Flight 1 :: SUCCESS!!

Wow, long time without an update and so much has happened. Quick synopsis goes like this:

• Previous flight 1 launch attempt failed due to accidental gpd groundplane to battery hot line contact (just a recap of 2 weeks ago...)
• planned second attempt for the following weekend, picked up additional small Helium tanke and attempted to connect other GPSes that I have lying around. For some reason, this was unsuccessful and attempt number 2 slipped.
• broke down and ordered a direct replacement of the original gps unit from Argent Data
• gps arrived last wednesday. Friday forecast and trajectory prediction was looking promising.
• Connected gps on Thursday and performed new end-to-end functionality test. All was well. Friday weather still looking good. Went to sleep early.
• Friday, March 2, up at 0400 and starting drive to launch point in Frostburg, MD

We got to Frostburg not long after sunrise and started balloon inflation. We took the time over the last two weeks to make a few design changes to the C&DH electronics board to add jumpers on the main power lines. This allowed us to do all the payload setup and make all electrical connections and then disconnect power until the balloon was ready. Launch then only required that we put two jumpers in place and start the camera script. We had enough experience with the process now that we were able to get everything pre-flighted while the balloon was filling and the whole process was much more relaxed.

Like last time, we gathered a bit of a crowd in the park including a little visit by the local police. No arrests were made. :-)

We were a little more aggressive with the balloon fill this time around and, before long, we had a balloon more or less ready to go. The payload started up and was communicating with the ground station exactly as it should so we taped the payload shut and cut the mooring lines.

  Then, ProjectDEBO soared free on it's maiden voyage.

Our plan was to sit at the launch site and record telemetry for around 20 minutes to ensure that we had a dataset my daughter could use to complete her project. Then, we would pack up the antenna and give chase in the car.  We recorded our 20+ minutes and then, lost the telemetry signal. I still don't know if the balloon just got out of range or was blocked by the hilly terrain or if something else happened. We assumed it was range and that we would pick it up again once we hit the highway. We did hear the signal as we crested one hill but couldn't decode telemetry from it. Then it was silent.

We drove further and still didn't hear it. We had enough telemetry to know that our actual flight path was going slightly to the north of predicted which just lengthened our transmission path to the highway that we were on. There were no good options to try crossing the border into PA and then continuing the drive east so, we decided to pack it in. If we heard it again during our drive home, we would give chase. If not, we hoped that someone would find the payload and give us a phone call. it would be sad to lose the payload but we had data to almost 10km altitude and photos to prove how we got it. That, we decided, was worthy of one Cadbury Cream Egg in celebration.

Several hours later, we got the phone call we had hoped for. A farmer in Peach Bottom, PA (just over the MD-PA border, north of Bel Air, MD) had found the balloon and payload. Our celebration was now two cream eggs.

We drove up to PA over the weekend and collected the balloon and payload. All seemed to be quite intact and undamaged which was somewhat surprising given how much of the balloon appears to have come down with it. The only real exception is the missing UHF antenna. It looks like it got ripped off the solder joints at some point. Likely culprits are that it snagged something in landing or got twisted up in the balloon remains during descent. I doubt that we lost it early in the flight so, I don't think it's absence explains our loss of telemetry.

The onboard data record was quite complete so there didn't seem to be any operational issues during the flight. With a complete data file, I've done a quick reconstruction of the flight and the resulting data:

Our flight path from Frostburg, MD to the eventual landing in Peach Bottom, PA. Total flight time appears to be around 2 hours with a maximum altitude of 20689 meters or 67,877'. Not as high as we were hoping but, still pretty darned impressive. We definitely filled the balloon more than was necessary which easily explains the loss of ultimate altitude.

And, finally:

 Shea's measurement of temperature compared to the 1200Z balloon sounding done that morning at Dulles airport, about 50 miles to the southeast. We generally track well. We see the same low level inversion as seen at Dulles and roughly the same level for the tropopause. We don't track perfectly and I suspect this is a simple issue that our temperature sensors have a higher thermal mass than those used in professional radiosondes so our temperature response is slower than what was measured at Dulles.

Apart from that, I think this was a perfectly respectable showing for a first time building a balloon to measure something in the atmosphere. There's much to learn from for subsequent flights and, we will learn.

Launch attempt #2

New GPS is in and installed and the electronics are working again. Weather looks good for this morning so we are packing the car and heading to Frostburg to launch.

Project DEBO talks to the FT-8800R

Couldn't get two videos in the previous post. Here's a quick video of the Yaesu FT-8800R that will likely be the primary radio for the tracking and recovery effort since it is soon to be mounted in the car (I have access to an Icom 910 that has been used for previous balloon flights so, I can safely say I have three redundancy levels on the groundstation side of things). The radio is connected to a Pico-Packet TNC and to a generic serial-usb converter to my linux machine.

Project DEBO talks

Ahhhh, 'tis been awhile since I last updated the work toward launch. We are just a few pieces shy of everything we need. Only some PVC pipe and a few pipe fittings for a balloon filler and a relay to control our 'here I am' buzzer on descent/landing and then it's time for final assembly. Every sensor has been read by the arduino C&DH microcontroller. The temperature sensors are pretty much read directly from the analog-in ADC ports. Shea has that fully under control. The BMP085 pressure sensor is a little more complex but I found some example code on the Sparkfun.com website. So, even for that, Shea now has only to call two functions and then write the numbers into her telemetry string.

