PDA

View Full Version : Automated Greenhouse



Mitsukai
02-09-2012, 07:52 AM
I am really new to the Boe-Bot and all of it's uses. I have only worked through the book that comes with the Boe-Bot full kit. Anyways, one of my friends was talking about trying to make an automated green house. I know it is over my head right now...but I was wondering if anyone could point me in the right direction to figure out how to do it.

I know I will need to monitor light, temp, humidity.
I will also need to be able to turn on sprinkler systems, fans, heaters
I will also need some kind of mapping system I think so that I can monitor what plants needed what and have some sort of log for whats been done.

Can you guys think of anything else I'll need?

Where should I start with this project?

bsnut
02-09-2012, 11:39 AM
Welcome to the forums.

What it sounds like to me, you want two systems to work together to control your greenhouse. One system to control the fans, dampers and monitor humidity and the BOE-BOT to water the plants. Is this what you are thinking. If this is what you were thinking, then two X-Bees would work great by controlling the fans, dampers and monitoring the humidity since the X-Bees have I/O to handle that part and the other X-Bee can be on BOE-BOT to tell the damper to open or close, start and stop fans and receive humidity sensor values.

RDL2004
02-09-2012, 02:53 PM
Take a look at "Process Control" and "Applied Sensors 2.0", both are available as free .pdf downloads from Parallax. These books were a great help to me when I was getting started.

http://www.parallax.com/tabid/535/Default.aspx

kwinn
02-10-2012, 02:09 AM
I know I will need to monitor light, temp, humidity.

This is where to start. You need to measure these values in order to decide what needs to be turned on or off, opened or closed, etc. You may also want to measure soil moisture content.


I will also need to be able to turn on sprinkler systems, fans, heaters.

There are solenoid valves, relays, motors, and other actuators available for all these functions and more. Another thing you will most likely need to do is open and close windows/shutters to control temperature and air flow. No point wasting energy running fans when opening a window accomplishes the same thing.


I will also need some kind of mapping system I think so that I can monitor what plants needed what and have some sort of log for whats been done.

Most automated commercial operations divide the greenhouse(s) into sections that are set up to provide specific conditions and grow plants that require those conditions in that area. This minimizes the amount of manual logging required and maximizes the the automation possible.

Mitsukai
02-10-2012, 04:04 AM
Thanks for the quick replys you guys!

I was thinking about using the Sensirion Temperature/Humidity Sensor to measure the temp and the humidity. Is this a good choice? Is there a way I can mount it close to the plants and run a wire back to the microcontroller or is this one of the parts that needs to be soldered onto the board itself?
The other two parts I was looking at were the DS1620 Digital Thermometer and the HS1101 Humidity Sensor.

I was thinking of running the light of a clock (on for 12hrs off for 12hrs). I figure this could be done in programming? Would it be more efficient to use a external timer somehow?

What could I use to measure soil moisture? I looked in the parallax store but couldn't find anything?

I definitely don't want to waste any power on a fan that I don't need to. I honestly have no experience with solenoid valves, actuators, or any of the other things you mentioned. All I've worked with are the little servo motors that come with the Boe-Bot full kit. I figure I could use those to open and close a window using some sort of pulley system. Is there and easier way?

I guess if the greenhouse is divided up I wouldn't have to map anything because all of the plants would need the same conditions.

Duane Degn
02-10-2012, 05:42 AM
I was thinking about using the Sensirion Temperature/Humidity Sensor to measure the temp and the humidity. Is this a good choice? Is there a way I can mount it close to the plants and run a wire back to the microcontroller or is this one of the parts that needs to be soldered onto the board itself?
The other two parts I was looking at were the DS1620 Digital Thermometer and the HS1101 Humidity Sensor.


I've been using a Sensirion sensor a lot lately. I'm very pleased with it. It's an I2C device (at least sort of). I read it might cause trouble on an I2C line but so far it has been fine attached to the same pins as the Propeller's EEPROM.

