> I have never seen a pressure sensor that far off. > I suspect confusion between reduced and station pressures. > But maybe you just have bad hardware.
But I see them right now :) One shows 969,7 hPa, second 979,8 hPa. Sorry, it's not my fault that I see such values on the displays. And I know what ABS and REL are, really. > You said 10 earlier which is not 6. It seems you didn't understand what I wrote: "I have two (WH25A and WN32P), and each shows a different value. Their readings differ by exactly 10 hPa". WH25A shows 969,7 hPa, WN32P shows 979,8 hPa. So it's 10 hPa. Correction +6 hPa is needed for WH25A which I use with my station. For WN32P correction is -4 hPa. > I suggest reading the source code. Also, in weewx.conf, services are > listed in order, and the standard approach is to do StdCalibrate before > StxWXCalculate. So *if* the console reports station pressure, and *if* > barometer is software instead of harder, then just adjusting station > pressure should be sufficient, because StdWXCalculate should see the > adjusted station pressure. But if the console is *also* reporting > barometric pressure, then StdWXCalculate will not be run for barometric > pressure. And that's the answer that's slowly leading me to some kind of conclusion. So should I set correction for pressure (ABS) only *and* pressure = prefer_hardware to get pressure reported by console corrected by +6 hPa? if yes, what to do next to have barometer (REL) also corrected, ie. near to METAR value? > You should understand what is going on with what's reported and what's > calculated. > > If you don't want to try to understand that, then it's going to be much > harder for you to get a good outcome. That's what I'm trying to do from the very beginning :) As I said earlier, I take care of several stations and trying to find some general way to correct pressure using only weewx.conf, without physical access to hardware/consoles. wtorek, 7 lipca 2026 o 18:55:31 UTC+2 Vince Skahan napisał(a): > Since you are using ecowitt you can apply your correction byusing the > wsview app if you have that. I have had to slightly adjust some ecowitt > temperature sensors vs. my multiple other reference sensors. > > On Tuesday, July 7, 2026 at 9:14:39 AM UTC-7 Greg Troxel wrote: > >> "'Tomasz Lewicki' via weewx-user" <[email protected]> writes: >> >> > Thank you for your reply and for your willingness to help. I've been a >> > satisfied QGIS user for several years now; I manage a few stations, so >> I >> > have a pretty good idea of how things work. I deliberately didn't >> provide >> > details about the station model, console, etc., because I'd like the >> > solution to be hardware agnostic. I'm familiar with the Wiki article, >> but >> > unfortunately, it doesn't explain what I'm asking about. Or maybe I >> just >> > can't read between the lines :) >> >> I don't think it's reasonable to leave out relevant information on >> purpose, and if you are asking a theoretical question that should be >> separate. >> >> > As for the barometer readings, I have two (WH25A and WN32P), and each >> shows >> > a different value. Their readings differ by exactly 10 hPa. I have a >> Python >> >> Something seems very wrong. I have never seen a pressure sensor that >> far off. I suspect confusion between reduced and station pressures. >> But maybe you just have bad hardware. >> >> > script that retrieves the current METAR for a nearby airport, and based >> on >> > the specified altitude above sea level and the entered pressure, it >> > calculates the "correct" pressure and reports the offset of my >> barometer >> > from the airport's. >> >> It would be nice to publish that, so that you contribute to the commons. >> >> > It saves the results to a file and calculates the mean >> > and median from several measurements. That's why I wrote that I'm >> > interested in a pressure correction of +6 hPa, because that's how much >> the >> > WH25A sensor underreports (I borrowed a WN32P for verification only). >> >> You said 10 earlier which is not 6. So please dig into your hardware >> and debug logs and examine and then describe >> - what the documentation (+ weewx wiki, internet) says it measures and >> reports >> - what kind of calibration can be done on the hardware >> - what values you have set for those calibrations >> - how you determined your elevation and what your estimate of alitude >> uncertainty is >> >> - specific values reported by your hardware, your elevation, and >> corresponding values/elevation from nearby official stations >> >> > I'd simply like to understand how the correction calculation workflow >> > works, as specified in the [StdCalibrate] / [[Corrections]] section of >> > weewx.conf. At what point is this correction applied, and to which >> > "measurement" method—only software? Only hardware? Both? >> >> I suggest reading the source code. Also, in weewx.conf, services are >> listed in order, and the standard approach is to do StdCalibrate before >> StxWXCalculate. So *if* the console reports station pressure, and *if* >> barometer is software instead of harder, then just adjusting station >> pressure should be sufficient, because StdWXCalculate should see the >> adjusted station pressure. But if the console is *also* reporting >> barometric pressure, then StdWXCalculate will not be run for barometric >> pressure. >> >> You really need to understand what is being reported vs calculated, and >> how weewx is choosing to calculate or use hardware for each item. >> >> > My reasoning is as follows: >> > >> > 1. I set the pressure correction to pressure = pressure + 6.0 in >> > weewx.conf; I don't change anything in the station console—no offsets, >> no >> > entering "correct" ABS or REL pressure >> > >> > 2. After setting the pressure correction, should I set pressure = >> > prefer_hardware or software? >> >> You should understand what is going on with what's reported and what's >> calculated. >> >> > 3. What should I do with barometer then, to avoid applying a double >> > correction or causing a calculation error? >> >> You should understand what is going on with what's reported and what's >> calculated. >> >> If you don't want to try to understand that, then it's going to be much >> <harder for you to get a good outcome. >> > -- You received this message because you are subscribed to the Google Groups "weewx-user" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion visit https://groups.google.com/d/msgid/weewx-user/a8e7918d-4306-41f7-9dcf-d54683aa5b07n%40googlegroups.com.
