My shot at a "Food SuperComputer"


#1

I saw @Caleb famous Ted Talk nearly 2 years ago and immediately knew I must build something like the Food Computer. My original goal was to build what would basically just be a larger food computer, based on high pressure aeroponics. Nearly 2 years later, I have gone far beyond this, but it’s roots and similarities with the Food Computer can still be easily seen. My dream is that one day these systems can make large data contributions to the Food Computer network, and more broadly, the #nerdFarmer community. More to come on this.

I wanted to share my prototype system with you guys and get your thoughts and feedback!

I call it the SpaceGarden.

It utilizes a variant of high pressure aeroponics that I developed to solve a lot of difficultuies that I found with my HPA system. It’s about the size of a bunk bed and has sites for up to 96 plants. It has 4’ of shoot growth area, and 1.5’ of root growth area so you can grow tall plants like tomatoes or cannabis, as well as rooty plants like potatoes and carrots. It is capable of maintaining optimium climate even in freezing or extremely hot environments (like a basement or garage in summer or winter).

Features:

  • Temperature Control (+/-)
  • Humidity Control (+/-)
  • Reverse Osmosis water filtration
  • Co2 Sensing and Boosting
  • 3 mode LED Lighting (Or lighting of your choice)
  • Carbon Filter Air Exhaust
  • Locking doors
  • Built in 1080p HD Camera
  • Air Circulation
  • 3x high pressure aeroponic root chambers
  • Root temperature control (+/-)
  • Water temperature control (+/-)
  • 5x Nutrient injectors accurate to <1ml
  • pH Sensing
  • EC/PPM Sensing
  • Backup battery

These photos are of the most recent completed prototype. The production model is in the works now and will be available for pre-order in Spring. Will be unveiling more info about the system in the coming months, check out the website if you’d like to stay updated!

www.ElementAgriculture.com

I would love to answer any questions you have!


Grow your own algae protein: Spirulina Photobioreactor
#2

Hey Mason,

Thats pretty amazing, too bad you cannot fit more plants. What is the price you are thinking of selling it at? I am building an environmental computer, Climate Computer, Check it out, I included the code for a webapp program you could use to control your Supercomputer if you want. How are you automating all your components?


#4

Thanks for sharing, looks like you’re using mostly php whereas we’re pretty strictly Python. It’s really able to fit quite a bit of plants (up to 96 if they’re small), and can accommodate relatively large plants as well. The SpaceGarden will cost about $6,000 with the first 5 buyers getting a very large discount + early delivery.

The automation is done with a custom PCB that emulates the Raspberry Pi / Arduino combo in a lot of ways. The controls are all hosted through a Flask webapp.


#5

Hello,

It is really cool, and your picture of prototype is looks operates well.

In my short experience with PFC, until now i’m more focused making “Recipe” in the “Monocropping” environment. In your “SpaceGarden” operate well in the “Polycropping” environment, how to control or using “Growing Recipe”?

It is a good project to think how to integration “Polycropping” in the PFC!

I hope take answering from you any idea and experience about your “SpaceGarden”.

Thank you.


#6

For the most part, the entire system will be operating on one grow recipe. However, the watering intervals and root temperatures can be adjusted on 3 separate root chambers individually. Good thing is that many plants share a happy range for temperature and humidity. The trickier part is the nutrients, as that’s probably the most powerful single variable of the grow recipes. I am working on some tech to try and isolate a particular nutrient solution to a particular root chamber but it is difficult to do effectively and even more difficult to test.

There will be an option for the SpaceGarden to include 3 totally separate environments, but this will cost considerably more.

I will certainly have to give polycropping integration a lot more thought, especially on the software and logic side of things.


#7

@mason
I absolutely agree with you about the if you are making separate 3 environments, it is really high cost.

I understand what you are saying, in my humble opinion it have a some strength in the polycropping envrionments, but dividing root chamber invoke some high cost. If we are using some recipe and control each chamber they have to connect each individual actuators(peristaltic pump, air pump) and sensors.

But, until now i usually think PFC only used for “Monocropping” usage, it is difference with “Farmbot”, Farmbot explain their one of strength is “Polycropping”.

