Hackaday Prize Entry: A CPU For Balloons

Launching a high altitude balloon requires a wide breadth of knowledge. To do it right, you obviously need to know electronics and programming to get temperature, pressure, and GPS data. You’ll have to research which cameras will take good pictures and are easily programmable. It’s cold up there, and that means you need some insulation to keep the batteries warm. If you ever want to find your payload, you’ll also need an amateur radio license.

There’s a lot of work that goes into launching high altitude balloons, and for his Hackaday Prize entry, [Jeremy] designed a simple embedded data recorder capable of flying over 100,000 feet.

This flight data recorder for balloons is based on the ever popular ATMega328, and includes humidity, temperature, pressure, accelerometer, gyroscope, and magnetometer sensors. All of this data is recorded to an SD card. The Real Engineers™ who are wont to criticize design decisions they disagree with might laugh at the use of a 7805 voltage regulator, but in this case it makes a lot of sense. The power wasted by a linear regulator isn’t. It’s turned into heat which keeps the batteries alive a little bit longer.

This balloon data recorder has already flown, and [Jeremy] got some great pictures out of it. It’s a great piece of the puzzle for an exceptionally multidisciplinary project, and a great entry for the Hackaday Prize.

17 thoughts on “Hackaday Prize Entry: A CPU For Balloons

  1. [sarcasm] Drones, Balloons should all be illegal. Lets keep the guns, makes shooting down drones easy :-) [/sarcasm]
    Serious note: You cannot let a balloon fly in commercial flight paths. 1 accident and no more tinkering or exploration for mortals is allowed.

    Don’t cause wrecks!

    1. We’re screwed when space balloon kits appear in Walmart.

      Though, it’s not just when they’re “off the shelf” and in the hands of the great unwashed. Remembering an argument I had a couple of years back with a drone “pilot” though pilot is too grandiose a word, he knew enough to put his own together, but he was being a total jackass, flying it in the overlapping approaches of 2 heliports and one international airport, highly controlled airspace, and over crowds and was all “Those rules don’t apply to me…”

  2. I’ll be “that guy”…

    While you might think you’re being as cunning as an orange toque to use a 7805 as a heater, doesn’t it have a negative feedback problem? It becomes more efficient at lower temperatures, resulting in less heat, the battery voltage also droops at lower temperatures leading also to less “waste” in the 7805.

    1. No, it is negligible. There is a certain heat dissipation in the atmosphere that settle the system to a given value. For whay concerns the battery efficiency, depends on the battery and on the settled temperature mentioned herein.

    2. I’m confused. Isn’t most of the heat a function of the load current through it and the voltage drop across? If they stay the same, the amount of heat generated (not temp) stays the same.

    3. Hey all! Glad you guys enjoyed this project. I’ll settle the 7805 debate real quick for you– I used the 7805 because I had extra lying around and it is very easy to use. I guess it could act as a ‘heater’, but that did not factor into the design decision. Sometime engineering requires you to build off your inventory, and forgo best design practices. I appreciate the discussion on use of the 7805 though!

      1. To settle a debate you might want to comment on what’s being discussed.
        RW and Enrico S. are still talking out their ass.

  3. The Meade hand controls for their go-to telescopes us the heat from the voltage regs to keep the lid warm so it does not black out in cold temps outdoors. It does not take much.

    1. Hi Macona! I actually used hand warmers inside the payload box to keep non-sensor electronics warm, like the trackers and cameras. Our next project will monitor the temperature inside and outside the box so we can accurately compare results.

    1. Accurate temperature readings of what? It’s taking an accurate temperature reading of the board’s ground plane. It definitely isn’t taking a measurement of the ‘outside’ temperature.

      Remember, there’s no real ‘ambient’ in the stratosphere, because there’s virtually no air, which means virtually no convection. So the fact that it’s stuck in an “insulated” package doesn’t mean anything – it’s all the temperature of whatever it’s conductively connected to.

      Which, by the way, is why it’s heating up steadily over time. In general balloons don’t have a problem keeping things *warm*. They have problems keeping things *cool*, although a balloon project I work on has parts which are efficient enough that you have to paint them black in order to get them at a decent equilibrium.

      If you really wanted to measure the temperature of the *stratosphere*… stratospheric temperature measurements don’t use temperature probes like that. They measure how much power is emitted in the microwave (e.g. a microwave radiometer), because the heat transfer from the atmosphere to a normal temperature probe is practically nothing.

      1. “Remember, there’s no real ‘ambient’ in the stratosphere” is a false statement. Yes there is no convection, but temperature increases with altitude in the stratosphere. The ozone is located in the stratosphere, and those molecules absorbs energy from the sun and turns that into heat.

        If you look at the document write up on the payload box, you will see that the SI7020 is located outside the box. This should give you a very accurate temperature of the atmosphere.

    2. Hi RichV! I tried my best to place the FlightCPU outside the insulated payload to get the most accurate readings. To achieved this, we cut out an inset on the outside of the box, and placed cling wrap over the opening to protect the FlightCPU from moisture. Of course this method isn’t perfect, but the graph of the temperature readings matches nicely with the expected result, so we are satisfied for the first trial!

  4. Today all payload electronics run fine from 2.8V to 4.2V the voltage available from ONE lithium cell, so offcourse no one find it smart to use 7805 regulators, since no one will use more cells that nessasary, weight is also an issue. The battery must be keept warm, at least over a minimum temperature, else it will fail ! the solution : leight weight isolation AND temperature controlled heating of the battery. I designed a VERY small and simply solution for exactly this, glued it directly on the cell, since it is temperature controlled, it draw no current at ground level, we monitor all temperatures live via APRS so before and after curves exist for our research. Also be warned about isolation, not all types works the way, check performance versus weight.

    1. Thanks for the comments Thomas! The next revision I am working on separates the battery control from the sensor board, thereby permitting us to regulate the power without interfering with the sensor readings. Also, I like your idea of using a lithium cell, that is much more practical than using a 7805 and AA batteries, although more complex!

Leave a Reply to jcope207Cancel reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.