Sign me up for one of these! I'm assuming "R 2500.00" is 2,500 rand, or USD $ 239. for the 80 square foot greenhouse, complete with working hydroponics system and controller and sensors, and maybe a few months of fertilizer. Hydroponics is worthwhile in remote areas of the Kalahari Desert, so if that is an intended location sure. A pH sensor and conductivity sensor are doable, but the budget will already be hit hard. A good pH sensor probably can be $100 USD is a real luxury item for a small system, and requires maintenance. Let’s say you build the circuit board and use an economical PIC, you probably could get away with a 12 bit chip (12f), though something like an 877A (which I've used), gives more flexibility if you wanted to attach a RTC for example, which would be quite useful. For argument sake, let’s say the circuit board and chip and programming are free. Also you may need a humidity sensor and a temperature sensor. There are cheap ones that work fine; I've used LM35 and a DHT11, nothing that costs much. But then you need some hardware, starting with a greenhouse structure, the plastics are going to cost. We make "hoop houses", Google it, or search this forum, by using thin PVC pipe (must paint) you can anchor with something like rebar (iron construction bars used to re-enforce concrete). You can break down (disassemble) these pretty quickly. You'll perhaps need a fan and something to open a vent in the top to control the temperature.

Thinking of all that, I don't think you will be anywhere near your budget even in production, unless you just use the controller to adjust the fan/vent, and leave the pH and conductivity to the users after teaching them to use a cheap pocket pH probe and EC meter, besides, to control the pH and EC automatically, you will need at least two actuated valves and maybe three, to add acid, and one or two parts nutrient solution, and also water.

Look into sustainable agriculture projects by searching for aquaponics. There are some interesting rural applications that work out with these, though you'll need to fill the fish tank with water and keep it aerated and filtered, so hopefully there is some rainfall, and a major part of that is catching the rain, though other Central African projects have actually reprocessed their waste water and I believe there is almost no need for fertilizer input - usually these are large systems though, say for a village. If you are just designing for a family in the wilderness, I can see where you are coming from), but this is outside of the scope of just a digital electronics project and falls more under sustainable agriculture.

A microcontroller probably isn't even necessary if the people are trained but it could be integrated feasibly for sure if you are growing in the greenhouse for just temperature and humidity lookup/open close vent/fan/heater stuff.

Fertilizers are going to be a cost with hydroponics and someone really needs to be around to supervise, so first you need to define carefully, the basic need for automation, so you can design something practical. For example, you mention portability, and once or twice are ok with cheap materials, but if the users are mobile and need more than that you will need to look into new materials.

To cycle the water, simply, all that is required is a submersible pump, perhaps on a timer, perhaps not. Timers are commercially available very economically, like $10-$20 USD. So I don't think you simply want to re-invent the timer unless you survey the need carefully and determine that some strange combination of controls is necessary. Photoresistors or photo-semiconductor can be used for day/night, if you don't want a clock.

I guess the msot successful thing to do would be first spec out your need and growing conditions. A typical pump is only about 25W (typical) to 50W depending on the system, and you would need channels or pipes to mount the plants inside where the water runs. Normally the pump just lifts the water a meter or two and gravity does all the rest. By specifying the growing conditions and also what is being grown you will be able to design a hydroponic system most efficient to whatever you are growing. The tank itself isn't too big of a deal, for your calculations consider between a minimum of around 2 liters to perhaps as much as 8 liters per plant to size the tank. 2 liters works ok for plants like leafy greens, other veggies can require more. Plastics are usually used in the tanks, from drums like the sort vinegar is shipped around in to totes for storing things around the house. You want a plastic that is non-toxic for agricultural use, or people if such a thing is available.

An irrigation controller is more necessary if you are growing in soil, because you can put a soil sensor (plaster of Paris, when home made) in the ground and get a signal for the water content of the soil and water on demand. With hydroponics you pretty much avoid this problem by just cycling the water at regular intervals. They always need to be wet, so most of the time you can just set up the timer according to the system. I can't see the need to put a sensor in the rootzone on one random plant and make the watering frequency based on that one particular point, since but I suppose conceptually it is possible, to cut down on some watering, but in hydroponic systems we like to have as much water moving continuously since when it stops, it tends to get oxygen depleted. So it is not so much the dampness of the roots-they are always damp), but more the aeration of the water, which we never seem to get enough of :-(

I know you started by asking about irrigation, but it sounds more like you have a solution looking for an application, so I thought this would be helpful.
Good Luck!

This post was edited by PupillaCharites on Wed, Mar 12, 14 at 12:30

"An irrigation controller is more necessary if you are growing in soil"

I have seen them used on container plants with success. The actual media that greenhouse plants are grown in does not contain any soil. I always thought pro-mix should get an honorable mention in the hydro world, since it is technically soil-less growing. The controller could be useful for a flood and drain setup with a more traditional hydro media like rockwool. A greenhouse fluctuates like mad in temperature without careful venting, so the standard practice of just using a timer would be improved upon by using the controller.

Alternatively, the OP could build a run-to-waste system, and use the controller to minimize waste. Run to waste systems are common in commercial use, and as the name implies, they are wasteful. The runoff can become pollution as well.

Pupilla Charites,

Wow, do I appreciate your comment summing up in a nutshell everything important that I need to look further in to!