I have a curiosity about your project, if you have a more experience your “SpaceGarden”, tell me about the story!

Thank you.


#8

Sure :grinning:

In 2016 I built a high pressure aeroponics system, and when I used it in comparison to my hydroponics system and soil gardening, I found that HPA was the winner by a long shot. However, it is also the least forgiving and most finicky. There is a lot more to go wrong in HPA, misters clogging, pump failure, bugs in the code, etc… And to make it worse, as little as 30min of downtime can kill your plants, there’s really no room for error. Getting the misting frequency down, and adjusting it as the plants grow is truly an art of its own if you do it right.

After building my system I came up with a lot of ideas to improve the typical way of doing HPA by solving issues and adding new elements. I ended up changing it so much about it that aside from misters and a pump it really wasn’t traditional HPA anymore. I call the new method Full Environment Flow Aeroponics (FEFA). Which is what the SpaceGarden uses. It is extremely resistant to algae growth, has multiple fail-safes built in, can change water and root temperatures quickly and efficiently, and holds only 1 Gallon of water on board at all times. But the coolest part about FEFA is the air flow patterns and methods. The first FEFA prototype has been running for about 7 months now and I have not had 1 mister clog, it has also gone months without cleaning algae from the filters.

In late 2017 FEFA became patent pending, and we got accepted into some exciting startup accelerators/programs. 2018 will be even bigger! I am very excited to show you all the production model when it is complete in Feb, I think you will love it!


#9

@mason,

It’s a perfect and kindly explanation about your project. It is really interesting “FEFA(Full Environment Flow Aeroponics)”. I think “Aeroponics” is more exciting things in the PFC. Probably between Aeroponics and Hydroponics have a strength/weakness, in my opinion Aeroponics helping makes a good design of PFC(When i think design new prototype of PFC for familiar vertical frame.).

Any reason for making PFC which using HPA methodology, it needs different recipe for efficiently way for growing plants. And it is really awesome your prototype continuing about 7 months.

Do you have any issue in the operating your PFC?

In the recent i purchased some sensitive Sensor like a PH probe/ EC probe, CO2 measurer which have a short lifespan and their maker’s public datasheet, they warned that should take a time for calibration on regularly.

Anyway i love it, your team and prototype showing up to others about the one of the possibility of PFC! :smiley:


#10

Thanks for your kind words! There has certainly been a lot of trial and error, but every day it gets a bit better :slight_smile:


#11

Hi Mason,

Thanks for sharing your project, this looks awesome! what plants have you successfully grown? and what metrics have you found are the most important to have dialed in? I’m having some trouble with my HPA system and would love to pick your brain a bit!


#12

Timing is one of the most important factors in HPA, if you don’t have your timing down then you basically just have a fancy DWC or LPA system. Filtration is critical, as well as the misting nozzles you choose. Feel free to send me a PM :slight_smile:


#13

@mason My hats off, that’s a fantastic setup you have there! Any plans to publish a BOM, release code, or otherwise make available the design in any way? This could be a great resource for people attempting to build larger systems not to mention real contribution to both the Food Computer network and the #nerdFarmer community. With all due respect, I’m curious how much of the $6K price tag has gone towards curb appeal. Congrats, and I’m looking forward to where you go with SpaceGarden!


#14

Thanks!
I have my timing at 5 sec on, 3 min off. Don’t know if that’s the reason my plants keep dying on me. They tend to stay alive for a few weeks, but their growth is severely stunted by something. I’ve tried different nutrients, nutrient concentrations, timing, light height, etc. Some roots will start to go brown, then the plant will grow some but then start wilting or turning yellow. I thought it might nutrient deficiencies, but change in nutrient concentration didn’t seem to help.


#15

Thanks so much!

It is not likely that I will publish the BOM, there are a few reasons for this. The first and foremost is that the pricing on it is dependent on my bulk pricing and negotiations I have made with suppliers. To truly emulate the product would cost well over $6k, not to mention the ~$40k of machinery and tools, and the hand full of custom parts (Like PCBs) that aren’t really attainable for most. To put it simply, the BOM wouldn’t be pertinent to most people aside from a few parts.

