Total 36mA (receive)
290mA (transmit)
(about 120mAH averaged out)
92.7 g 3.27 oz
Balloon 300 g 10.6 oz
Flight System Total
Grand Total

2006/03/25 - The night before. Well, weather permitting, I'm going to launch tomorrow! I decided to add some extra insulation because it is winter, and from what I read the tropopause can be around -90F or so in the winter (I'm using bubble wrap as insulation, and I expect it will pop at some altitude, but it was a last minute thought and the best I could do on short notice.) So the final weight is 92.7 grams (3.27 ounces.) A friend from work and someone from our local Ham Club are going to meet me and help me launch it.

The final weight is 92.7 grams

2006/03/26 - Launch Day!
I met with two guys (one from the local Ham Club, and one from work) to help me launch the balloon. Everything went pretty smoothly considering that no one there had any experience with this stuff. The laptop gave me fits with the sound card driver, I had to uninstall/reinstall it several times (with a reboot each time;) it seemed like forever while I had people waiting on me (and I had the deep fear that I was going to have to abort and tell these guys they had wasted their day.) The other thing that gave me problems was sealing the balloon; I tried several times to fold the neck and tie a string around it. I finally gave up on that and just wrapped Duct Tape around it as tight as I could (good enough.)

One final check that we were receiving data, and I released the balloon. We brought binoculars and a telescope, and followed it until it went through the clouds and could no longer be seen. Every so often I checked the data and called out the Altitude and Temperature. It was pretty thrilling to see my little contraption actually work and return data. Using the Morse Decoding Software we got good data up to around 42,000 feet, where it was about -41C; after that the decoded data was too garbled to make any sense of. I have the audio saved in a WAV file, and one of the guys said he was good at weak signal CW copy and would take a crack at decoding it.

When I thanked my friends for helping me, they both were like "Thank You for inviting me! That's the most fun I've had in a while!"

Here's the Raw Data as returned from the little craft and decoded by the software, and here it is plotted on a graph after a little cleaning up. The spikes are missing data points.

Data Plotted on a Graph (the spikes are missing data points.)

Here is the Preparations Checklist that I used before leaving home on Launch Day, and after arrival at the Launch Site.

Here is the Launch Checklist; step by step instructions that I used to Launch the Almost There 1 Weather Balloon.

If you have any comments, suggestions, questions, or anything you want to say or ask, please E-Mail .

---=[ It may not be Space, but it's Almost There! ]=---

Note: While working on this project, I was selected (by invitation only) to test a prototype of a new, very powerful little Micro-Controller that I will almost certainly use on all future projects. It was all very hush, hush until the official public release (I had to sign Confidentiality Forms and all.) It is hardware and software compatible with MetMedia's Basic-X processors. While it has a Basic-X compatibility mode, it also has a native mode with a much more powerful set of commands than Basic-X. It's faster, and has more memory than NetMedia's products. If you're at all considering getting a Basic-X, you owe it to yourself to check out http://www.ZBasic.Net/ first! I have used it and can highly recommend it.

Good Luck!