I’m working on integrating an altimeter into a lightweight kite (simple square type - generally made of paper) to track altitude over time, as well as any other data I can get (possibly an accelerometer). The sensors themselves and even Arduino are quite light, but the power source is a bit tricky. I need advice on the following:
– what Arduino is best for this, the smallest that needs the least power (and makes it somewhat easy to get the data off tho I can handle that with USB or bluetooth)? would lilypad be a good solution, as I can sew it into the kite more effectively?
– what’s the smallest battery power source I can use? 1 coin cell won’t cut it - better to do 2 coin cells or Li-ion? the Li-ion seems heavy though I have not yet tried to fly a kite with it attached
– any other solutions come to mind or hacks that might not need Arduino?
How large would the kite be? I haven’t flown one in years but as I recall the one’s I flew would lift quite a lot. I would think a small Arduino and a lipo battery would be fine.
The other parameter would be for how long do you want to fly or record data? If it’s only an hour or so then again a small lipo like [this one would do fine.
Will the Arduino record the data for later downloading or will you want to download in real time? If the latter you might consider a radio integrated into your Arduino like [this one. Unfortunately the kite is likely to fly too high to use Bluetooth (too far between transmitter and receiver).
I think the flying and recording time can be kept fairly short - though burning through batteries could end up being an issue. For now, I think batteries will suffice.
In terms of transferring the data - for this initial experiment, I think it’s fine to get the data off after flying. I’m going to be working with kids in neighborhoods where they fight kites - so I need to try to make the process as fast and easy as possible, but it can start out being a little manual and then more automated as I test.
Thanks for the radio tip - I’m going to check that out! I think it would be amazing if the kite could tweet it’s data, but again I think I’m going to quickly get into too much weight.
Hmmm, some of those look pretty small. If you are talking about one the size of a sheet of A4 paper then the electronics and battery might well be too heavy. If you are talking about something 45 to 50 cm (18 to 24 inches) on a side then I don’t see a problem. Properly configured an Arduino with altimeter, accelerometer and radio powered by the battery I referenced should last several hours on a charge. There are smaller and larger batteries available.
The small radio in the kite Arduino would need a matching receiver connected to a computer on the ground. If that computer had an internet connection there is no reason why it could not be tweeted. (some programming required )
Hmmm, if you are talking about fight kites you need to consider a couple of additional things. One: where the weight of the electronics and battery is placed will affect the maneuverability, stability and agility of the kite. It will be quite the balancing act to place it correctly. I was thinking to dangle the electronics from the string or the tail. Both of these would actually increase stability but that would severely hinder a fight kite. Two: you will need to consider how to protect the electronics from crashes and collisions. On a stable kite I would say just tie it on and go. On a fight kite you might need some sort of enclosure which adds even more weight.
The OP might start by putting some weights on the kites and seeing just how much can be carried. To minimize the weight, it cries for a custom board w/only the absolutely needed components. No reset switch, no voltage regulator, no FTDI, no crystal, set the onboard oscillator to the lowest frequency useable … the bare minimum. If the flight times are really short, perhaps use a coin cell battery.
But I think it all starts with some idea of the weight that can be lofted.
Too bad the uLog isn’t offered any more. It might have been a good start.
thanks EVERYONE! super useful input from everyone. i think you’re all right in that I’ll try with the Li-ion and see how I can get it balanced and flying. I am looking at the lilipad too because it’ll enable me to stich it into the frame of the kite in different ways and it might be easier to use it as a teaching tool in workshops with the kids.
quick questions re coin batteries - I need more than one to power the Arduino Mini or Lilipad right? if 2 it is getting close to the weight of the Li-ion so may not be worth the hassle…?
Some coin cells are lithium based, a CR2032 for example, and so are around 3v nominal voltage. They’re quite limited in output current but perhaps just enough for an MCU, memory and a barometer. A CR2032 weights about 3g.
What’s your guess for sampling rates and recording times ? How many samples/sec of altitude readings and for how long ?
Ah thank you! I may try that if the Li-ion is too heavy.
I have to get a sense of how long kids are flying those kites - but my guess is that it’s not too long a time period. I think I’d want to capture a fairly constant stream of data with some delay so that I can play with some visualization of the kite’s path. I want to get one up and running quickly so I can see if the altitude is enough to be interesting or if I need other data in the mix.
cbernett:
I think I’d want to capture a fairly constant stream of data with some delay so that I can play with some visualization of the kite’s path. I want to get one up and running quickly so I can see if the altitude is enough to be interesting or if I need other data in the mix.
Yeah, I wonder if just "dynamic" altitude data is going to be interesting enough ... let alone for a kid. I was thinking that some IMU type of instrumentation (9 "DOF") would yield a more interesting flight reconstruction, but then you're adding more weight and needing more power, which is more weight. And even then I'm not sure you'd get a good representation in 3D space of the kite's position vs time.
But then I thought … what if you measured the amount of string paid out ? Perhaps measured the compass heading of the string (somehow) ? Even given the string will form some catenary shape, perhaps that info would be enough to correct drifts, etc, in the IMU data ?? I don’t know but I raise the questions hoping to stir some discussion on the matter. No doubt there are people who’ve done this type of thing before.
- so I need to try to make the process as fast and easy as possible, but it can start out being a little manual and then more automated as I test.