Remote Temp/Humidity Monitor

[richyrich_one]

Hi

Just thought I would share my recent attempt at monitoring my observatory temp and humidity. It had to be remote via the web so I could check it any time. I had a Raspberry PI lying around doing nothing so...

http://www.instructables.com/id/Raspberry-Pi-Temperature-Humidity-Network-Monitor/

This project seemed to fit the bill so gave it a go. I wil stress this isn't my project, I just put it together following the instructions. I used a standard PSU rather than the POE adapter and kept the PI user rather than running everything as Root. With a few minor tweeks to the python script, not experienced enough for much else, it works very well. I have added the dew point calculation and display to the code and would really like to plot that on the graph too but can't make a lot of sense of the code to implement it. Anyone?

This is what I get on the webpage...

As you can see the occasional spike gets through somehow, not sure what's going on there. The quite sudden steps are from a slight coding error that crashed the python script that generates the graph, all sorted now. Not sure why the humidity is so spikey compared to the temperature. I'm guessing as the the observatory has quite a bit of ventilation, if there is a breeze it brings the humidity down quickly. It did seem to coincide with a windy period so could be it. I can now easily monitor what if any effect any further ventilation measures might have as well as keeping an eye on those potentially high summer temperatures.

The modified python script.

GraphTH.py

Hope this proves useful to someone.

Clear skies

 

[Tinker1947]

That's a nice system if your into tinkering with a soldering iron ect, for those that don't there are ready built units, Google will show quite a few this link is just one of many...

 

http://www.measurementsystems.co.uk/measurement-computing/mcc-data-loggers/wifi-502?gclid=CjwKEAjw55K4BRC53L6x9pyDzl4SJAD_21V1VyFkfgpAfp0xZDhr8i2GJJT__HO3nELmN_eGbCSGihoCGcrw_wcB

[richyrich_one]

Oh yeah, the other requirement was it had to be cheap

[Gina]

Very interesting - thank you for posting

[Tinker1947]

1 hour ago, richyrich_one said:

Oh yeah, the other requirement was it had to be cheap

I picked a dearer type so i wouldn't have to spend more time looking for one, but your system is likely to be as cheap as it comes......

[baggywrinkle]

A 0.5 Deg C variation in ambient temperature can show up as a 2% shift in Rh displayed.

Wind and drafts can make small pressure variations and this could possibly cause spiking. Rh is a ratio of actual water vapour pressure compared to the theoretical WVP that can exist under different temperatures, As both parts have a pressure element and temperature too variations in temperature and pressure can cause rapid changes.

What type of hygrometer are you using?

This too can have an influence. Some hygrometers are not checked by the manufacturers at anything other than a nominal 50%.

Based on the data calculated I can see the firmware is using a standard Sonntag formula for calculating Rh and Dew Point, this is very much the norm.

Interesting results.

[richyrich_one]

Thanks for that.

I should have mentioned, the sensor is a DHT22.

[baggywrinkle]

I guessed right, a polymer sensor. As water vapour is adsorbed and desorbed it causes a  small expansion and contraction in the polymer and alters the capacitance.

Accuracy +/-2-5%Rh which in terms of dew point is still very good. Ignore the 0-100% Rh range, 0% does not exist and 100% is liquid water which would kill the sensor.

Based on experience these are generally calibrated/checked 30, 50, and 70%. They tend to have a drift at low and high Rh values.

In case this all seems a bit nerdy it is what I do for a living...selling moisture measurement sensors and calibration systems.

 

[richyrich_one]

On 07/04/2016 at 14:00, baggywrinkle said:

I guessed right, a polymer sensor. As water vapour is adsorbed and desorbed it causes a  small expansion and contraction in the polymer and alters the capacitance.

Accuracy +/-2-5%Rh which in terms of dew point is still very good. Ignore the 0-100% Rh range, 0% does not exist and 100% is liquid water which would kill the sensor.

Based on experience these are generally calibrated/checked 30, 50, and 70%. They tend to have a drift at low and high Rh values.

In case this all seems a bit nerdy it is what I do for a living...selling moisture measurement sensors and calibration systems.

 

Interesting, thanks

[richyrich_one]

After a crash course in C and Python it looks like I have ironed out the anomalies and managed to include the dew point trace as well.

 

[richyrich_one]

Can anyone recommend a better quality sensor than the DHT22? I now have 2 running(inside and outside) and while the temperature sensing seems consistant and reliable the humidity side of things really isn't.

Quite often the outside one maxes out(99.9%) for long periods and the internal one has done so occasionally. The outside one is shielded from the weather so it doesn't get rained on. I gave them both a blast of a hairdryer which seemed to bring them back down to more realistic readings but I fear they are no longer accurate.

Any suggestions?