Ever since I got my Davis. I have loved it… worth every penny.
EXCEPT - my barometer always seems to wander off the normal. I had to re-calibrate it every couple of weeks. Here I am with a $1,000 weather setup and it doesn’t keep barometric readings as well as my old $100 WS-2300. Very very frustrating!
Well, I was looking through the wxqc archives the other day and found someone else with the exact same problem. And even more amazingly, the solution from Steve with SoftWx.
It seems that if you fill out your altitude in the Davis console, then the unit does not use just the raw barometric reading + your offset. It wants to use temperature and humidity to help figure the amount.
If you just reset your altitude on the console to 0 and keep your offset, the barometer quits wandering from the correct amount and actually works correctly! See Steve’s response below.
In the past 3 days since I “fixed” my altitude, my barometric reading has been off -.1 mb from what MADIS expects it to be. A lot better than the old days where I sometimes was off the entire day by 5-10 mbs. CW2433 readings
To reset the altitude, you select DONE and down arrow (-) at the same time. Then you scroll through options with the DONE.
To exit, you press and hold DONE.
To set your offset, you select BAROMETER then you select 2nd and press and hold the Set (Alarm) button until the barometic reading stops flashing. Then you change the numbers with your arrow keys. Select DONE when you have your changes complete. (I’m very familiar with this procedure, having done it dozens of times in the past year).
Every piece of equipment seems to have its own quirks… I have a WS2308 (2310 series) and yes it was for under US$100 when I purchased it on a special, but it also has is problems of high wind gust spikes especially if you have the external power supply plugged in (I still get about 1 a month even on batteries) and the humidity limit is 96%, and the console hardware is prone to being flaky (one serial com port out failure, replaced unit, now has LCD display has part failure…)
BUT - it has got me off the ground and running with my new hobby, one I have been interested in as a 7yo or so on our windswept farm when dad always tapped his index finger on the barmoeter in the hallway every morning… the weather ruled his life with sheep and crops and cattle…
And I now belong to a great forum of helpers - well done Tom for sharing the hints found - its just like your templates and Ken’s (Ktrue) AJAX versions and everyone else who has a tip or hint or software utility or code that has been developed specifically for helping folks display their data on the internet…
After that thread whose link you posted, I got the code worked out, and Brian incorporated it into WeatherDisplay. So you can leave your VP console’s elevation set to your real elevation, and WeatherDisplay will convert the SLP from the console into a true altimeter value just prior to sending it to APRS/CWOP. It is an option on the CWOP setup page of WD.
It was great that Brian made that change, making WD one of the first weather programs to have that feature. Some time after that, WeatherLink finally added the altimeter/CWOP feature.
Leaving the console configured with the true elevation, and configuring your software that does the CWOP submission to calculate and send altimeter is the best solution.
Regarding the elevation of the sensor…
It’s best to use the best elevation information you can find, but it’s not critical. If the entered elevation is within around 30 ft. the elevation difference from true elevation is compensated when you calibrate to a nearby station. Note that ASOS weather stations at airports have two pressure values: station pressure which uses the altitude of the highest point on any of the runway surfaces, and sensor pressure which uses the altitude of the actual barometric instrument. Station pressure is calculated from sensor pressure, and temperature is part of the formula the NWS uses. But over the normal range of temperatures you’d likely see, the delta between the sensor and the station pressures remains essentially the same if the elevation difference is 30ft or less, which means it can be covered by a fixed calibration offset. You can read more about the details of these formulas from a programming point of view here: http://www.softwx.com/weather/uwxutils.html
Setting elev kind of works. What you’re doing is setting your elevation to sea level, which means your sensor outputs the raw pressure, not adjusted for anything, and then calibrating a big offset to approximate the altimeter value. That offset represents the difference between raw pressure and the altimeter (which is raw pressure corrected for elevation). The problem is that the difference between raw pressure and altimeter is not the same over the range of normal pressures you’d see. If you do the 0 elevation thing your console will show “pseudo altimeter”. The “real” altimeter value is not calculated by adding a fixed offset to the raw pressure. It is calculated via a standard formula that is more complicated than that. So when the pressure is what it happened to be the day you calibrated, you will be spot on compared to the nearby NWS station. But as the pressure gets further and further from that, you will track increasingly off. But the difference will not be as great, nor as noticeable as the SLP issue. How much it could get off is dependant on your actual elevation. The higher you are, the more you would notice it.
Click the link and scroll down to barometer… I made the change in mid/late October… notice how the deviation went from all over the place to right in line?
I also have the wandering barometer (7500 ft), but for me the wind seems to be the biggest problem with my quality control. I haven’t done a formal comparison, but I’ve found I can usually predict when my readings will be “off”. I assume it is a compression thing due to the mountains? I slightly changed the offset once, and that gave me a better “average”, but I’ve seen no reason to do that again. I will have a few “off” days when there are fluctuations, but then things settle and stay fairly close to the “predicted”. Also, the number of the fluctuations will vary with the day. Sometimes there are quite a few, other times it is a few several hours apart. That is another reason I think the wind might have something to do with it.
Other than my cwop quality checks, what would I gain by setting my elevation to 0. Wouldn’t that make my “adjusted” reading (meaning what WD shows in the graph) off?
Looking at my recent data I guess I can’t totally blame it on the wind. Some windy days are fine. I guess the other calculations are also playing a part. On the other hand, my data (so far) always comes back and tracks close to predicted so I haven’t worried further about offsets. Unfortunately there are no close stations to compare me with so I have to assume the checks and thumbs up are as good as it’s going to get.
Thinking further, if I was to set the console at 0, how would that affect WD and the rest of my data? Would I need to alter other things in WD or would my WD barometer graph just be “wrong” or?
