Bypassing PFC, straight to (mini) Food Server


For various reasons explained below, I have come to the decision to (temporarily) take a different approach to this project that will make it more possible for me to be a part of it. I’ve decided to post this in case my approach might help others get started on this journey.

Basically, I’m turning a small cabin into a grow room, and building a flood and drain hydro setup inside. I am computerizing the climate control and automation using some combination of Arduino, Raspberry Pi, and Particle’s Photon. I’m writing all of the initial code to operate it, but I’m trying to stick with PFC compatible hardware since the end goal is to shift to OpenAG software once I have the correct hardware in place. The end result will basically be a (very) small food server (7.5’ x 11.5’), running 100% on the OpenAG platform. Of course, my media based growing system (versus the PFC’s DWC approach) will necessitate a basic shift in operation theory regarding movement of the nutrient solution.

So, here are my reasons for taking this approach:

  1. Growing Knowledge - Despite a successful soil garden this past summer, I know very little about growing. It was my first garden, and hydroponics is new to me. If I magically had a PFC in front of me right now, I wouldn’t know what instructions to give it. I would be reliant on some else’s climate recipes. You can tell me to “grow it this way”, but I am always going to ask, “Why?”. To steal & adapt an old food adage - Give a man a climate recipe, you feed him YOUR crop. Teach him to grow, and he creates HIS crop.

  2. Growing Capacity - The whole reason I got into growing was to feed myself (and eventually others). I have an ongoing goal to grow as close to 100% of my own food as I can. The “grow bed” or “grow space” in a PFC is quite small for this purpose. The increased capacity will also increase the pace of knowledge. More plants, more experiments, more data, more results. Also, my decision to use a media system as opposed to DWC should open me up to a wider plant variety.

  3. Cost - I know the team is working hard at cost reduction, but right now the cost of the PFC seems a bit out of my reach. Even if I could afford it, my approach allows me to get started sooner, and should result in a higher food per dollar ratio. The gains in capacity and knowledge make this a more efficient use of my time and money, while still resulting in the ability to participate in the Open Ag Initiative.

I’m sure this journey will result in much knowledge I can contribute back to the community. The effort I’m seeing here is quite beautiful and inspiring. Open source collaboration never ceases to amaze me, and I think we have an opportunity to take this project to great places, together.

If there is interest, I will update this topic with my progress. And, naturally, any code or schematics I generate will be freely available to any and all.

Here is a very preliminary pic.

Localhost and port forwarding

Nice. Sounds like a fun project. I’m planning on building my own non-conventional PFC out of different hardware parts than V1 or V2 plans specify. Partly to save on cost, but also because i already had a bunch of t-slot for another project that i never finished and i think it will make a nice frame that i can turn into a grow box of sorts. I’ve 3d printed most of the connectors i need and ordered a few more bolts, but i need to 3d print a few more and hopefully i will have my frame assembled soon.

I’m more interested in a grow box because i’d like a place to grow a few plant breeding projects indoors during the winter. Like my purple-seeded and red-podded pea projects.

Anyway, good luck. Look forward to your success.


Very interested and would like to follow your progress please keep posting.

I’m curious to know where you are building this at.



@Webb.Peter - I’m building this in Caribou, Maine. Our cities tag line is “the most northeastern city in the us”. We’ve got long winters, short summers, lots of snow, and bitter cold temps. The thought of year round growing makes me ecstatic. I think that even a space this small could go a long way in feeding me through the winter with the proper high intensity and/or vertical techniques.

Progress - There hasn’t been much besides planning lately as I tapped my available funds. But between xmas and extra work , I’m about to make the final push that will at least get me started.

To Do:

  • Grow Light - I was about to buy a grow light, but I have a lead on a fluorescent that my dads friend is giving away. Going to check that out first.

