CL System for Pressure calibration explained


I guess I was concerned that the instrument has issues such that it needs such large (3mb) calibration regularly. If you’re saying this is typical when a new sensor comes online and it calms down over time (i.e. learns) then I’m not as worried.



You have been online 5 days and that looks like your first and only calibration. It takes several days of data to fine tune the sensors.

I think you are fine and in time you will not see any big jumps.

1 Like


Hi @eric.guill25 . Awesome! This is actually a good thing…it’s the CL process doing it’s thing. Since your station was installed ~5days ago, our system has been collecting and analyzing comparison data in your immediate area. We don’t simply look at a couple point values and then make a blanket % adjustment … that isn’t how sensors perform. Instead, we evaluate numerous data sources over a period of time across a spectrum of conditions. Only when the data source are in adequate agreement by virtue of statistical analysis do we determine a linear calibration for pressure across that spectrum. Today was the first day that your comparison data was in adequate agreement to apply the initial calibration — in your case the initial calibration adjustment was slope = 1.0062, intercept = -3.6206 with a solid r-squared value of 1.0 on baseline data. As you witnessed in the data, your initial calibration was automatically applied on 1/11/19 at 9:11:46 am UTC.

It’s important to realize that the initial calibration is by far the most dramatic. Going forward, the CL system will continue to monitor your data against reference sources and apply small tweaks if and when warranted to keep your sensor always in tune. Think of it as guardrails. You likely will not notice most of these minor tweaks ahead.

Preemptive: To date, your station 6079 has received initial cal for pressure and humidity. UV cal has not been applied due to lack of suffient number of clear days, esp around solar noon. Rain cal is yet to be deployed beyond internal testing.

Keep checking in! Love your questions. Cheers.



Thanks for the quick response, really appreciate it and keep up the great work!



Is not WeatherFlow great?

David is one of the best company leaders I have had the privilege in dealings.



Since we’re talking about weird data stuff, might be a little off topic, but what would cause this data gap that I observed today? I’m missing data for 3 minutes in ALL fields from both my SKY and AIR. Even if my WiFi screwed up during that time, don’t the sensors store the data for eventual upload? Thanks, appreciate it!



Hey @eric.guill25 . It’s not you, it’s us. We had a rare but necessary API update earlier this evening that caused 3 data slots to drop. Some of you who had rain yesterday may have noticed a brief period where Yesterday’s Rain values displayed “—”. This was a hiccup in the API value summary.precip_accum_local_yesterday which was quickly recognized and resolved by our engineering team (during Friday happy hour!) . :beers:

1 Like

WeatherFlow PiConsole

Maybe, just Maybe @dsj you will consider a page where you will proactively post this information and action as it happens.



When you say Sea Level Pressure are you meaning SLP that uses temperature in its calculation or is it altimeter pressure as CWOP requires in its posting string? This formula:



The formula we use for SLP (and others) can be found on our Derived Metrics page



3 days after installing a replacement Air, Pressure CL applied an offset to perfectly align readings with trusted local sources… and now matches my 2nd Air too!



SLOman, were you having the issue I had, temperature linked pressure readings? It seems the new units have totally solved that issue.



Yes, and the replacement Air works great.



The formula you posted does not use temperature.



Gary, I think you are aware, but the problem some of us were having was that as a result of an apparent hardware issue, temperature changes were having a significant impact on pressure readings. Obviously a properly functioning Air shouldn’t do that, but some were. The newest units are now fine.

1 Like


That’s not what the poster is referring to.

The poster is enquiring about the formula used to convert station pressure to Sea Level pressure.



Gary, I was just addressing SLOman’s success with his new Air and inquiring if he had the same link between temp & pressure in the old unit.



I haven’t been part of that conversation. I know about it but have nothing to add.


pinned #40