I've been using my sensor on the end of a meter of wire (four wires). I was worried about using such long wires, but the Prop seems to be reading the sensor just fine. I wouldn't trust wire over a meter long for I2C communication.

I have a couple of the HS1101 Humidity sensors but I couldn't find any objects that would give calibrated readings. I'm not opposed to writting my own objects, but since the Sensirion sensor was working so well, I didn't spend time getting the HS1101 sensor to work (yet).

I bet I have a DS1620 some place. I haven't used it myself yet.

The HS1101 sensors cost so much less than a Sensirion, I think it would be a good idea to get a couple of the HS1101 sensors and one Sensirion sensor to use to calibrate the HS1101 sensors.

Loopy Byteloose
02-10-2012, 09:38 AM
Parallax has good automated greenhouse solutions - the BS2 and the Propeller. Distances are involved, so some sort of network is called for. While I prefer CANbus, simple RS-485 networking can do quite a bit. The prime advantage is that twisted pairs of low voltage wire can place the sensors where they will do the most good.

You might study up on dew point, as well as simple humidity and temperature measurement. Orchids can easily be damaged by a drop in temperature crossing the dew point and creating standing water on the plants.

Senitron is a useful sensor, but Parallax also has others that are simpler to interface and read only humidity or temperature. Light is a more complex issued as wave length sensed has some bearing on the usefulness of data.

Duane Degn
02-10-2012, 10:03 AM
some sort of network is called for

I also like RS-485 for wired communication. JonnyMac wrote a SpinZone article about using RS-485 with a Prop that helped me start using RS-485 in my projects.

For wireless communication, I'm a fan of Nordic nRF24L01+ modules. You can purchase the modules for less than $3 on eBay. There's link to the eBay modules in post #1 of my index. Post #2 of my index has a link to the nRF24L01+ driver.

Loopy Byteloose
02-11-2012, 03:19 AM
An automated greenhouse can quickly expand as a project. At first, one might just acquire data in real time; but sooner or later this is not enough.

If you want reports emailed to you at periodic intervals, some sort of internet interface is required AND you need a real-time clock to apply time and date to your emailed data files.

kwinn
02-11-2012, 07:10 AM
@Mitsukai

Soil moisture can be measured with a simple resistance probe made with two conductors mounted on opposite sides of a plastic stake. It will need to be calibrated for the type of soil it is going to be used in. Calibration is done by drying out a volume of soil and measuring the resistance values while adding adding measured amounts of water. Simple but time consuming since you must wait for the water to disperse before each measurement.

Loopy Byteloose
02-12-2012, 04:20 PM
Soil moisture probes are sold by several agricultural supplier, but they do seem to wear out over time or require recalibration. A simple resistance probe is affected by ion content in the soil. Some sensors are actually intended to measure water level in the sub-soil; others are intended to estimate soil humidity. The better solution is to have multiple redundant sensors and to try not to rely heavily on their precision.

Much depends on what you want to grow in a greenhouse and knowing what is necessary to optimize at each point in the growth cycle.

kwinn
02-13-2012, 02:40 AM
Soil moisture probes are sold by several agricultural supplier, but they do seem to wear out over time or require recalibration. A simple resistance probe is affected by ion content in the soil. Some sensors are actually intended to measure water level in the sub-soil; others are intended to estimate soil humidity. The better solution is to have multiple redundant sensors and to try not to rely heavily on their precision.

Much depends on what you want to grow in a greenhouse and knowing what is necessary to optimize at each point in the growth cycle.

All good points Loopy. The resistance probe is more of a qualitative rather than quantitative measurement and the system can be self calibrating within certain constraints. By knowing the flow rate of the irrigation system and starting with dry soil initially known volumes of water can be added incrementally and resistance measurements taken after each increment. These measurements can be used as a base line for on going irrigation as well as calibration correction.