External Fogponic system design


A number of years ago i experimented with mist timing based on humidity and imho, its a non starter :wink: Water vapour is a gas and plants need liquid water, Secondly, there is a sizeable delay between the end of the mist delivery and the peak RH reading. The time it takes to drop back down is much longer and, worse still, it never falls back to the value it was before the mist delivery. Its an interesting experiment, all you need is a container, stopwatch, hygrometer and some mist :wink:



Juan`s experiment confirms the nutrient is present in the atomised droplets :wink:


I did not think a higher ppm would be necessary tbh, but now that you mention it I would like to know what happens too, it will have to wait, tho. Next step is to start planting using fogponics, I will let you know if it works. From the research I have done, I believe fogponics works pretty good, but you have to be very careful with the fog density and make sure the whole root is being fogged. I believe the brown roots people see in fogponic systems are because of oxygen deprivation from the fog is too dense. It’s easy to screw up and not notice.


Hey Guys,

I am using a modified brain box as the computer for my system. Here are some pics.

I connected two AC outlets ( each is controlled individually) and 10 12v DC outlets (orange) all controlled by 12 relays and connected to the pi. I am using an Arduino to connect the sensors to the pi too. The sensors come out through the white brush plate in the last pic. The atlas sensor connectors are on the right of the plate. The smaller gray box in the pics it’s a modified ATX 12v power supply.

The idea is to write a program where you can choose how to control any of the AC and DC outlets, either by time or through sensor readings. The computer could potentially automate any hydroponic system but is intended to work with the FC. It has 10 DC outlets, 5 for the peristaltic pump, 1 water pump, 1 axial fan, 1 ptc heating element, and 1chiller system (ion pump and fan connected to same power source). The AC power supply can be used for the lighting system. Because it requires more power than what the atx provides.

What do you think about the system? We may start pcb prototyping soon, do you have any recommendations on the computer before we cannot change the design? I am hoping that the 10 DC outlets would be enough.


Very interesting. What can that Brain Box control?


Anything you want to power either with 12V or to a wall plug. Turning on and off the fan or pump system depending on the time interval you set, turning on and off the chiller based on the water temp, adding more pH if pH of water lowers. Technically you can turn on and off almost anything that uses electricity, based either on a set time interval or reading from any sensor connected to the brain box.


The brain box is working! Right now it has only 4 relays connected. They are programmed to be controlled by a timer, I am changing two of them to be controlled by the temperature probe and the conductivity probe (maybe it can control nutrient dispensing??). I am designing the brainbox so it could eventually replace the harness in the FC2.0. Could someone help me know how many relays the FC uses and what does each relay control? So far I have come up with 9:

5 peristaltic pumps controlled by a timer
1 axial fan and 1 ion reservoir pump (both are controlled by the same relay, water temp)
1 Axial fan and 1 PTC heating element (both are controlled by the same relay, air temp)
1 circulation pump (timer)
1 lighting system (Timer)

For the sensors I have connected:
Temperature probe

Is there anything I am missing?

The sensors are connected to an arduino. As soon as its done I will opensource the code.


Here is a render of the app I want to develop to independently control each relay in the brain box. The idea is that you can program each relay separately and very easy.

We are currently developing three programs to control de relays, a cyclical timer that turns on at time intervals, a daily schedule that turns on at a certain time of day, and a program that uses sensor data to turn on and off. What do you guys think? any feedback is welcome.


Very clean/easy to use it seems.

I’m curious if you’re developing all of this from nothing, or starting with something as a platform? I know you used the configuration of a brain_box, but are you basing your app off that software?

@ferguman has noticed that the atomizers we are testing only are rated to last about 120 days if they operated 24/7. We were thinking it would be possible to have them turn on/off based on a program? We are working on an aeroponic system here as well @webbhm has been planning to modify the MVP software for us to use. Is your software on GitHub? Let’s not duplicate our efforts.


I have some initial software that the developer made from scratch. Its very simple, lets you control each relay independently to turn them on and off, You set the time you want them on or off, ranging from 1 to 100 seconds. I will post it on GitHub then. I think that software is what you need. It is not on GitHub yet, we are still debugging it a bit, but it should be available soon.

