Re: altitude from battery/sensor temperature
Jeffrey Warren
temperature... maybe it's the EXIF program I'm using, "exif" (linux).
Attached is the output and the first image of the flight. Ideas?
On Fri, Feb 10, 2012 at 9:13 AM, Jeffrey Warren
<je...@publiclaboratory.org> wrote:
> What altitude resolution does 1C resolution get us? Is that the max
> resolution of the sensor? We should try this with an existing balloon flight
> dataset.
>
> On Feb 9, 2012 8:24 PM, "andrzej zaborowski" <bal...@gmail.com> wrote:
>>
>> On 8 February 2012 22:46, Jeffrey Warren <je...@publiclaboratory.org>
>> wrote:
>> > truly original, noah.. very cool.
>> >
>> > Is it possible to insert temperature into the exif tags of each image?
>> > Then
>> > we could parse it in MapKnitter and when you place the image the first
>> > time
>> > it'll make a guess at the scale.
>>
>> By default Canon cameras do store the CCD temperature at 1C
>> resolution, in an exif tag. Makes for nice graphs.
>>
>> Cheers
andrzej zaborowski
> I checked a set of images from a Canon a490, and didn't see
> temperature... maybe it's the EXIF program I'm using, "exif" (linux).
> Attached is the output and the first image of the flight. Ideas?
My desktop is down and I'm on travel, so I went installing all of the
exif readers for Linux on the netbook and it seems like each one gives
you a different set of tags. "exiftool" is the only one that gives me
the temperature and a whole lot of other things the other programs
don't.
For the example image it says 31C, pretty hot :)
The theory is that, on average and up to some altitude, temperature
decreases 1C per 100m up. I have this winter graphs on picasa [1]
that seem to confirm it, you can see when I'm unwinding about 500m of
line and then when winding it in although the change is slow. The
bigger cameras have more "inertia" and the really old ones actually
generate enough heat the you can't judge on ambient temperature at all
:(
The other day I grabbed the JPEGs Tim Zaman posted from his
university's "near-space" balloon launch with a Canon point-and-shoot
and then later Google published similar pics from their "android works
even in space" balloon, and got those graphs with some extreme
temperatures: [2][3] :)
Seems like -20 is the minimum the camera can capture.
Cheers
1. https://picasaweb.google.com/108793263944050035068/AerialShots#5565593371272275986
https://picasaweb.google.com/108793263944050035068/AerialShots#5565595774058283218
2. http://unadventure.files.wordpress.com/2011/01/hohoho-temps.jpg
3. http://unadventure.files.wordpress.com/2011/01/android-temps.jpg
andrzej zaborowski
> On 10 February 2012 15:56, Jeffrey Warren <je...@publiclaboratory.org> wrote:
>> I checked a set of images from a Canon a490, and didn't see
>> temperature... maybe it's the EXIF program I'm using, "exif" (linux).
>> Attached is the output and the first image of the flight. Ideas?
>
> My desktop is down and I'm on travel, so I went installing all of the
> exif readers for Linux on the netbook and it seems like each one gives
> you a different set of tags. "exiftool" is the only one that gives me
> the temperature and a whole lot of other things the other programs
> don't.
>
> For the example image it says 31C, pretty hot :)
>
> The theory is that, on average and up to some altitude, temperature
> decreases 1C per 100m up.
By the way the cameras also save the target distance in EXIF. I don't
know how it is measured technically, but in low flights (assuming the
camera is pointing down, is not set to focus on infinity, and is below
65.536 m from the ground), it'll surely give you better accuracy.
Cheers
Jeffrey Warren
http://publiclaboratory.org/notes/warren/2-10-2012/temperature-during-thatchmore-farms-flight
Kinda weird, it goes up towards the middle of the flight.
andrzej zaborowski
> Cool, i extracted temperatures and times and graphed them:
>
> http://publiclaboratory.org/notes/warren/2-10-2012/temperature-during-thatchmore-farms-flight
>
> Kinda weird, it goes up towards the middle of the flight.
Yes, kind of the opposite of what you'd expect. I've been using Canon
A490's and they didn't generate much heat, although they did a little
more than the newer ixus ones.
Can the difference be between balloon (which, in some range, moves
together with the air) in full sun, and kite, where there's a constant
ventilation?
In the Tim Zaman "near space" graph the camera was I believe a dslr in
a styrofoam box, ccd temperature apparently reaches 30C despite the
-60C outside. (but very low air density)
BTW, the hypr3d model is awesome!
Cheers
