switch pump on based on varying resistance of another cirucit

[twinturbotom]

I still have more testing and research to do but I'm turning to the experts for some direction; I appreciate any feedback...

I would like an irrigation pump to turn on based on the amount of water saturation inside of a garden bed. Been a fun project to build everything up and now I need some advice....

high level view: Solar panel recharges a battery which runs a pump. I would like to run a low voltage line (1-3 Volts) through my garden bed. The amount of water in the bed will change the resistance through the line. I need to do some testing here to figure out what the max and min resistances / currents would be based on water in my soil and then come up with the threshold value for pump on/off...

for arguments sake lets just say it's that simple... run a wire through the bed and the resistance goes up or down. I'm having trouble figuring out the best way to capture that information and use it to switch on the pump.

I'm pretty sure the resistance will go DOWN as the water content goes up. Therefore the pump should turn on when the resistance of the "sensing" circuit goes above a critical value and turns off when the resistance is below it. Not sure if it makes more sense to talk in terms of the circuits resistance or current here... Key is the pump should normally be OFF.

I imagine there being two or three circuits... the low voltage sensing circuit, pump circuit and some switching component or circuit that turns the pump on or off. I'm really unclear as to what that switching circuit or component is to work the way i want. Seems like some Relay? Anyway... any direction on the switching or relaying of a signal so that a pump goes on when the resistance through a sensing circuit goes up / current goes down would be helpful.


Thanks!

 

[strantor]

I think you could use a light/dark sensor circuit and replace the LDR with a flowerbed.

 

[Reloadron]

I agree with strantor. The biggest problem you face with circuits like this and keeping it simple is you end up using a few probes as electrodes to sense moisture. So you have a small current flow through the ground (dirt). Depending on the soil mineral content you get a mineral buildup on your probes which will start responding differently and need cleaned. The probes need the same spacing all the time and experimenting will tell you the resistance between clean probes based on soil moisture content. You need to get initial numbers to work with. The more accurate and corrosion resistant you get, the more complex things get.

Ron

 

[twinturbotom]

This is all great feed back. Thanks for the direction strantor. I'm working out the LDR circuit in a simulator now. Just bought this icircuit for ipad... pretty cool to drag and drop comps and simulate while on the couch! I was actually thinking of something much simpler than the LDR; although I'm not sure that's possible. I'll have some time to deveop this further in the next few weeks so I'll get back to you with my progress. Thanks!

 

[alec_t]

I don't think a continuous length of wire surrounded by soil will show much resistance change between its ends, because the wire conductivity is likely to be much greater than the soil conductivity (wet or dry). It's more usual to sense the soil resistance between two separated probes, as mentioned above.
DC energisation of the wire/probes has the disadvantage of their rapid destruction due to electro-chemical effects. AC energisation, for brief periods, can largely overcome that. The wire/probes should be of a corrosion-resistant material.

 

[twinturbotom]

I think the sensing portion will definitely change from my initial simplistic description of a wire in the soil. Right now I'm more concerned with the switching circuit based on some information from a sensing circuit... although it is all intermingled so I should start dialing in on cheap probes and other methods. Next week I'll start putting sensors in the dirt and mud!

Thanks!

 

[Reloadron]

When it comes to circuits like this I see the actual sensing circuit as the least of the problems. Give this link a read as to the use of comparators. The bigger problem is the sensors chosen. If this is a small simple garden and you don't mind cleaning the probes on a regular basis then things are easy. However, if you want set them and forget them probes, as mentioned then the circuits get more complex.

Ron

 

[alec_t]

I agree with Ron; the sensor is the tricky bit. The type of sensor (e.g. AC or DC energised) determines what circuit is needed.

 

[twinturbotom]

Reloadron... that link is filled with great content. Thanks! I need more time and education to digest it all. I'm a bit tied up on other things for the next week but after that I'll be digging in much deeper. Alec_t it'll be DC as this is ultimately powered by a DC battery recharged with a solar panel.

 

[crutschow]

You might consider a soil moisture sensor, such as this.

 

[strantor]

You might consider a soil moisture sensor, such as this.

AWESOME! I think I might get one.