Arduino Watering by Moisture

ShirkGoldbrick

Active Member
Did a quick search and haven't been able to find anything quite like this setup. Can someone tell me if they've done this and if they can paste the code here it would be great!

Objective: Automatically water plants based on moisture need during lights on period, replacing the need to fiddle with timers.

Enabling Objectives:

1. Place 4 moisture sensors strategically around garden points where plants will likely dry out first. Some examples include: in a pot directly under a light that will cause it to transpire more, in a pot nearest a fan that will transpire more, in a pot containing the largest plant which will transpire more.

a. Once targeted sensors drop below the "water me" resistivity as detected by a moisture meter it will trigger a solenoid to turn on a pump in lieu of a timer.
b. This will only happen during lights on.

Supplies
-Arduino board and expansion board(for easy connections)
-4 soil moisture sensors
-5 3 pin jumpers
-photosensor
-power supply

misc:
-tools for re-pinning connectors
-extra wire
-enclosure for board
-enclosure for photosensor
-cable glands
-weather proof plug
-relay board
-etc.


Theory of Design:

Apparatus should be contained in a water proof box with one power supply plug into original power source and an outdoor weatherproof socket to supply power to pump.

There will be 4 moisture sensors that exit cable glands and have a length of wire for choice in placement.

One photosensor shall also exit cable gland and have a length of wire while placed in a smaller weather proof enclosure. This will determine whether the lights are on or not.

Theory of Calibration:

In one example using a rockwool hugo block water with 0.125L of 1.0EC nutrient solution (equivalent to 15 minutes of drip cycle on a .5Lph drip nozzle as recommended by Grodan). With a sensor in place write down the reading from the sensor and weigh the block. Once the block has lost half its weight write down the reading from the sensor again, this will be "water me" time.

Theory of Operation:

Once one of the most susceptible plants requires watering a drip system will turn on and water for 15 minutes (if using hugo blocks and .5lph nozzles) or until the saturated moisture reading is reached. Not all plants in the garden may "need" water at this time but they are expected to be within an acceptable range to prevent over watering, prevent under watering, and allow an adequate "dry cycle".

Benefits:
No fiddling with timer as plants grow or as temperatures/light intensity changes. Plants are provided with an optimal watering regimen for growth taking all the guess work out.

Cost to build: I estimate $70-90 for the main supplies and $20-40 for the miscellaneous. Total cost to build is estimated at $90-130.

Optional Advanced: Include a screen that displays the garden moisture in a % as averaged between the sensors and with "water me" at 0% and saturated at 100%
 
I didn't read all of your post, don't want to.
Here are some reasons why no body gives a fuck.

Not everybody grows in a soil like medium...voiding the application of said moisture sensor.

There's really no point to automate with a moisture sensor. When you can just press a button, plug in a pump, or hand water etc...

I don't know what sensors your using but any exposed metal cheap sensors will erode.

the voltage readings of the sensors are also pressure sensitive i.e. compacted soil will give off a higher voltage than looser soil. With good aerated soil that is needed uniformity becomes less. Can you guarantee that the pressure distribution and the soil imprint will remain constant for the entirety of the grow? If not then it becomes less accurate.

At this point it really becomes just a switch.

Most people are going to inspect their gardens on a daily basis given the fact they have a good amount of money or soon to be invested. It isn't a hastle to visually/physically check and press a button, set a timer, plug a pump, etc...
 

MikeGanja

Well-Known Member
I found various Arduino hydroponic projects on Youtube. I think Arduino is interesting because it makes it possible to control a lots of parameters, or just collect data at a low cost. But I must admit that I wouldn't trust my grow op with a device I built myself. Perhaps as extra safety like high temp close down, SMS alarm if there is a waterleak in the grow area.
 

damnsmoker

Well-Known Member
I have also considered a similar system but wanted to incorporate multiple Arduinos, a Raspberry Pi and some relays to control the lights, humidifers, dehumidifiers and possibly a CO2 tank. I'd also like to connect a touchscreen for manual overrides, status displays for temperature, soil moisture levels, humidity, and to keep and display data and error logs.

I know that ASCO carries a line of solenoid valves that can be controlled pretty easily, some of them are even food grade.

Many years ago, I built a tweakable timer for a humidifier using a Motorola 555 chip, a couple of trim pots and a relay. It could turn the humidifier on and off for time intervals set by each of the pots (one pot for the on interval and one for the off interval); worked like a charm. Now, with this newer tech, it should be much easier to control an enclosed environment.

Sadly, as far as projects go, it's pretty far down on the priority list.
 

sonson176

Well-Known Member
Ever look into a programmable logic controller? Might be a little better suited for the given application.
 

damnsmoker

Well-Known Member
Ever look into a programmable logic controller? Might be a little better suited for the given application.
Not really, I'm just a gear head, mechanical type (I can fix this with a hammer, right?) and haven't any experience with PLC programming. I'm not saying that I couldn't learn, but that'd be yet another thing I have to do before I can set up some grow room automation.

The thing I like about the Arduinos is that they are easy to use, and they are very inexpensive. I also have a couple sitting around, as well as some stepper motors; I may even have a couple of Pololu motor drivers still kicking around somewhere. And I know how to use them.
 

bicit

Well-Known Member
I didn't read all of your post, don't want to.
Here are some reasons why no body gives a fuck.

At this point it really becomes just a switch.

Most people are going to inspect their gardens on a daily basis given the fact they have a good amount of money or soon to be invested. It isn't a hastle to visually/physically check and press a button, set a timer, plug a pump, etc...
Who pissed in your cheerio's this morning?

1: Quite a few people grow in soil and soil like media like coco-fiber. I'd argue that even today hydroponics still remains a niche compared to the number of garners using soil and soilless media.

2: The point of the sensor is to trip the switch for you. While you're not around. Even if you forget to water, end up in the hospital, go on vacation,

3: Captivate soil humidity sensors can be coated with an epoxy to prevent corrosion. More than one way to skin a cat there, lots of different sensors to measure soil humidity. The cheap resistive sensors on ebay will corrode however.

@ShirkGoldbrick This is the most inexpensive capacitive soils moisture sensor I've been able to find. I've been meaning to pick some up to pair them with some solenoid valves to take care of watering.

https://www.tindie.com/products/miceuz/i2c-soil-moisture-sensor/
 

ShirkGoldbrick

Active Member
Nifty article. Of course it can be done but you'll have to pay attention at first to figure out which pots are going to dry out the fastest and base the watering on them. I've noticed that it's not always so obvious which will dry out first heh heh.
 

bicit

Well-Known Member
Nifty article. Of course it can be done but you'll have to pay attention at first to figure out which pots are going to dry out the fastest and base the watering on them. I've noticed that it's not always so obvious which will dry out first heh heh.
Put a sensor in each pot with a drip ring. Hook the drip ring up to a solenoid valve triggered by the sensors.

At least thats what I want to do. I want a digital blumat lol.
 
Top