Humidity Measurement

[Kiwi Bruv]

I have the rev ed retailed Honeywell sensor and it is giving me up to 122% Humidity readings. Power is 3 x AA cells, actually giving 4.92V.

Here are my calcs, I'm sure they are OK but at full 10 bit ADC (1023) it will allow the sensor to indicate 135% RH. Any one else have this problem?

VOut = Vsupply ((0.0062 x SensorRH) + 0.16 )

Therefore VOut/Vsupply = (0.0062 x SensorRH) + 0.16 = ADC/1023

So, ADC/1023 = (0.0062 x SensorRH) + 0.16, dividing thru by 0.0062

SensorRH = (ADC/1023) x (1/0.0062) - (0.16/0.0062)

Therefore SensorRH = (ADC x 0.1576) - 25.80646

Or for code for the PicAxe :

63xSensorRH = 10 x ADC - 1665

The sensor seems right, at midday it was showing 66% humidity, (local airport 30 Kms away near the sea was showing 73% on the net).

This increased to 98% at dusk, then this morning it was 122%. Weather was foggy, so was definitely 100% RH.

I'm sure the sensor is Ok so I might need to limit the 100% reading in software. Just wondered if there are any similar experiences.

Dowmload cable is not connected, so it's not that skewing the number.

 

[Jeremy Leach]

Hi

Haven't got time to chew it over at the mo, but I got slightly different equation in my weather station calcs:
64xSensorRH = 10 x ADC - 1652

See:<A href='http://home.btconnect.com/PicAxe_Projects/SensorCalculations.htm' Target=_Blank>External Web Link</a>

Not saying I'm right though.

 

[Kiwi Bruv]

Thanks, it's just different rounding to get the integers for the Picaxe. I rounded down, I think you rounded up. Anyway that difference doesn't explain why my sensor doesn't stop at 100%!!!

The sensor seems to be spot on when the air is not at the dew point. This afternoon when I got home it matched the local airport reading exactly - 76.0%.

Does yours indicate &gt; 100%?

 

[Technical]

The senosr datasheet does state slightly different graphs at different temperatures.

Therefore for a completely accurate system you should adjust the calibration figures according to temperature (and to supply voltage). The numbers you are using give a 'best fit' figure for general use but will never be exactly 100% accurate at varying temperatures.

It could also be that the sensor saturates at 100% humidity.

http://www.rev-ed.co.uk/docs/sen008.pdf

 

[Jeremy Leach]

I take account of temperature when I download my data to my PC and print out the graphs. Yes, I've often got over 100%, probably not much over 110% though, even taking account of temperature, which is measured by a DS18B20 right next to the humidity sensor.

I just put it down to Humidity being a tricky thing for a sensor to measure accurately - although maybe this is being too lenient ?!

 

[Jeremy Leach]

Also, another thought ...not sure if this is correct : Fog is beyond saturation level so wouldn't you expect over 100% ?

 

[Kiwi Bruv]

Techncally, saturation is the limit. Below this the water is water vapour, above it at the dew point the excess water condenses out and exists as liquid. So more than 100% humidity is not possible at normal temperatures and pressures.

I think I am seeing the reaction of the sensor to saturation. But the actual humidity (Kg water/Kg air expressed as a %)can't be more than 100%.