PCB humidity measurement circuit design

[strawberry]

how to design spaced PCB tracks to measure air humidity with +/-10% accuracy
something that could work atleast with one dual OP

[GigaJoe]

I would consider it as a joke ,
for a very simple reason ...  any analog humidity sensors have a special compound that change resistance absorbing water molecules.

from your question I would assume that epoxy foundation of PCB significantly change resistance , ( or capacitance ,  or some other  electrical parameters ...)
And that would be real fun ....  to see electronics behavior in real life ...  after rain outside ... for example ...

...   take a real sensor intead ..
https://www.newark.com/multicomp-pro/mp007422/humidity-sensor-w-o-case-90rh/dp/45AJ6216

[strawberry]

5X5 mm PCB sensor area and feature size 0.1mm
pure water = 1.8E9 /m
A=12.5^2mm
l=0.1mm
R=~8E9 (100% humidity)
resistivity measurement will be hard
maybe capacitative

are commercial sensors reliable(>100kh)? some say, need to dry them to make them operable again

[DaJMasta]

Are you actually measuring humidity, or are you measuring wetness?  Humidity (in the air, noncondensing) can be reliably measured for years by a variety of commercial sensors, but it's not the same as detecting the presence of water on a surface (condensing, liquid level, etc.)  PCB traces in parallel but not touching can be used to detect water droplets (easy detection does rely on water contaminants, but even trace amounts brings down the resistivity very dramatically), and they can be used to detect water concentration roughly in a medium that soaks it up (like soil.)  You can capacitively detect the presence of water or liquid level too, but is again a different sensor design and consideration.  To some extent, you probably can measure ambient humidity with open traces and some amount of contamination and you can probably get extra information from capacitive sensors, but those are going to be precision measurement tasks that will probably hinge on tightly controlling and characterizing other variables that will interfere with the measurement, since the meaningful measurement values will be much smaller.

The short of it: different sensor designs depending on whether you want to measure air humidity, soil humidity, or the presence of water.

[strawberry]

to measure PCB condensation level . dont know much about this type of sensor
how to damage commercial humidity sensor
 Water condensing:
If there is a condensation on the humidity sensor surface over 6 hrs, the polymer of the sensor
(humidity sensitive) membrane can be diluted by the condensation.
 Drenching (solvent .. etc.):
If the humidity sensor is soaked in the liquid material (solvent .. etc) over 30 min, the polymer of the
sensor (humidity sensitive) membrane can be melt by the solvent.
 Ionic atmosphere:
SYH series humidity sensor is made by using ionic polymer membrane. Therefore if the humidity
sensor is exposed ionic atmosphere (salty air, anionic ionizer.. etc) for long period of time
 Organic / Inorganic gas:
If the humidity sensor is exposed directly to organic / inorganic gases (i.e. SOx, NOx, Ammonia,
Alcohol, Glycol .. etc.) for long period of time, reactive polymer membrane can be damaged

[jonpaul]

$25 contractors humidity probe for wood, driwall

j

[zapta]

Is this an exercise in building a humidity meter without buying an off the shelf humidity sensor?

If not, there are many humidity sensors on the market with analog and digital interfaces. For example , this one with analog outputs.  https://sensirion.com/media/documents/74DC2069/655379E3/Sensirion_Datasheet_SHT4xI-analog.pdf

[CatalinaWOW]

Your answers imply that your manufacturing environment or operating environment will ruin commercial humidity sensors.  Quite possibly right.  But it is probably easier to change/modify the manufacturing environment than to design a "good" humidity/moisture sensor.  The simplest modification is a cap or seal over the sensor that is removed on delivery.  If it is the operating environment your problem is tougher. 

Inferring wetness or humidity by measuring conductivity is almost impossible in the kind of environment that kills humidity sensors.  Ionic contamination can change conductivity over several orders of magnitude and there will be contamination in such environments.   50% error bars are likely.  Capacitance measurements imply that you are measuring the change in dielectric constant of the substrate (or possibly air) with humidity.  The change in air is small, hard to measure.  The change in the board may be large, but if it is I suspect that the same kind of changes from solvents/contamination will occur.

One older method is to measure dewpoint by chilling a mirror (thermoelectric cooler) and detecting the change in reflectivity when moisture condenses on the mirror.  With ambient temperature and dewpoint, humidity is a table lookup.   But in the kind of environment you postulate some kind of mirror cleaner may be necessary.  If the solvents you fear are present in any concentration they may condense out and interfere with this method also.

These humidity sensors can sometimes be revived by heating in a dry environment.  Perhaps a series of sensors with a method to isolate them from the environment while recharging would be feasible.