linusakesson.net
Hardware projects

The pot plant monitor

This pot plant is connected to my computer. Every five minutes or so, the resistance of the soil is measured. The resistance varies indirectly with the dampness of the soil, as you can see in the graph.

Graph

I water the plant every few days, as indicated. It would be fun to extend this into a full-blown irrigation system by building some kind of motorized watering contraption, but I don't want to risk the wooden floor. =)

How the sensor works

This is the world's cheapest ohm-meter:

Schematic

The microcontroller drains the capacitor by providing a 0V output on the pin. It then re-configures the pin into a high-impedance input pin and resets an internal clock. Now the capacitor is slowly charged through the resistor (soil), and when the voltage level has become so high that it corresponds to a logical 1, the microcontroller receives a pin change interrupt, reads the clock and reports this value to the computer.

Posted Wednesday 9-Jan-2008 17:20

Discuss this page

Disclaimer: I am not responsible for what people (other than myself) write in the forums. Please report any abuse, such as insults, slander, spam and illegal material, and I will take appropriate actions. Don't feed the trolls.

Jag tar inget ansvar för det som skrivs i forumet, förutom mina egna inlägg. Vänligen rapportera alla inlägg som bryter mot reglerna, så ska jag se vad jag kan göra. Som regelbrott räknas till exempel förolämpningar, förtal, spam och olagligt material. Mata inte trålarna.

Anonymous
Mon 31-Mar-2008 14:43
While I quite like the idea of the plant monitor, I wonder what the effects of the sensor will be on the system.
Current flows. Electrolysis will happen. The electrodes will dissolve over time. There's a magnetic field, etc.
Anonymous
Thu 3-Apr-2008 09:40
At University of Gent (molecular genetics department), we weigh the plants to determine if they need water. This way we can water them very precisely and there's no milliliter of water given to much or to little.

-femeref
Anonymous
Wed 9-Apr-2008 15:47
One interesting thing missing in the schematics is a suitable value for the capacitor.
Anonymous
Wed 9-Apr-2008 15:56
The amount of current would be very small. And the time it flows can be minimized by providing 5V on the sensing side when not measuring resistance, in effect giving very ner zero volts of difference between the electrodes.

Electrolysis would happen to a very small extent, but it can be minimized in the same way as the current.

Time to put that gold jewelry to use! It won't dissolve if used as electrodes.

The magnetic field would probably not be strong enough to be measurable with affordable instruments, given the small current.
Anonymous
Thu 10-Apr-2008 11:42
Plants tend to grow, thus gain weight. How do you calculate the amount of plant growth?
-webpresence04 at soulsurf se

At University of Gent (molecular genetics department), we weigh the plants to determine if they need water. This way we can water them very precisely and there's no milliliter of water given to much or to little.

-femeref
lft
Linus Åkesson
Thu 10-Apr-2008 20:25
One interesting thing missing in the schematics is a suitable value for the capacitor.

Good point. I'm using a 10 uF electrolytic capacitor.
Anonymous
Mon 14-Apr-2008 21:35

lft wrote:

10 uF electrolytic capacitor

Thanks!
Anonymous
Tue 22-Jul-2008 11:51
The amount of current would be very small. And the time it flows can be minimized by providing 5V on the sensing side when not measuring resistance, in effect giving very ner zero volts of difference between the electrodes.

Don't even need to do that, as the capacitor will charge up until there is close to zero volts across the resistor (in this case, the soil).
Anonymous
Sat 9-Aug-2008 09:49
Now all you need is a low power RF transmitter so that you can monitor all flowers with great convenience.
Anonymous
Sun 17-Aug-2008 21:33
Very nice idea!
I have got two suggestions on how to improve it a little bit:
1. There is no need for a high frequency of measurements and therefore also no need to always provide +5 V. Just connect this path to another uC pin and while no measurement is in progress, let both uC pins provide GND level.
2. Put a resistor (suggestion: 220 Ohms) in serial with the measuring uC pin to protect the uC. Although the capacitance of the capacitor is not too high, a large amount of current will flow in a very short period of time while it is discharged.
Anonymous
Fri 23-Jul-2010 20:17
"It would be fun to extend this into a full-blown irrigation system by building some kind of motorized watering contraption,"

I've always thought the same thing. I've been dreaming about an automated greenhouse. I have all the control side down, but the moisture sensing circuitry was a little bit of a problem for me.

Thanks to you, I have that almost all figured out. The next step is to blend the sensing and controls together.

The problem I'm having is to get an output signal send to the controller once a setpoint is reached.

Thanks

p.s. you can really mislead someone with the title "pot plant" hehe

andrerouellette@gmail.com
Andre
Anonymous
Mon 7-May-2012 13:49
You could eventually use a piezo pressure sensor to weigh the plant in a similar manner.
once you know the 'ideal' weight for the moist level that the plant needs.
to dry then switch on a little airpump (aquarium pump) that pushes the water out of a closed jar, or a stepper motor pump (like in the old xerox inkjets) that pushes the fluid trough a medical rubber tube.
Anonymous
Tue 5-Jun-2012 05:07
Does the input of your microcontroller have a Schmitt Trigger?
Anonymous
Sun 13-Oct-2013 06:37
At University of Gent (molecular genetics department), we weigh the plants to determine if they need water. This way we can water them very precisely and there's no milliliter of water given to much or to little.

How do you account for the plants growth?