Comment by Andrew Rabbitt on February 22, 2012 at 3:24am

Interesting comment Monroe!  Thanks for the tip.

I guess that means there are a whole bunch of conflicting requirements to make this thing work.  I need to get it through the tropopause without icing up and dying on me, but also I need to keep it from cooking when it gets into sub-orbit as well as at sea level.

I will think some more! :)

Comment by Marco Glattfelder on February 22, 2012 at 3:55am

I launched a HAB Balloon with an arduino board (as camera controller) and some cameras up to 22'500m. I had -53C external and +9C internal temperature. I used a simple foam box (about 1 cm thick... a box used in Italy for fresh mozzarella's) and an one of these gold-like plated emergency blankets. My experience is that there is no real need for an internal resistor. The best practice is to enclose all in an well insulated box and the external temperature won't be an issue.  

Comment by naish88 on February 22, 2012 at 5:42am

Have you ever thinked about peltier cell? keep warm when it needs and keep cold just reversing polarity, quite simple!

Comment by penpen on February 22, 2012 at 5:55am

Well insulated box works very well, indeed, 15km without any prob

Comment by paul hubner on February 22, 2012 at 8:39am

I had the pleasure in participating in a few high altitude amateur launches.  Cold was a worry the first time too. We added a chemical hand warmer pack, like Monroe mentioned, it turned out to be way overkill.  In fact on decent, the temperature differential caused a ton of condensation on the electronics on the way back down.  There have been four launches reaching from 27 to 32km. never had any issue with cold. Check out the data data at www.HAPB.net.

Comment by Helldesk on February 22, 2012 at 9:06am

Does the barometer function reliably in an enclosed box? Obviously the thermal insulation won't become an airtight pressure vessel, but can the waste heat and other internal temperature generation in general skew the altitude calculation? That is, is it able to do its temperature compensation correctly in every situation?

Comment by David on February 22, 2012 at 9:16am

I generally agree with Paul; I've flown nearly 100 HAB payloads, many of them with very exposed electronics, and have never had a temperature-related hardware failure.  There are temperature sensitive components (for example, we use TTL timing sources rather than crystal oscillators to send clocks into our microprocessors, and you want to stick with a lithium-based battery chemistry), but most components are pretty OK with getting cold.  Is there a specific temperature sensitivity that you've identified on the APM?

If your APM will be in the same box as your transmitter, you'll have a heat source "built in."  20C temperature differentials aren't uncommon in our relatively uninsulated containers that contain the transmitter, though, admittedly, we're running the TX pretty hard. 

Alternatively, you could use dry ice to get the APM that cold at standard atmospheric pressure... find out what stops working.  Just be careful about how you cool it down if you go that route - it is possible to break stuff by transitioning the temperature too quickly.

On very early flights, we ran resistive heaters out of an abundance of caution, but we quickly found that the weight (particularly the battery) and complexity of those systems could be much better used elsewhere in our payloads.  So... if it's necessary, absolutely implement it, but I'd suggest making sure that there's not an easier solution.

Comment by Jonathan Lussier on February 22, 2012 at 9:18am

I agree with Paul, penpen and Marco's comments, no worries if the box is well insulated.  

That said, the atmospheric model shown above is the 'standard.'  If you're looking to see if your box will perform well, you need to add standard temperature excursions as well (+/- 15 deg C at that altitude) to your model.

To give you an example.  At sea level, the temperature can be classified (on the hot side) as ISA, Hot Day (ISA+20degC), or extreme hot day (ISA+35degC).

Good luck!

Comment by Andrew Rabbitt on February 22, 2012 at 4:22pm

All good info - thanks guys!

I am mostly concerned with keeping the gyros and accelerometers at a relatively stable temperature to avoid the need to run an extensive calibration exercise.  The gyro gains are apparently particularly sensitive to temperature extremes.

Thanks for the links to the HAB projects.  I enjoy reading through other's experiences.

@Jonathan, the temperature at the tropopause doesn't change that much, so -56°C is a good minima to work with.

Developer Comment by Randy on February 22, 2012 at 10:51pm

The barometer has a built in thermometer and the AP_Baro library has a get_temperature call that might be useful for your PID controller attached to the heater.  Perhaps you've already seen that but though I'd point it out in case you hadn't.

Comment

• ‹ Previous
• 1
• 2
• 3
• 4
• Next ›
• Page