Humidity has never been recorded as Zero%

Device AR-00007537 is showing Humidity at 0% yet the rh_humidity sensor has not showing failed status.

@dsj is this in-fact a bug in the firmware or maybe a case where the firmware just didn’t see this as a failure?

This smells like the overflow bug. It is the calibration that catches the overflow but once it goes way over, it starts to fail. This is what I remember for my nasty air.

In any case the sensor seems to be malfunctioning. David will tell :sunglasses:


WeatherFlow is sending a replacement so one can infer that the sensor is indeed in a failure mode.

1 Like

It’s not a bug in the firmware since 0% RH is a valid reading. This sensor was not reporting a failure directly. Prolonged 0% RH is definitely suspect, of course, and is actually something that gets flagged by the auto QC system (which is part of the continuous learning system). For now that makes a great diagnostic tool and in the future we’ll be able to use it to proactively alert users of issues.

Then you have an issue in your javascript code that is used on the website.

    }, e.Observations.prototype._getDewPoint = function(e, t) {
        return 243.04 * (Math.log(e / 100) + 17.625 * t / (243.04 + t)) / (17.625 - Math.log(e / 100) - 17.625 * t / (243.04 + t))

If t = 40 and e = 0 the return is NaN

Math.log(0 / 100) is not a good value.


The web javascript code does handle the NaN return but it clearly should never happen.

Searching websites gives a lot of information on humidify and 0% but I am unable to find any trusted authority that states it can’t be 0%.

However every article I do find leads to the conclusion it has never happened and likely never will. The lowest calculated value, of all I am able to find, recorded is 0.33%.

Since a humidity value of 0 causes calculation errors in other formulas I think it’s safe to say, a reading of 0 is cause for concern.

How about Mother Nature’s division by zero error? I bet that the humidity is 0% near the event horizon…


In Sapce No One Can Hear You complain about humidity. And of course, no oxygen means no moisture.


Right. An RH of 0% means there is no water vapor present in the air. If there is no water, there is no temperature at which dew will form. That is, the dew point temperature is undefined when RH is 0.

Given the above (dew point is undefined when RH = 0), and given that NaN is what Javascript uses to represent undefined values, I’d say the formula works perfectly!

Yes, this is true. Like 100% RH, the odds of true 0% RH somewhere on the surface of the earth are extremely low. But it’s still a valid RH value.

But NaN it not error! See above. :slight_smile:

Definitely true. And, as I said above:


That is logical. Thank you for taking the time to answer.