I do hope to make this project feasible and maybe even get some patent rights on it ~ early retirement at the end of the tunnel (some pun intended with the tunnel). :)

Ok, before I get carried away completely, let me comment on your comments. The intended area wont be as harsh as that of the Kalahari Desert, luckily. Average midday temperatures of 18 - 29 °C and night temperatures between 0 - 15 °C throughout the year.

My first concern is the structure, after your post and some more research I got a light bulb idea of combining a, A-Frame and a Hoop Greenhouse like you mentioned. https://www.dropbox.com/s/2rkstr9ghdhfcak/A-Frame.jpg and https://www.dropbox.com/s/juzxfmnbtrelric/Hoop%20House.jpg

The only concern I have with the above possible combination is higher growing vegetables, such as tomatoes for example. Even beans as they grow quite tall.

As for the type of Hydrophonics, aquaponics being the most idea as far as I could research, I am considering the common Flood & Drain (Ebb & Flow) method. (Comments on this method very welcome).

Another unknown area for me is the power needed by the pump to get the water from the tank to the top of the structure. Will a fishing pond or fountain pump do?

Then, ventilation and heating. Do I need a heating source inside the greenhouse or only some fans?

Above pretty much cover the physical greenhouse part. As for the controller a RTC will be useful like you mentioned. So I think I will stick to the PIC you used or find a similar one. The DHT11, DHT22 are both temperature and humidity sensors which will then work just great. There is a well written library usable on an Adruino board which might lessen some troubles with the programming part.

As for the pH readings, the intended market is that of a rural area, thus the person operating it, most likely won't be schooled. This said, as little as possible should be done by the user.

Ok, sjoh.. that was a mouthful for now! Thanks allot again for all the comments and help thus far! Appreciate it

Cole Robbie,

I will definitely be using a controller to control the temperature and humidity inside the greenhouse at least. Due to the fact that this is a Computer electronically project after all, I might just get carried away on a mechanical part of the greenhouse and miss the whole point.

As replied to Pupilla I think I am going to stick to the Flood and Drain system or also known as Ebb & Flow making the controller useful as you mentioned.

Thanks for the feedback and comments!

When you get the controller worked out I would be very interested.

Hi SP,

For your hardware specs, yes the pumps are usually just fountain/fish pond mag drive submersible pumps. 15W - 30W pumps deliver reasonable flowrates for small setups in the 1 - 2 meter lift category, just be sure to get a look at a manufacturer's pumping "Performance Curve" and factor in around 30% margin since they are optimistic in a marketing way.

Find the pump performance curve you select, or use these from a guide generally for the wattages for most of the similar efficiency pumps, from the horticulture source, one rare supplier who actually gives specifications of what it sells without leaving us guessing, right on the item page:

Horticulture Source EcoPlus pumps series with complete performance specifications

Don't forget to cover the initial surge required to overcome the pumps resting inertia for on/off cycling, which is greater and something to consider if you have a battery/solar power source, too low of a surge might lead to pump replacement sooner, but don't know that for a fact.

Ebb and Flow might be decent, or earlier Cole mentioned run-to-waste. That doesn't sound great, but if you have a system to put in the hands of people who aren't critical thinkers, pH and EC control might be a little to much trouble. If I were in this situation, really focusing on my project, I would not attempt to put in an acid/fertilizer/water adjusting system. They will require their own pumps of a different kind, probably peristaltic pumps with stepper motors to give you a programming routine converting sensor readings into pump turns, for precision metering for control of pH and fertilizer strength. It will get complicated and costly and then there will be more tanks to keep around and more chances for errors.

Now, taking the run-to-waste idea and going further, follow the keep it simple philosophy, I think you could simply allow the "waste" to run into a recipient reservoir and someone in the simplist case can just dump it back, when mixing up a new batch of fertilizer. As soon as salty (more conductive than water) fertilizer solution from the on cycle reaches and passes over a sensor at a certain point in the system, it cuts off flow, so there is less waste except excess that drains out after the pump is off.

They would just need to mix a measured amount of fertilizer with water and perhaps acid and fill up the main tank.

Since energy is critical, you may have the controller moisten the growing media two or three times a day with a low power fountain pump (or possibly, mounting the fertilizing reservoir on say a wood platform and just actuating a solenoid valve switch and let gravity do all the watering, no fountain pump necessary).

I think ebb and flow can be a good one too, but it requires more attention than this, I think :-) There may be other contrasting opinions, that's what fun about agriculture. No one has yet designed one system that optimally works for all crops.

The temperatures you mention near zero C mean a heating source is necessary as does a fan for radiant warm days, but that is a whole different field. Usually when farmers design their system they first specify the vegetables they will grow in the temperature and humidity conditions they have, then decide upon the simplest, most economical system to do it, and the controls are adapted to that. It is quite a task to flip the design strategy inside out. Hey, if there can be an internet interface and internet, your first prototype controller could also relay back the sensor info along with a webcam even so you could watch and improve...

As for hoop houses, the one you have a picture of is more a "low tunnel", and they are cheaper and quicker. But ir you want to keep to the 2m X 4m size, you could try to make straight up sides with the rebar re-enforcement, and then crown it with the low tunnel to get a "high tunnel". Otherwise you need to increase the width, if PVC is the material of construction. Others use pipe benders to make keystone like center roof connectors and then run pipe straight down. Perhaps there is an economical source of pipe like electrical conduit which will meet the need where you are..