That being said, I will be disclosing something much better/more useful in the coming months. That is the general designs and functionality for FEFA (Full Environment Flow Aeroponics). This is the modified variant of HPA that the SpaceGarden uses. You can read a bit more about it in the comments above. I will be releasing the full designs when the patent is issued, but some more general ones before that (since it is pending). Of course anyone in the DIY community is free (and encouraged) to utilize it! It’s truly a next gen grow method and I am very excited to share it.

The exterior/appearance of it certainly has a decent weight on the price. However, the materials aren’t just chosen for looks. They are durable, industrious, insulating, and play an important role in the manufacturing and ease of assembly. It’s true that I could make it cheaper – but what I am going for is really a high end product where the look and feel are important.


#16

I’d have to know a lot more about your setup, like the mister positioning, pressure, etc… are the temp/humidity acceptable? What nutrients are you using?

5s./3m. isn’t bad, certainly not enough to kill your plants. Try 1.5s./2m. for young plants.


#17

I have a tote with misters on the sides facing in. It’s not ideal, I wanted to reduce cost as much as possible for my first prototype so I used an outdoor misting system with plastic misters.

They’re running on 40-60 psi, and I have a reservoir for the pump so the pressure is more constant, specially at the start of the cycle.

Temp and humidity haven’t been ideal since I’m in Ontario, and it’s been a cold winter, and I haven’t built systems to control them. Temp ranges from 65-75 F, with water temps at 60-70 F. Humidity has usually been around 30-40%

I’m using House & Garden Aquaflakes A/B, with a CalMag additive to supplement since I start with 0 ppm water.

Why do you think certain timings could kill plants? over watering, like you said, would just make this a fancy DWC or LPA system, which should still grow plants. That’s been my thought process this whole time, I’d rather over water than under water while I get all the other parameters right, but maybe I’m shooting myself on the foot, lol.


#18

You should for sure be at a higher pressure, atomization begins at 65psi.

The only way timing would kill them is if its WAY too infrequent. It’s pretty hard to over water to the point of death with HPA, maybe even impossible, but yes you do loose the magic of HPA when you over water.

Your nutrient choice seems fine but what’s more important is how much you’re adding, how often, and how big your tank is.

HPA is super tricky, best of luck! :slight_smile:


#19

@Juan141 I suggest looking at this thread:

@mason I would assume you’re familiar with INFARM?

I’d like to learn more about your automation system - how can your product overlap/work together with our Food Computers? How are you defining a “farm” is it one SpaceGarden, or are there three climates (arguably 3 farms). I’d like to get you/your software developers thoughts on this and some potential use cases here:


#20

I’m glad you ask :slight_smile: a lot of this is up in the air still so I’ll err on the side of keeping it overly simple and not promising anything we don’t plan on delivering. We still have a lot of research to do on the topic and this would be a lifelong commitment for the company so we’re taking it very seriously.

  • VISION
    My team and I believe that to truly access the wonders of data based climate recipes we will need a platform where grow recipes can be downloaded and shared across systems and products with as much integration as possible. Of course this would need to be open source and free to use by any system. It would need methods for distinguishing the value of the incoming data, and tossing out data deemed “bad”. Over time the recipes would become fluid by adapting to user and sensor feedback. Of course this kind of a platform would be vastly complex, and may even seem insurmountable, but we believe it can be done.

  • BEGINNING
    We are open (and would even prefer) to aim our efforts towards this at a preexisting platform, however we haven’t really found any. This has led to discussion on starting it ourselves with the help of people who envision the same goals for the future of CEA. We want to begin with a very rudimentary system that is made to work for a small handful of grow methods/systems. One of the first orders of business will be to come up with a common format for the grow recipes and returning data. Long story short, we just want to start with something really simple, probably pretty shitty, and see what we can make of it.


#21

Do you plan to share your GitHub/PCB/BOM? I’d also be interested in your HPA design. What is the status of this software stack? How large is your team to develop it?

I’m planning to try and take the MVP/$300 Food Computer stack to the “next level” and would be interested in collaborating if you’re willing.

Posts will be coming shortly attempting to establish some standards. I’d love to share as much in common as possible.