siddacious
commented
5 years ago @dachshund-digital I understand that you may be frustrated but none the less please phrase your request in a manner that is less accusatory and more constructive.
Open dachshund-digital opened 5 years ago
siddacious
commented
5 years ago @dachshund-digital I understand that you may be frustrated but none the less please phrase your request in a manner that is less accusatory and more constructive.
dachshund-digital
commented
5 years ago To your point about being constructive, valid point. Thus, below is the formula for proper calculation of sea level pressure, for a given elevation, temperature, and pressure at said elevation location. This should save some time and effort when the code is revised appropriately to address the issue at hand.
ladyada
commented
5 years ago @dachshund-digital thanks, please submit a PR and we'll check it out!
barbudor
commented
5 years ago While I was in the others BME280 issues, I started to have a look at this one.
What I understand so far is that there are mainly 2 main formulas to link pressure and altitude: The "full" version is the one given in the above link and the "Barometric" which works fine for altitude below 11,000 meters. The formula currently used in the library is the "Barometric" formula with an additional approximation using a static temperature at 15°C (seems to be a common approximation) I trust 11,000m is enough for most of the applications of Adafruit products, with the exception of stratospheric balloon. This limitation could probably be written somewhere. 2 points seems to remains from my view:
1) Should the formula be changed to offer the ability to take temperature into account ? It seems temperature has a measurable impact on the 44330 constant used in the formula (44300 = (15°C + 273.15)/0.0065 where 15°C is converted in Kelvin and 0.0065 is a constant). What I understand is the temperature should be the outdoor temperature, which is not necessarily the temperature measured by the BME280 if set indoor. So probably not a good idea to directly use BME's temperature.
https://en.wikipedia.org/wiki/Barometric_formula
2) As of today the library is using a fixed value for sea-level pressure (101,325 Pa). As we know, this may vary a lot between days. I feel the major improvement to the lib should be to offer a way for the user to provide current-day calibration by proving a reference altitude and pressure. Most altimeters, either in avionics or even used in sports like mountaineering offer a way to set a reference point.
The Arduino library have such a function :
float Adafruit_BME280::seaLevelForAltitude(float altitude, float atmospheric)
I believe this would be worth a dedicated issue.
Happy to help is you want. If anyone with knowledge in thermodynamics and physics could jump in, this would be appreciated... let me know.
ladyada
commented
5 years ago hiya, for Q2, you have to set sea_level_pressure that's a property - the default is just the default. so if you want to make an example for that, we'd love it.
for Q1, we use the calculations from the BME datasheet - which takes temp into account, the self-heating isn't too intense, maybe a few degrees
The current Adafruit code for ESP32, ESP8266 BME280 libraries/APIs support correct calculation for sea level pressure by accepting an elevation value and pressure value. And that calculated pressure at sea level is then used for altitude result.
This library only supports a crude static value for pressure at sea level. This is functionally incorrect, and should be address. For example ESP32 and ESP8266 library by Adafruit correctly reports sea level pressure at an elevation of 41 meters for my location. But this library reports the altitude for my location as 68 meters.