SF-HAB Launch @ HackaDay Superconference

We will meet prior to HackaDay’s Supercon on Sunday, November 6 at 9 am in front of the the Los Angeles College of Music, 14 East Del Mar Blvd, Pasadena 91105

If all goes well you can track it on 20 m WSPR,  WSPRnet.org or some of the alternate sites listed on WSPRnet as  W6MRR & W3HAC.  It will be moved over to APRS as soon as it shows up on WSPR .  WSPR callsign W6MRR-26 & W3HAC-11.

New Trackers Underway

Kazu AG6NS and I have been working on a couple of trackers. One is a dual band 2/20 m APRS/WSPR with other modes coming. It is based on an ATMega1284 processor, which is a 328 (used currently) with more memory and and running at 10 MHz, the safe limit for 3.3 V operation. Kazu has come up with code for it and I need to catch up with a revised PCB.

The other is a Raspberry Pi Pico based tracker. This is a longer range project. It will use the same Si5351 based modulator and filters for 2/20 m. The Picos are attractive as they have a dual core ARM Cortex-M0+ @133MHz which is far more powerful than the ATMegas.

The RP2040 chip has instructions for DIYing a board but I went the Raspberry Pi as it was the easiest path to start working on a design.

Due to the chip shortage, I may need to use through hole components for the output filters.

We are looking at a lot of fun things that can be done – SSTV data and cam, voice ID over VHF, crossband repeaters, and latex floaters. If you’re interested in helping out out or at least join in please get on our chatter list.

Launch on September 4th at 9am

We’re launching a pico balloon this Sunday morning, September 4th at 9:00 AM pacific time.

If you’re in the area and would like to see the launch, please join us in the Milpitas, CA area in the Ed. R. Levin County Park.

See the map link below for the exact location.


HAB Launch Site On Google Maps

New Tracker Underway

Kazu AG6NS ran out of space for more things like 2 m APRS and 20 m data over SSTV. I had tried a dual 2 m/20 m tracker before but gave up due to self interference. I was using a 6 g 0.5 W FM VHF transmitter module and it caused the ATMega1284 to lock up. QRP Labs had this problem and solved it with a metal can over the microcontroller chip. I had some leftover 1284s and am spinning up a new board. It will run at 10 MHz/3.3 V to conserve power. The 1284 has 128K flash vs 32K on the 328s. Kazu has already demonstrated APRS on 2 m but the 328 didn’t have room for WSPR and SSTV. A Si5351 with be used to synthesize the output signals. A separate LPF is on each of the the 3 outputs and a BME 280 is used for temperature, air pressure and humidity. I think we’ll need amplification to get anywhere with APRS but not 0.5W. If you’d like to help out with this – let us know.

Launch on May 14th at 9am

We’re launching a pico balloon this Saturday morning, May 14th at 9:00 AM pacific time. The callsign on WSPR will be K6RGG-14.

If you’re in the area and would like to see the launch, please join us at the Berkeley Marina on the north side on Marina Blvd. See the map link below for the location.


Choosing the time and place

For a launch

The time and place to launch your picoballoon is important. Things to consider are location, surface wind, cloud cover, the path the balloon will follow during ascent and the path it will follow one at float altitude. Consider the things that can go wrong. The balloon can drag across the ground, snag a tree or power line, drift into lee turbulence from a bluff, dip into water, get into clouds and get wet, get drawn into an updraft in a cumulus or cumulonimbus cloud from below, float over a growing thunderstorm and so on. With a little thought the list would get much longer. So, plan the time and place…..

Choose an open spot. It can be a field, dock or a beach. If the day is very still, almost any place will work. At this point, you should have a pretty good idea of the float altitude and some idea of the rate of ascent. If it is an SBS-13, they mention the rate of ascent as a function of free lift. I you are using clear chinese balloons, you probably should learn by experiment. For a first guess, if you have to guess, use between 1 and 2 meters per second.

Use the windy.com to see what the surface winds at your chosen location will be. You can look days in advance for planning but look for wind of 5 miles per hour or less. You can probably launch during higher wind speed but the risks of problems increase.

Having chosen the launch location and time, go to to http://predict.habhub.org/ which will predict the trajectory during the rise to float. Input the location, the launch time, and where it asks for burst altitude, put float altitude. Put in your expected rate of ascent and for the descent rate, put in a very small number as the program exects a bursting balloon with a payload returning by parachute. If you put in a very low descent rate, the line it draws will be the direction it will follow at float.

