[wxqc] Sensor drift

[spamfree]

Hi,

What he (and Davis) are saying is that you should set your console elevation to 0 ft (this effectively turns off the internal sea level reduction algorithm so only sensor/station pressure is displayed) and then calibrate the barometer to match a nearby airport. The calibration sets a fixed barometer offset that is added or subtraced from the sensor pressure to obtain the displayed pressure. Based on my understanding of pressure, I think this is not good advice, especially at higher altitudes. Here's why:
 
Let's say I have a VP2 console with a perfect barometer, and I have it sitting right next to the barometer at an ASOS station at an airport at 4000 ft. elevation. I set the elevation on my console to 0 ft., and since it's perfect, it shows a sensor pressure of 860 mb which perfectly matches the sensor pressure of the airport's ASOS barometer. A week later I see that the sensor pressure now shows 890 mb which again perfectly matches the ASOS barometer's sensor pressure. 

Now that I am sure my barometer sitting next to the ASOS barometer is perfect, I decide to take Davis' advice and make it show altimeter. Although the ASOS barometer has a sensor pressure of 890 mb, the airport's current altimeter value derived from that sensor pressure and elevation show 1029.74 mb (see http://www.srh.noaa.gov/elp/wxcalc/altimetersetting.shtml for calculation). So, I set my perfect console to 1029.74 to match the airport's altimeter setting. Internally, the VP2 sets a calibration offset of +39.74 mb (i.e. 890 + 39.74 = 1029.74).

A week later I return to the airport. The pressure has since fallen, and the ASOS sensor pressure is now back down to 860 mb. The derived altimeter value for the airport is now 995.96 mb. Internally, my perfect VP2 console also has come up with a sensor pressure of 860 mb. But, much to my amazement, my perfect console is displaying a pressure of 899.74 mb (860 + 39.74 = 899.74). Somehow my perfect console is off by over 6 mb. It doesn't match the airport's altimeter that I calibrated to, and now my CWOP quality check has a big red X. Hmmmm!

If you have a VP station, and you report data to CWOP, and you are not near sea level, then if you set the console's elevation, and submit the console's sea level pressure you will drift off the surrounding stations that report true altimeter as temperature changes. If you set the console's elevation to 0 ft and calibrate to a nearby station you will drift off the surrounding stations that report altimeter as the pressure changes. The only way you'll stay where you should be is to use the accepted pressure formulas to convert the VP's SLP or sensor pressure to an altimeter value and submit that. Currently WeatherDisplay, VPLive, and PC-WeatherStation do that. I don't know of any other software that does.

Steve
  
======= At 2006-05-23, 17:30:36 you wrote: =======

>Actually, either Davis misspoke or your misunderstood them.  If you don't
>apply an altitude correction to the barometer - either via the console or
>via software (which will then only reflect the correction in the software
>and not on the console), you will display "station pressure" - the actual
>barometric pressure at your station.  Use of an altitude entry allows for a
>correction to sea level, which is what is reported in media sources and by
>NWS.  Altimiter pressure is calculated slightly differently as Steve has
>stated (and is indeed what CWOP ideally would like to have reported, but at
>present, only VPLive and Weather Display can report altimeter pressure to my
>knowledge.)