dolezsa / thermal_comfort

Thermal Comfort sensor for HA (absolute humidity, heat index, dew point, thermal perception)
Other
590 stars 109 forks source link

Not calculating right #217

Closed Traxeronus closed 1 year ago

Traxeronus commented 1 year ago

Hi,

I see Simmer values exactly same as input sensor values.

Screenshot 2022-11-11 183954

Should be 22.7 for floor 1 and 10.8 for floor 2.

matze338 commented 1 year ago

I recognaice exactly the same issue :/

vaandefanel commented 1 year ago

Same issue too.

rautesamtr commented 1 year ago

Currently ssi is not really calculated if bellow 70°F since it's meant for higher temperatures.

https://github.com/dolezsa/thermal_comfort/blob/90a797669a63abc49a0b46bb821b9eba9ee91772/custom_components/thermal_comfort/sensor.py#L869-L884

Not my code but it's a good idea to discuss how to handle such situations in this issue.

Traxeronus commented 1 year ago

Hi, I'm bit lost why? 70F is standard room temperature and when I try to calculate it using some of ssi calculators (f.e. https://www.vcalc.com/wiki/rklarsen/Summer+Simmer+Index), it shows me the right value. There should be no temp limitation. Or am I mistaking? M.

rautesamtr commented 1 year ago

@lymanepp do you remember why you added the temp limit?

lymanepp commented 1 year ago

@lymanepp do you remember why you added the temp limit?

The limit should be 58° F degrees instead of 70° F. That was an error in my implementation. I'll submit a PR to fix it.

Traxeronus commented 1 year ago

Please, can you explain why the limit at all? The calculation of SSI has no limit and can count with lower values. Thanks

lymanepp commented 1 year ago

Please, can you explain why the limit at all? The calculation of SSI has no limit and can count with lower values. Thanks

SSI is a form of heat index and was not designed for lower temperatures. The SSI author only presents results for temperatures above 70° F. But my implementation from around 15 years ago used 58° F as the cutoff value. I think I changed it to 70° F when implementing in HA to align with the author's descriptions of temperature ranges (slightly cool, etc).

@dolezsa I'll leave it to you to decide what (if any) lower bound to use here.

lymanepp commented 1 year ago

I changed my local copy to use 58° F as the cutoff value and it looks much better as the SSI and air temperature converge together at that temperature. No more big jump at 70° F.

lymanepp commented 1 year ago

This issue was changed and integrated a while again and can be closed.