Starting from scratch

Having found out that the Davis system I’ve been using for over 10 years is inaccurate for temperature when the sun shines, I decided to start again from scratch on a new computer with a different weather system and a better-sited temperature sensor. Re-siting the Davis was not physically possible because of its construction. Attached is a screen shot of the new non-Davis system showing the max temp today at 37.0


Expecting dead nutz ball accurate readings from hobby stations is unattainable. You need to spend a lot more money if you want scientific results #-o

[quote="NorCal Dan post:2, topic:64294"] Expecting dead nutz ball accurate readings from hobby stations is unattainable. You need to spend a lot more money if you want scientific results #-o [/quote] I agree 100% but we can try. Even professional Stevenson screens vary in their results http://onlinelibrary.wiley.com/wol1/doi/10.1002/joc.4287/abstract https://wattsupwiththat.com/2008/01/14/a-typical-day-in-the-stevenson-screen-paint-test/ http://server1.nuge.com/~weather/WX-Meas-of-Temp-shelters.pdf etc. etc.

I quote from the last article:

Accurate measurement of air temperature in the presence of strong radiation presents difficulties that are not easily resolved. The factors affecting error due to radiant exchange were analyzed, and the importance of sensor size was shown to be critical in this context. A number of alternative designs for instrument screens were evaluated: though none was found to give full protection from radiant exchange, several jury-rigged screens were found to give similar results to some commercial products. Until thermometer screens are developed that eliminate error due to radiant exchange, a universal method of calibration and labelling should be adopted, so that errors in measurement resulting from the response of the specific screen to combinations of environmental factors may be assessed.
Re-siting the Davis was not physically possible because of its construction.
why do you say that? I have my Davis temperature sensor in a stevenson screen (by extending the wires)

I applaud your desire to attain the most accurate readings possible. We all do the best we can, but in the end we are still hobbyists, our data will always be viewed as suspect by the professionals. Best of luck in your quest for exact readings.

Thanks, Dan! I don’t claim my readings to be exact but what I’ve learnt from the exercise is that the professionals are not exact, either. It is telling that the pros still use shields invented in the first half of the 19th century and that the results vary from model to model and the paint used on them! If our data are viewed as suspect by the pros, this amateur is beginning to view the data from the pros as suspect, perhaps to a lesser degree.

@Brian. I did not know that it was possible to move the temp/humidity sensors from a Davis system. In my ignorance, I thought it was accurate to

If you can keep running your Davis setup alonside your new system for a years or two you should be able to work out what corrections are needed to bring you old data in line. Well worth doing rather than throwing away all of your historic data.

[quote author=Devil link=topic=65454.msg525584#msg525584 date=1501657235]
In my ignorance, I thought it was accurate to

Thermistors are quite stable:

[quote]U.S. Sensor has conducted studies with similar results to studies performed by NIST (National Institute of Standards and Technology) that show thermistors can be produced with typical drift of only 0.001

For the record Davis has not used thermistor temp sensors in the ISS for some years, as I recall they were phased out quite early in VP2 production. Since then they have used the Sensirion SHT series of sensors. Currently they are using an SHT3x device as the SHT1x series are being discontinued.

The Sensirion has a “band-gap” temp sensor. The “band-gap” is a physical property of silicon and is very stable.

The majority of humidity sensors, including the Sensirion are capacitance based.

Thanks for that info, Niko. Very interesting.