This is how the current program looks like, it displays the sensor readings at the end.


Aeroponic growth system with nutrient fog stabilization.

This invention provides apparatus and methods for stabilization of ultrasonically generated nutrient fog. To eliminate the sensitivity to the humidity and temperature of the intake air, the fog is recirculated from the growth chamber to the fog-generating reservoir. The chamber and reservoir are connected by a fog supply conduit and a fog return conduit which form a fog circulation circuit.


Thought you might appreciate this. Thanks to @wsnook for the link.

@Atom @yusuf.khan.su @ferguman


I think you have the three basic uses (cyclic, schedule and sensor).
The two key problems I have found are security and servers. Security can get complicated if you have a class of students, and several of them need to have access; or open access for viewing, but secured for making changes. Setting up a server is an issue for schools because of the need to expose the computer through the firewall. The other option is like Particle, where there is an ‘external’ server that everyone communicates with.
You have a good starting point; I am just interested in how you intend to handle the other issues that go along with it.


Thanks for the info! its very helpful. The patent was filed in 1999, it already expired!

It claims that,“The humidity level of ultrasonically produced fog is 100%. When fresh air is passed over the nutrient solution reservoir, the humidity level of the air determines how much of the nutrient solution is vaporized to increase the humidity back to 100% and how much is atomized into fog. Changes in the humidity and temperature of the intake air change the nutrient concentration in the fog. Additionally, as the nutrient solution reservoir is depleted, due to vaporization and atomization of the nutrient solution combined with resultant precipitation of nutrients from the fog, the salt concentration in the nutrient solution reservoir increases significantly. The fog can become so concentrated that it is toxic to plants.”

What do you think about the claim? It doesn’t sound very effective having the fog going through all those pipes because of condensation, but the idea is pretty cool.


I have no computer background so thanks for the input! Right now I am solving each issue as it comes along. I have been wondering about using a cloud platform such as aws, azure or from what I see, particle. This will provide the external server, security capacity, and scalability. It gets expensive pretty fast tho. Do you think cloud platforms is the way to go? Or how would you scale it up?


I am waiting till OpenAg goes to a cloud platform. No point in making a different path, and if they make it a community platform, then the MVP doesn’t have to foot the bill. Individuals may want to go the particle route if they need to make a move now, though unless the Particle is running its own actuators, it doesn’t play well sharing GPIO pins on a Raspberry. So far, I do not see a clear solution; and since I am working from home and can configure my router about any way I want, I don’t have a personal stake in the issue at this time.


That’s a really good point. I need to find out more about their cloud platform plans.


@Juan1 have you seen this? http://nutramist.com/


I am building my own version of a nutramist for around CAD $100, the high cost is due to top tier water atomizer @ ~5 microns and a water proof fan. the rest of the stuff is standard home depot items. I will post my design soon and include a google sheet with daily measurements and notes.

I will be growing lettuce, a strain of kale, green onions and leek.


Please start a thread about this! Take pictures as you progress.

I’m super interested to hear about your research so far that has led you to the conclusion that ~5 microns is the ideal size. Help us learn with you!

@Juan1 Did you ever get this system working?


Hey Peter sorry for being absent, I have been working on something. Yeah, I got it to work and actually compared it with a Kratky method.

@cohenoshri I am assuming you are using an ultrasonic fogger. They tend to heat up the water but not enough to be a problem. I set mine 2 minutes off 15 seconds on. My setup consisted of a 5gal bucket with a 1 disk ultrasonic fogger floating in the water. I noticed that compared to the Kratky, The roots in the ultrasonic fogger were longer so they could reach the water, maybe the fog was not enough?

I decided to check if that was the case. I made an external fogponic system. I used two buckets connected by a hose. One bucket made the fog and used a fan to send it to the other bucket (2 buckets w/ lid, 1 hose, 1 timer, 1 fogger, and 1 fan cost around $55 dls) where the plant grew. Unfortunately, the new system did not work and my plant died the next day. I am not sure if it was a leak, or if there was not enough fog.

Fogponic systems are very delicate, maybe you can make it work. Good luck!