Anyone Keen in Green House Sensors and Control System DIY?

Grow house maybe...?

"I am going to develop number of things and if suggestions match for the need then I can place information for discussions here."

Why not place it first and let the discussion ensue naturally?

If your talking small scale DIY, then why not use an Arduino?

Simple, cheap, effective, and there is a LOT of information (all open source) on the web.

Hi Kilowatt0:

Yes, any small SBC will be fine but sensors and control system DIY type planned. I think Arduino is a very good suggestion.

There have been several times that I have started a similar investigation and so far I have not done well at finding moisture sensors and the data to make me feel good about their performance. There are very cheap sensors and very expensive sensors but not much that is a little off the bottom. To buy a really cheap sensor and adapt it to a microcontroller based system seems to me like I will probably be disappointed.

Beyond sensing moisture the Arduino suggestion is a good one. I'm assuming you are basically interested in measuring moisture and temperature, turning water on and off and possibly controlling fans or vents. Arduino and other projects (PIC is strong) exist for all of these, but I always found moisture sensing in different soil types to be weak.

My interests are just my wife's garden so the projects never went beyond my first stumbling blocks.

Hi BruceFlorida:

You have brought out points nicely so I can have a starting point on this now.

1) Soil moisture sensor - desire of water or mist for the plant

2) Water tank level sensor - perhaps for water management

3) Sun light or day/night sensor - to supplement light source for increased growth

4) CO2, NH3 gas sensors - optional for trapped air gas sensors

5) Nutrient in water sensor for water based nutrient system

6) Temperature and pressure - optional atmospheric condition sensor

7) Plant growth sensor - optional

8) Plant leaf color sensor - optional

We have this experimental farm where right now we grow potato and I will like to have few green houses for experimental purpose. Another one is planned on roof top and both will come up soon immediately after we get rid of this rain season here. In white dress is my son who is in USA right now. I will also be experimenting in a small green house my daughter has in UK. I have large amount of farm land for this purpose. I also have bore well water source. As I am doing from scratch level zero we can share all information.

There are many 'open source' controllers available, and companies like 'Sparkfun' sell temperature, humidity, light level, et cet. sensors for a reasonable fee...what I used to build in an STD bus box for $6K 20 years ago can be put together for a couple of hundred. Controllers can be linked with 'xbee', wifi, or be hard wired. Check out 'PcDuino', a unix based, Arduino compatible board used at UC Boulder for teaching. There is a plethora of software and hardware design open sourced through the 'Maker' movement, do a search on 'Instructables' for similar things, you'll be amazed at what's been done with microcontrollers. I'm waiting for the new 'Propeller' 16 core controller which should be out in sept. of this year.

Hi geraldpaxton:
For a small system, few sensors and few controls, any micro-PLC can do great job.
For large number of sensors and distributed controls looks a bit tricky. Most of the green house use a common water drip irrigation system or locally flow adjustable water outlet. These systems are cheap and work fairly well.
I thought of using valves operated by the moisture desire. As this requires monitoring and control over many places, it is going to be expensive.
Another thing to consider is the variety of plants one may have within one zone that may need different level of water irrigation.
Air temperature is to be maintained along with humidity in the air and 400nm to 700nm light source, less IR and less UV.
Some kind of steel frame structure dome and perhaps strong polyethylene sheet cover may help to create a good green house chamber.

'any micro-PLC'

check out: http://makezine.com/magazine/make-36-boards/which-board-is-right-for-me/

it's a 34 page article, written Feb.,2014 (there have been many added since, seems many are jumping on the bandwagon)

The other thing to consider is IDE(integrated development environment). What language do you want to become familiar with? There are subtle differences in compilers which can be frustrating.

The microcontroller approach can be cheaper and you can make your own cables to interface to standard industrial controls.

PLC's are vendor specific, and the interfaces are vendor specific and that can get expensive, and they all have their own programing environment to learn.

Green house requires larger integrated system and hence standard PLC is much better idea. Ladder diagram is a very simple language and one can learn in one day.

Hi geraldpaxton:
For initial phase, microcontroller, moisture sensor, tank water level sensor, air temperature sensor, room light sensor have been planned so let us see how all this go with few plants of tomato and capsicum.
Small group of engineers became keen so they will be helping me in this project. I think within two weeks we can have a working plan now.
I am going to consider the design of the moisture sensor now.

For the tank level sensor, I would consider the Parallax 'Ping' (appx $30 US) ultrasonic sensor mounted with a window on the top of the tank.It outputs a timed pulse relative to distance.

For the moisture sensor, I would consider multiple, spaced sensors, tied to something like an Arduino Uno (5 analog inputs) and a comparative routine. IMHO it's more accurate to develop a routine that determines rate of change/distance (due to chemicals that change absolute resistance) and soil type.

You might check into the use of the pallet sized tanks where the top 12 inches is cut off and used for plants and the lower part used for fish(perch/tilapia) with the water recirculating for the benefit of both.

