Wrong Sea Pressure calculated

[valdus2]

I use a BME 280 sensor and accurately displaying 1018.6 hpa at 9m high with 20.5C.

I have also declared in Tasmota the correct latitude, so it can calculate the constant of gravity.

Tasmota shows 1020.6 Sea Pressure when it should be ~1019.6.

I used the following calculators and they all agree:

http://stacjepogody.waw.pl/kalkulator_cisnien.php

http://www.meteo.tomkii.net/pressure_calc/?alt=9&latitude=38.18&abs=1018.5&temp=20.5&x=10&y=11

https://www.mide.com/air-pressure-at-altitude-calculator

Thank you in advance.

[s-hadinger]

I think the code is here:

Tasmota/tasmota/tasmota_support/support.ino

Lines 956 to 974 in 7ef93b2

	float CalcTempHumToAbsHum(float t, float h) {
	if (isnan(t) || isnan(h)) { return NAN; }
	// taken from https://carnotcycle.wordpress.com/2012/08/04/how-to-convert-relative-humidity-to-absolute-humidity/
	// precision is about 0.1°C in range -30 to 35°C
	// August-Roche-Magnus 6.1094 exp(17.625 x T)/(T + 243.04)
	// Buck (1981) 6.1121 exp(17.502 x T)/(T + 240.97)
	// reference https://www.eas.ualberta.ca/jdwilson/EAS372_13/Vomel_CIRES_satvpformulae.html
	if (Settings->flag.temperature_conversion) { // SetOption8 - Switch between Celsius or Fahrenheit
	t = (t - 32) / 1.8f; // Celsius
	}
	float temp = FastPrecisePow(2.718281828f, (17.67f * t) / (t + 243.5f));
	const float mw = 18.01534f; // Molar mass of water g/mol
	const float r = 8.31447215f; // Universal gas constant J/mol/K
	// return (6.112 * temp * h * 2.1674) / (273.15 + t); // Simplified version
	return (6.112f * temp * h * mw) / ((273.15f + t) * r); // Long version
	}

I suspect this might be due to the imprecision of FastPrecisePow.

Could you try to compile with powf instead? It will grow the firmware size significantly, but will confirm if that's the cause of the issue. If so, we'll try to find an alternative more precise power function.

  float temp = FastPrecisePow(2.718281828f, (17.67f * t) / (t + 243.5f));

[arendst]

On the other hand, I have several pressure sensors around the house showing deviations of many hPa so I wouldn't mind a difference of 1 hPa as long as I do not plan to use it as the main height measurement in a plane.

[valdus2]

@s-hadinger You mean, changing this:

float temp = FastPrecisePow(2.718281828f, (17.67f * t) / (t + 243.5f));

to this:
float temp = FastPrecisePowf(2.718281828f, (17.67f * t) / (t + 243.5f));

@arendst I'm not talking about deviations of hPa, which of course happens between different sensors,
I'm talking strictly about the conversion inside Tasmota of Pressure to Sea Pressure.
(Whatever hpa to use the coversion-equation seams wrong)

Thank you.

[s-hadinger]

No. I mean powf()

[valdus2]

Maybe a little more help please, cause I'm not a C programmer...
What to change exactly?

[sfromis]

powf instead of FastPrecisePow

But the code bit quoted by @s-hadinger was the wrong one, the sea pressure calculation is instead in
https://github.com/arendst/Tasmota/blob/development/tasmota/tasmota_support/support.ino#L1010-L1015

float ConvertPressureForSeaLevel(float pressure) {
  if (pressure == 0.0f) {
    return pressure;
  }
  return ConvertPressure((pressure / FastPrecisePowf(1.0f - ((float)Settings->altitude / 44330.0f), 5.255f)) - 21.6f);
}

[sfromis]

BTW. Those sea pressure calculation can only be an estimate, as local meteorological conditions can vary. And the formula already does not use things like temperature and humidity.

[valdus2]

does not use things like temperature and humidity

Ah, I see.....Nevertheless, I'm going to compile with powf and if any difference, I'll report it here.

So in your bit of code I have to change FastPrecisePowf to powf ?

Thank you.