  • Nutrients - still researching options based on desired crops

  • Control Circuit - for starters, I’m simply building a bank of outlets controlled via micro-controller/relays. I’m using Particle’s Photon for its onboard wifi and it’s arduino “compatibility”. At first, it will only control pumps and lights. The temp will initially be controlled by the furnaces onboard thermostat. I will monitor temp and humidity via the Photon remotely (probably using the Blynk app). I’m hoping I can delay control of humidity until my next wave of cash. Then I will take control of the furnace as well. A final stage will look at sensors of pH, EC, etc… but at that point I think I’ll be positioned to incorporate the Arduino/Pi setup and start running the PFC software.

I expect to start seeds in the next week or 2. I will update and post new pics then.


You are taking the right approach in getting started growing quickly. All too often I think people focus too much on the technology before learning the hydroponics portion first.

When it comes to nutrients depending on your crop Jacks make’s a great hydro base nutrient.

What types of crops are you going for? If you plan to have anything fruiting like peppers or tomatoes fluorescents may not be intense enough to produce much. For leafy greens though it should get you started just fine. I would definitely be very careful when purchasing grow lights, just remember if something seems too good to be true, it almost certainly is.

It sounds like you have some experience with electronics, you may be interested in my DIY lighting thread. I won’t claim that they are the best/most effecient lights, but they are easy to DIY and much, much cheaper than anything available commercially.

Hope that helps,


Hello @Campergeek! I am taking similar approach as you. Last year I’ve found an old foil greenhouse in my house attic and decided I will use it to grow some chillies. I have some experience with chillies, but not with hydroponics.

So here is my progress so far. Maybe we can share some ideas.

I do not want to build complete PFC, because of complexity, cost and small size. My plan is to start with simple system, few sensors and lot of manual work and as I get more experienced I will add more sensors and automation.

The idea is to make several boxes (3 meters long, 30x30cm) with growing medium (perlite or expanded clay). Flood and drain hydroponic system. It is quite simple to build and operate.

The flood and drain needs only water reservoir, pump and some timer. I already have fountain water pump which I got for free. The 100 litres reservoir for water will be buried in ground (I think this will help to insulate from fast temperature changes). For timer, I will use my Arduino-compatible electronics I designed last year for one project.

It contains ATMega328P and it can be connected to RS485 bus. I can connect several of them to one bus and then to Raspberry Pi. Each electronics module will be autonomous, It will switch on and off pump, measure temperature of water and air, later I can add EC/pH/light/etc… sensors.

The Raspberry will be bus master and it will read measured data and check if system is working. I plan to run OpenAG brains on this Raspberry, so I can use database and web front-end. Right now, I am waiting for RPi3 to arrive so I can install ROS and OpenAG brains on it.

At first I plan to use OpenAG only for measuring. The automation will be added later. I do not plant to use light or heating/cooling in my greenhouse but I think it would be possible.

As for nutrients, I plan to use FloraMicro with FloraMato. The water will be from well in my garden (it should be quite soft water, but I have to check first).

I have time until next month, then I have to start growing seeds in rockwool cubes. I think it will take until March/April (?), then it will be time to move it into greenhouse. So I have few months to prepare everything.


@Webb.Peter - Thanks for the info! I have been keeping up with your Cobb thread, I have a feeling my lighting setup will be the element that changes and improves the most over time. I’ve been playing with cobb’s for my dad, who is using them for general shop lighting, but I haven’t yet explored them for my garden use. I’m trying to keep it simple to get started, which is why I’m willing to buy a light rather than build one. I’m curious to see what happens between the cobb lights and the GE lighting that is spec’d out in V2.0. Do you have any experience with the Erligpowht lights from V1.0? I thought about getting a few of those since I would still have a use for them after upgrading to better solution. I HOPE to grow lettuce, tomato, and cucumber crops. I’m being a bit ambitious, but I’m prepared for the trial and error approach.

