I ran across this very issue when trying to calibrate my barometer chip against 
the NWS station located less than two miles away.  Their numbers for millibars 
and inches of mercury do not agree.  I sent them an email and asked what was 
going on.  They said their instruments read out in millibars (to three decimal 
places) The reported value is converted to sea level pressure and reported to 
two decimal places. They are also converted to inches of mercury for their 
reports.  Only problem is their conversion constant is NOT the proper value.  
They consistently report around 0.02" too high.   I reported this back to them, 
 but have received no further responses.
Note that the conversion between true pressure readings and sea level pressure 
involves an equation with about a fifth power/root (depending upon the 
direction of the conversion) so it can be quite sensitive to true chip 
calibration.  The pressure chip that I am using (MP5611) is factory calibrated 
and has calibration constants stored on-chip (the Bosch BMP085 and BMP180 chips 
also do this),  but the soldering process can affect the chip so you need to do 
some final calibration.  The MP5611 can detect the air pressure change seen by 
raising the chip less than 6 inches...
Relevance of temperature/humidity/pressure sensors to time-nuttery?  We all 
know the comparatively massive effects of temperature on our equipment.  But 
humidity and air pressure also affect them in many subtle and not-so-subtle 
ways.  I'll post some recommendations/observations on various sensor chips in a 
while.
---------
One funny thing about weather measurements is that the data that NOAA reports 
is not what it would seem. The standard ASOS data (which is what NOAA reports 
in its local current conditions) includes barometric pressure in inches of 
mercury and in hectoPascals. It turns out that neither is the actual barometric 
pressure. 

                                          
_______________________________________________
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Reply via email to