Tom somehow I missed the attachment last night.
I've not checked on how this is implemented under the hood but the formula I'm using makes calls to a log function.
I've not got my head around the impacts of atmospheric pressure on dew point nor have I looked at wet bulb/dry bulb calcs but my impression is that I don't need great accuracy.
http://en.wikipedia.org/wiki/Dew_poi..._approximation
There is some notes on using a 1-wire device as an A/D for barometric pressure sensing at
http://www.sensorsmag.com/articles/0501/34/
A range of I2C barometric pressure sensor's at
http://www.futurlec.com/Pressure_Sensors.shtml could work well if barometric pressure is important. Cheap and I'm already running a variant of the I2C for the Temp/Humidity sensor.
//--------------------------------------------------------------------
float calc_dewpoint(float h,float t)
//--------------------------------------------------------------------
// calculates dew point
// input: humidity [%RH], temperature [°C]
// output: dew point [°C]
{ float logEx,dew_point;
logEx=0.66077+7.5*t/(237.3+t)+(log10(h)-2);
dew_point = (logEx - 0.66077)*237.3/(0.66077+7.5-logEx);
return dew_point;
}