@vasco - Thanks for sharing. We have a lot of similar ideas. Your buried reservoir is a neat idea. It wouldn’t work for me here as it would freeze unless it was buried deep, and then it would be to hard to work with/repair. I look forward to your progress.

Here is my latest:
I’ve started to build my relay controlled power board:

The left set is wired as 4 individually switching receptacles, while the right set switches in pairs. So 6 relays will control the outlets (timed switching for pump and lights, mostly). That will leave 2 channels on my 8 channel relay board for the eventual low voltage switching of the furnace and de-humidifier. It will all be controlled by the Photon so I will have remote monitoring and control. Remaining pieces are arriving Monday, including the AM2315 temp/humidity sensor. With any luck, I’ll be starting seeds next week.


No, we have a PFC V1 here in STL but have only used it to grow lettuce. Not sure if the lighting would be intense enough for anything else. What I’ve noticed is that while you can get plants to fruit, it’s hard to get produce that is large enough to sell commercially with indoor lighting. That isn’t to say it can’t be done, but it takes a much more intense light than most leafy greens.

Very curious to see how it turns out, I know part of the goal with V2 is that you can turn on/off the red/blue/white lights and perhaps even dim them as well. Please keep us posted, love to know how it turns out!



It’s been two months since my last update here, and a LOT has happened! My system has been fully operational for several weeks and I’m quite pleased with the results. Today I started shifting from vegetative nutrients to flowering nutrients, and scaled back my grow light timer. Here are photos of the progress…

The brain:

The first seedlings on January 31st:

The transplant process:

The monitor/control interface running in the iPhone Blynk app:

And what the plants looked like earlier today:


That’s awesome! I like your pictures.


Very interested in your integration to the Blynk app.

Could you provide any more details on that?


@Webb.Peter -

You can read a write up of my system here:

But here is a breakdown of Blynk, specifically:

So Blynk is an app available for iOS and Android. They have libraries available for various platforms (Arduino, Raspberry Pi, Particle, ESP8266, etc…).

You simply #include the library in your code. The app provides an auth token to put in your code to authenticate communication.

The Blynk site ( has docs on how to integrate, but here’s the gist:

  • You set up your own interface choosing from “Widgets” (buttons, LED’s, LCD’s, displays, graphs, etc…)
  • You can communicate with hardware pins, but Blynk also allows “virtual” pins that can control hardware pins through code, which greatly expands options.
  • You set each widget to the hardware/virtual pin you want to read from or write to.
  • There is no code to write in the app, and the code on the hardware side is quite simple. The Blynk site has pretty simple docs, and links to example code.

The app is free, but has in-app purchases. Basically, each widget costs “energy”. You start out with so much, and can purchase more if needed. The cost is low. I think I’ve spent about $6 on “energy”, and I run TWO systems on this platform (the garden and my campers thermostat:

I have no affiliation with Blynk, so have no bias to promote it, but it’s served me well. I actually discovered it here on these forums from another user. I’m weary of putting so much stock in it. If it gets purchased by a home automation concern, it will likely disappear or become proprietary. Crossing my fingers on that not happening.

Make stuff,


@Campergeek Your RoboGarden blog post is great. I’ve come to a similar conclusion that building a food computer according to the MIT design is not practical for me. I’m still excited about sharing knowledge about simpler DIY hydroponics automation and monitoring techniques though. Blynk sounds interesting–I wasn’t aware of it until today.


The garden is going well. I was pretty ambitious to grow three crops in one system, which invited extra challenges and failures, but I learned a lot in a short time as a result. My next crop will be tomato only, so I can focus on one set of nutrients and lighting schedules.

I have harvested my first lettuce. It did not grow as big or abundant as the same seed did in soil last summer, but it tasted better. Even as the plant began to bolt, it was not bitter.

The tomatoes are flowering. I waited too long to trim them, so they are pretty unruly. I also should have switched to the flowering stage earlier. I may actually scrap them and start fresh.

Here is the circuit and garden in action.