I'm such a nerd

[kwende]

Some of you may find this interesting, some of you might find it somewhat overly nerdy; but I thought it was kind of interesting.

I'm a software developer by trade and have started playing around with microprocessors and the like and recently purchased an Arduino and a bunch of electronic parts to go along with it. So I thought I'd combine that with my interest in carnivorous plants and create a temperature/humidity sensor system to record for me the trends of the temp/humidity levels in my terrarium. I only let it run (thus far) over night, but I plan to let it continue running for the next week to get some good statistical sampling.

At any rate, for those who are interested in the climate of my terrarium, here is a graph of last night's events. I'm going to credit the momentary spikes in humidity levels to noise inherent to the sensor itself. Nonetheless, the trends are obvious: as the temperature decreases, the humidity increases (not linearly either, which is interesting to me). My hope is to try to match the curves I find in each terrarium to a particular environment so I can shuffle the plants into their appropriate terrariums (Aussie plants go into the Aussie tank, South Africans go into the South African tank). My guess is someplace like weather underground will give me historical stats for the environments I'm interested in.

Anyway..here's the graph I'm working with so far.


plot by kwende.rush1, on Flickr

For those who are interested, I can supply the source code and the electrical diagrams I'm working with.

 

[Fish n Plants]

That's pretty interesting! You could probably also program things like "thunderstorms" with the Arduino.

 

[kwende]

You're not too far off. I had thought of basically hooking up the Arduino to weather underground and controlling the "weather" of the terrarium to match what's happening in their native habitat. That's a bit of a pipedream, though.

 

[divaskid]

Nice! It'll be interesting to see the set of data over about a month. Should get rid of that spike youre talking about and possibly make it more linear.
Did you open up the terrarium at all during this time period?

Power to the nerds!

 

[kwende]

Yes, I think it'll be interesting to see how it behaves over a period of time.

And no, the terrarium was kept closed the entire time. Unfortunately I don't have an on-board clock so I can't correspond it to time of day; but I figure it'll still give me worthwhile information.

 

[fullcollapse]

I have been thinking of doing something similar. My boss set up something similar to remotely monitor his freezers. He has a web page that updates itself and it will send him a text message if the temp goes above a certain limit. He lost about $300 in beef, good Nebraska beef by the way, when his son left the door to the freezer open one summer morning. All the code is there, just need to take the time and do it. I think it would cool to build a couple and look to see if there are any micro climate differences between shelve on the grow rack.

 

[Lil Stinkpot]

You're not too far off. I had thought of basically hooking up the Arduino to weather underground and controlling the "weather" of the terrarium to match what's happening in their native habitat. That's a bit of a pipedream, though.

That would be VERY interesting!

 

[richjam1986]

Very cool

 

[kwende]

I'm from Nebraska, fullcolapse.

 

[sarracenia_X]

would you believe that my parents just got me an arduino board for my birthday(gonna try and build me a robot!:grin, and i was programming it for the first time, like literally to minutes ago (just making it blink and pulse a bunch of LEDs and that kind of basic stuff), i close the Arduino software, and decide to check Terraforums. the first thread that caught my eye was this, "im such a nerd". i clicked on it, and would you look at that, its all about an arduino based temp/humidity logging thing-a-majig for a CP terrarium! what a hilarious coincidence! mabey i could try using something similar if i ever started up a tropical CP terrarium!

 

[fullcollapse]

I'm from Nebraska, fullcolapse.

Yeah, I read in another thread you were from Lincoln. I knew you would understand what i was talking about. What are the time units on the x-axis of your graph? Is the temp jump when the lights came on? The drop in humidity was much quicker than the rise in temp. It will be interesting in the next few days to see if the inverse temp/humidity relationship is as dramatic, or if this is an outlier.

 

[kwende]

I came home this afternoon and realized the 9V power supply I was using died (it was simply a battery pack); so tomorrow I'll have to spend a few minutes and drag an actual cable into the terrarium (I was hoping the 9v battery pack would last longer). So...I will have to restart my tests. Just an FYI.

The x-axis are 5-minute intervals; in other words, every five minutes I appended a new data point to the file. I don't have an on-board clock so I cannot translate that to wall time without using offsets (and I'm not sure how accurate the timer is on the Arduino so I can't account for drift).

I believe the temp jump is the result of the lights coming on, but I also think it's the result of the temperature increasing in the house itself. We have the house running on a thermostat. All of the things you bring up: the change in temperature vs. humidity interests me and I have an interest to try to understand if I can come up with a reasonable intuition as to why these things happen. Hopefully after I fix the setup tomorrow I'll be able to run it indefinitely and, after a week, I should have some good data.

It's worth noting I'm still slightly puzzled by the occasional spikes in the humidity sensor. Like I said earlier, I'm blaming this on sensor noise. However, the fact it is not constant is a bit interesting.