Internal pressure sensor

Christoph Mair ml at chonyota.net
Thu Oct 1 22:30:16 CEST 2009


Am Donnerstag 01 Oktober 2009 16:57:18 schrieben Sie:
> On 30 Sep 2009, at 22:18, Christoph Mair wrote:
> > ...
> > I successfully added a pressure sensor to my Freerunner. The BMP085
> > chip from
> > Bosch Sensortec is a small (5x5x1.5mm) chip which includes a
> > pressure and a
> > temperature sensor. Power and I2C is enough to get it working. I
> > glued it next
> > to the BT antenna. The wiki page contains some pictures:
> > http://wiki.openmoko.org/wiki/I2C_Pressure_Sensor
> > Sourcecode is available from http://gitorious.org/freerunner-navigation-
> > board/bmp085
> 
> Cool! How accurate do you find it?
I think I can get about 0.5m (the datasheed speaks from 0.25m and better with 
software averaging), but it seems that the temperature compensation is either 
buggy or does not work good enough. The running phone (means: not suspended) 
heats the sensor up and the measured pressure increases slowly. I'll try figure 
out what to do.

> This makes the Freerunner a nice mobile for glider / hang-glider /
> paraglider pilots. They most always carry a varioaltimeter with them,
> and last time I flew (some years ago) GPS was beginning to have a
> significant impact upon that (niche) market. The combination of
> barometric pressure with groundspeed allows one to easily find the
> optimal speed-to-fly, allowing for head- or tail-wind on a glide
> between lift sources (thermals).
Unfortunately I'm not a pilot, but this seems to be rather useful.

> I'm sure that this sensor must be at least as accurate as those used
> in the cheap commercial ($250) varioaltimeters of 10 years ago, so it
> looks really good for this sport, if someone has the time to make a
> nice interface.
I know some Qt and could probably hack a simple but ugly interface, but right 
now I'm busy with other stuff.

> It would be interesting (for me, anyway) to test by climbing a small
> hill and see if the pressure difference is reflected accurately. 30m
> should be a good initial test.
I will do this as soon as I can find a hill ;)
First I need to figure out where the temperature dependency comes from.

Christoph



More information about the community mailing list