Backyard Aquaponics
A place to discuss aquaponics

Forgot to say...
On the subject of IR, it's a bloody nightmare doing anything in daylight or with fluids around.
I have done a lot of work with IR sensors, one thing I noticed while developing an IR Lap Timer was that plant life reflects huge amounts of random IR noise.
All in all making it pretty much a no go in a GB, daylight, water and plants...

Hmmm, the only other thing i can suggest is and electrical float switch? when the float is high enough it pushes a switch in or similar?

Which MicroSD shield are you using, and how did you choose it? I've never done anything with one but didn't know they were only 10 bucks. At $10 I already need a few

Hmmmm, I'll have to remember that then. the thought if a level indicator did cross my mind at one point, but thought i would focus on the datalogger or before i know it i will be trying to do everything at once.
I prefer to do this sort of stuff in pieces, first log the data...then try it with an SD card, Next add an LCD, Next add a few more sensors.

I considered that too, but mechanical stuff always breaks.
Thailand is also very humid, leading to problems with electrical contacts.
Floating mercury switch would be perfect, until the glass got broken
I'll dig the stuff out of the box and start playing again, it's been about 9 months I think...

This is what i bought:
http://www.ebay.com.au/itm/MicroSD-SD-Shield-Arduino-UNO-Duemilanove-/250900735563?pt=LH_DefaultDomain_0&hash=item3a6ad9664b#ht_500wt_1413

Where it comes from/what it looks like (mines blue aswell):
http://www.ekitszone.com/Products/9-microsd-shield-for-arduino.aspx

Its pretty much an exact copy of this :
http://www.sparkfun.com/products/9802

Its pretty much fine the way it is. it has a 5V-3.3V digital logic convertor and a 3.3V power supply for the MicroSD and any 3.3V sensors. But I plan on soldering an RTC and RTC coin battery to it in the space provided, but wont solder any sensors because i may want to use it to sense something else some day

All good points. If you come up with something let us know

Of all the ideas, thoughts, and inventions I've played around with, I think I've spent the most time pondering the chicken door

I for one would love to read about it, although sadly we no longer can keep chickens so I cant see myself making one.

I even gave some thought to a fox trap as part of a few designs

-single chicken sized revolving door
-narrow tunnel u-turns too tight for a fox to navigate
-single chicken door with guillotine fox remover/second chicken harvester
-small solar fountain pump scales of justice door lock that stopped pumping water to the slow leaking balance after dark (and also kept their water moving)
-sprung perch that dropped a pin into a catflap to lock it when at least 3/4 of the chickens were on the perch by weight (my chickens were always all home for 1/2 an hour before any of them got on their perch)

and many others I cant remember

But now that I'm learning electronics, I might have to revisit it, because you can never have too many chicken door designs.

LOL

Well, thats a small project for my mums place. chickens kept getting eaten because people kept forgetting to close the door at night.
So simple solution with another photoresistor pressure switch and an XOR digital logic gate that can be found here...
http://australia.rs-online.com/web/p/gates/0442993/

XOR truth table looks like this...

light open/close | Out
0 0 | 0
1 0 | 1
0 1 | 1
1 1 | 0

So using another voltage divider circuit with a photoresistor and a calculated resistor to match.

If its dark, V over photoresistor will be high leaving voltage over the other resistor low (so = 0)
if its dark door will be closed opening the switch so no current and also = 0
both = 0 so no output

as sun comes up and resistance over photoresistor decreases enough, and that voltage then goes over that goes down, while the voltage over the constant resistor goes up, bringing it up to HIGH or a 1.
but the door is still closed and switch open so =0
Now the output will be 1/HIGH/5V this will go to a transistor controlling a 12V actuator that will open the door until the switch closes giving a HIGH reading. where it sits until the sun does down repeating the cycle.

I haven't chosen the resistor yet because i dont know if it will be on a separate circuit to the 12v battery or not. But i'm sure it will be now, so ill calculate what resistance ill need and post my results here.

This should hopefully work, Otherwise i have just invested in 5 attiny85 microcontrollers that should do the job is it doesn't work.

Haha, your can never have too many designs for anything

Just remembered, I didn't calculate the resistor because i didn't know at what amount of light mum wanted it to switch at, And i like my sleep too much to get up at the crack of dawn with a multimeter waiting for the chooks to decide whether it was the right amount of light to go outside or not.

Ok, I have been busy. Just finished coding it to read out a number in milli seconds from a file called commands.txt and to use that in the delay function to pause between reading the sensors. The code is a bit haggard but ill try clean it upp and comment it later.

http://dl.dropbox.com/u/19981222/MicroSD%20Datalogger/greenhouseSDReadWrite.pde

For some reason when it prints wait after the setup function its comming out about 10X what it have in the command.txt file, Don't know why yet but its not hard to divide by 10 just to get it to what it needs to be and i can't be bothered going through it at the moment, but if anyone notices anything let me know.
cheers

changed my mind, had another look and did some calculations by hand and on line 57 the "1" should have been a "2" so it all works perfectly now, will probably add more commands in the future, i just don't know of any needed now?

Good idea, switching the power one only when sampling would help, even better to do say 4-5 samples right after each other (switching off the power each time and take an average) Yes the high impedance of the water will be an issue. I would use as high a voltage as possible, say 12v at least (switched by a transistor). The higher the voltage the more noise immunity you will have. Then use a voltage divider to go into the ADC. Also important to add some noise filtering as close as possible to the ADC input.
Use a basic RC low pass filter, which is just a cap between the ADC input and ground, and a series resistor right before the cap. Should check values but something like a 100n cap and 50-100ohm resistor should be ok, just make sure its not too slow.