The rate of ascent for slow balloons like we send up is pretty much constant all the way up. It is also proportional to lift so, double the lift, double the speed. It does not act like an airplane where drag is proportional to the square of speed. It is too slow and the flow is laminar. Once the balloon is fully inflated it will slow and reach float altitude. The approach to float has math like the discharge of a capacitor. The time constant is the distance between full inflation and float divided by the rate of rise. There is no need to worry about the approach to float but it will be in the last 15 30to minutes of the ascent, depending on your free lift.

The above program does not take that into account but don’t worry, it is trivial.

At this point, you have a fairly good idea of where and when the balloon will begin float. It is time to run HYSPLIT.


Put in the altitude and time. Select how far in the future for the prediction you want to see and it will tell you where the balloon will go. This program is specifically written for superpressure balloons so it is perfect. Just what the doctor ordered. It will take a little time to sort out the inputs as there is a lot to select and it too much for me to detail here.

After you know where the balloon is going to go, switch back to windy.com. Look at the cloud tops information. What you want to understand is if you will be floating into a storm or a cloud. A storm can tear things apart and a cloud will probably get it wet so it will descend. Remember it has no net lift at that point so every extra gram is a problem. There is nothing you can do about these things but evaluating before launch might save the balloon for later.

Assemble the balloon and payload. Half the dipole is above and half below for wspr. Just a 2 meter whip is below for the aprs trackers. Make sure the line from the balloon to an aprs tracker is long enough that the beacon and balloon are unlikely to bump in turbulence.

So, you have assumed a location and time. If you are using 2 meters and aprs, the balloon is easily launched by one or two people. Release the balloon upwind from the beacon and release the beacon when directly below the balloon. You don’t want it to swing down and drag on the ground. You might not want to launch in a wind faster than you are willing to run!

If you are using a tracker on 20 meters and using wspr, the antenna is significantly longer. I suggest three people but it can be done with less. Line up with the wind, release the balloon, keep the payload below the balloon until the line is straight and overhead and release. The third person keeps the line from tangling in weeds and so on.

Now the balloon is floating into the sky and all you can do is watch. Make sure you get video to share as it is easy to get engrossed in the launch and get no video. That is a good job for spectators and they are usually eager to be there.

Now watching the progress is the fun part. If you have a 2 meter aprs beacon, it will show automatically on APRS.fi. Once it is over about 10,000 feet, it will be loaded to https://tracker.habhub.org/#!mt=roadmap&mz=11&qm=1_day&f=AD1L-11&q=!RS_*;

For APRS.fi, put in the balloon call sign to see where it is. For the habhub tracker, the column on the left lets you select the balloon you wish to see and clicking on it will zoom to it’s location. Both sites give some telemetry coming from the beacon. Using APRS will give good tracking over North America and Europe. It might be picked up in China and Japan and again as it comes back to North America. There is lots of nail biteing while you wait.

If you choose to use a wspr beacon, the coverage is wonderful and you will almost always know where the balloon is. Night and very high latitudes will stop transmission but it still is just great. Usually the signal is sent on 20 meters and can go as far as 8,000 miles. You will know where it is over the ocean and over the land. Some nations make it illegal to transmit in their airspace and the beacons stop sending in that space. That would be primarily England and North Korea. There may be GPS jamming over some conflict areas in the middle east. The problem with wspr is the signal does not get picked up by aprs.fi. It is picked up by the wspr network. You can run a python script to go to wsprnet.org and download the data, convert the format and upload it to aprs.fi. That script is on Github. You can also ask for the tracking to be done by http://lu7aa.com.ar/wspr.asp?other=lu7aa and the map on his site will give you a great display.

It is also interesting to see what the jet stream is doing and where your balloon is relative to that. The jet stream can be seeh here: https://www.netweather.tv/charts-and-data/jetstream . If you would like to see what the air pressure at high altitude to see if the balloon is at the altitude/pressure you expected, go to http://weather.uwyo.edu/upperair/uamap.shtml . If you would like to study what pressure the balloon floats at, http://weather.uwyo.edu/upperair/balloon_traj.html, written for tracking balloon ascents and descents will give pressure and altitude information that can be used to interpolate the pressure from the gps altitude.