Reading in the NMEA strings from the gps is proving to be a bit of a headache. I haven't dealt this in-depth with strings in the arduino before. For all that it looks like C, I'm finding it doesn't really handle strings like C. It looks like there are a couple of issues I have to get around: string handling is not quite what I expect and there are issues with timing when the serial buffer has data and determining if we have the right data before the buffer overflows.

Once that comes together, though, the radio is operating and transmitting AX.25 packets at 1200 baud. I have managed to hear and decode packets on both my Kenwood TH-D7 and an FT-8800R with a PicoPacket TNC. If we can build the string to feed into it, we can send it out. I guess that means our two main hurdles to launch are reading/formatting the gps data and getting the weather to cooperate.

Next weekend is President's Day and there are some extra days off from school. If the weather cooperates, this will be the launch weekend. One of us will be frantically watching the weather forecasts while the other finishes coding.

It really starts to take shape

One major step closer to launch today. All the ordered parts except for the temperature and pressure sensors arrived in the mail this afternoon. chdk loaded fine in the camera and a stock intervalometer script worked first time out. We'll modify it from there and add video triggering. The new GPS from Argent Data Systems is built to work up to at least 25km altitude. Shea wasn't expecting her parachute to be in her favorite color and monogrammed with the project name so that turned out to be a fun treat (it's the little things, right?). And finally, our two balloons. They are a bit heavy in stock condition because of the way the fill neck is designed and the skirt that hangs from the bottom. I've read of others using these particular balloons cutting off the skirt and modifying the neck to cut down on envelope weight. I'll unpack one of these later this week and we'll decide what we want to do modification wise. Right now, we'll be hard pressed to break 2.5 pounds in the payload so the extra lift gained by the mods may not be all that necessary. This is such a light payload, I really wish I had time and money to put more of my stuff in it. Next launch....

Pretty much the whole kit in unassembled form. Payload box, camera, C&DH microcontroller, radio, GPS, parachute and balloon.

More parts

Well, as of this afternoon, most of the major components for Project DEBO are ordered and either shipping or waiting to ship. Next week should be busy with payload construction and testing. Today saw an order go out for balloons, recovery parachute and the temperature and pressure sensors. Thanks to Dale at orionspyramid, Mark at Rocketchutes.com and the gang at Sparkfun for making spending money sooooo easy. (I'll post something of a technical breakdown for the geek crowd over the weekend.)

Shea had to work through some balloon ascent/parachute descent simulations last night to pin down which balloons and parachute we would order. Once we had that, it was all eCommerce. I think, the only major component still to be ordered is a GPS unit rated to work above 18km. Both units I have here will fail above that. Since we hope to make it to 25-30km, I don't want to lose tracking information for what could be a crucial part of the flight. We'll need some incidentals beyond that (and Helium, can't forget Helium...) but we should have everything needed to put the payload together and start testing.

Parts are rolling in

Well Project DEBO has a body and eyes now. I picked up a small styrofoam container (like the ones used to ship insulin) from the lab and we pulled off an auction win on eBay for a Canon Powershot A530 camera. This camera works with the Canon Hack Dev. Kit (chdk) so we will be able to script camera actions during the flight.

Now that we have a camera, Shea needs to work on a quick mass estimate and then do some ascent/descent predictions with online calculators so we can size balloons and parachute. I'd like to get those ordered in the next couple of days.

Shea's first arduino code

One more step toward her eventual balloon flight has started: teaching her how to write control code for an arduino microcontroller. She took to it pretty well and quickly had a couple of LEDs blinking on and off. Tomorrow we talk about reading sensors.

/*
Blink
Turns LED on and off repeatedly
*/



void setup() {

pinMode(12, OUTPUT);
pinMode(4, OUTPUT);
pinMode(2, OUTPUT);
pinMode(8, OUTPUT);

}

void loop() {
digitalWrite(12, HIGH); //1 on
delay(500); //pause
digitalWrite(12, LOW); //1 off
digitalWrite(4, HIGH); //2 on
delay(500); //pause
digitalWrite(4, LOW); //2 off
digitalWrite(2, HIGH); //3 on
delay(500); //pause
digitalWrite(2, LOW); //3 off
digitalWrite(8, HIGH); //4 on
delay(500); //pause
digitalWrite(8, LOW); //4 off

}

High altitude balloons + Middle school

When my daughter first started talking about her upcoming science fair, I didn't think it would go in this direction. This would be her first real science fair and her first fair at a new school. She brainstormed for ideas and, with mom a planetary geologist and dad an atmospheric physicist, she was not at a loss for them. She really latched onto the idea of measuring the atmospheric temperature and pressure profile with a balloon though.

I hope this blog will be a fun place to share the development process and record her work on it. We will use amateur radio for tracking and telemetry so this will also be the place for posting transmitter and telemetry decode information.

The idea is simple: launch a balloon into the stratosphere carrying temperature sensors, a pressure sensor and camera. An Arduino microcontroller will run command and data handling (C&DH) for data collection and storage. GPS coupled with amateur radio will provide position and data dowlink. In the event that we can not recover the balloon, the hope is to have the downlinked data for a backup. We will publish our downlink info and try to give warning about launch times. Radio amateurs along the flight track (likely Cumberland, MD to points east) are welcome to listen in.