I also agree the use of 'ladder logic' schematics and PLC compilers that generate software for PLC's are quick and simple, but I found myself linking assembly routines for faster execution for some functions.

The Arduino movement started a controller revolution, putting the use of microcontrollers into the hands of people for as little as $3 US (Atmel chip with a couple of resistors and a capacitor) dedicated to doing something.

Hi geraldpaxton:

Normally to save money people use common irrigation system with timer operated valve. Distributed irrigation system requires sensor for each plant and separate valve for water flow control for each plant which needs massive structure of sensor and control electronics. While I do have such resources but it looks like a waste of resources.

I do have about 500 numbers Atmel Microcontroller boards ready to be used and also 500 numbers soil humidity sensor boards and sensors. Hence for small green house with 500 plants I can manage it easily.

I also have water level sensors for water storage tanks, room humidity sensors and room temperature sensors, also some photo cell and LDR based room light sensors.

I need to get the water flow control solenoid operated valves manufactured. There is a local company which manufactures for about $2/each so I may go for these.

I also have water pressure sensors which can tell if water is flowing on not. Basically it senses the water pressure in the pipe and is good to know if pump is pumping water on not.

What else we can look into remains for the discussions so let us share more point. So far discussion is progressing very well.

On my automated welding systems with water cooled torch I use a flowswitch (1/2 gpm) on the return line that gives me a bit to test for water flow.( if water is returning, it surely must be getting there) You might use a minimal flow return from the end of the line to the input to determine if pump is delivering flow/pressure throughout the line. In my garden I use 'drip' system with 4 spaced sensors (If it's been raining, system locks out until soil dries to a point that the sensors(tied to one of Chip Gracey's Propeller chips turns on pump and starts cycling the drip system). The 1000gal tank is monitored and if I don't haul enough water to it, one of the processes running on the Propeller keeps a minimal level by opening a valve to city water to maintain a minimal level for irrigation function.

Good. Existing moisture need to be monitored. Polyhouse don't get rain water.

I don't know about others but I do find this interesting. It is useful (for some of us) and as individual tasks they are educational for both current students and those of us that haven't showed up for class in a few decades.

Yes, there are many project sites and postings on the Internet. I don't know if CR4 is the right place for it but most posted projects are basically "do these 6 steps" and don't go into any engineering details. The idea of having engineers post projects for engineers is somewhat unique.

Greenhouses are best if the sensors and instrumentation is very simple, KISS=Keep It Simple Stupid. They usually use ordinary house thermostats to control the fans which control large fans to cool the temperatures. Water content is usually small hygrometer cells which control the overhead sprinkler systems. Really though, water is usually applied by a low wages person without a green card. Heat is provided by water or steam space heaters connected to another household thermostat. A water tank level sensor us usually a float or buoyancy mechanical device. Depending upon the money that is to be committed some use the float and fill valve from toilet tanks to maintain a certain level in the storage tank.

Light sensors to determine the ambient light are best done with simple photocells. In greenhouses most of the light control is done to a prescribed length of time. In other words the grower determines how much light he wants and uses incandescent lights to supplement the sun or overhead blinds to reduce the sun's intensity.

Experimentation with plants is not rocket science. A good grower is hired and he/she will evaluate the conditions, many of them can not be done with instrumentation, and will adjust the parameters to achieve the desired properties of the plants, By varying the different parameters they can usually cheaply achieve a good crop.

Generally, greenhouse operations is a low profit and competitive business. For this reason the owners are very tight with their money and will operate on a very low budget. Whatever they save is profit in their pockets!

Suggest that you do not automate your growing until you have several seasons experience. This way you will have a better idea of what you want to do and how to do it. Depend upon a good grower before automation and after automation.

Good Luck, Old Salt

Hi old salt:

Greenhouse is not a rocket science. Nasa now has it in space also but let us not get involved with them as we are not wasting Government money.

Most of the items as sensors and control systems you have listed are good enough.

As I am experimenting to this level for the first time and I have access to so called experts who are not any real experts around, I and my team members are going to do both, like take help from vegetable planters, agriculture professors, and some of engineering help here and there.

Water level sensors I have already discussed at http://cr4.globalspec.com/thread/84531/Water-Resource-Management-in-India

Wireless devices for control you can see at http://www.asro.in/projects/

Yesterday I called local solar panel supplier to provide me an estimated 10kW 20% plus efficiency solar panel modules for water pumps. I do have diesel pumps and may also get power from grid. However, solar is a good idea to try out.

I have planned electrical fencing to protect from wild animals mainly and security cameras to protect from human interference.

As work on crop has to start in September end, I have purchased 25 truck load cow dung for compost fertilizer. Soon seed beds will also be ready with baby plants for plantation in the green house. I will plant 100 Neem Trees around the farm soon to get natural insecticide as no chemical insecticide to be used on our farms. Trees will also stop wind a little bit.