$300 Food Computer - MVP


I don’t have any personal experience with rock wool. But, based on what I read, I wouldn’t feel comfortable using it to grow food.


I made a hydroponic system several years ago and I just used pebbles that I purchased for around $3 for a 50lb bag. You need to wash them. You can reuse them as well. Should be a nice option for those worried about rockwool.


Good advice!

I’m curious @malonso2 - how did you germinate your seeds? Did you start them directly in the pebbles?

Here are some other articles as an FYI while we’re on the topic of substrate/medium @sjolliff :


Here’s what NASA has used, they’re called Plant Pillows. A little far out there, but definitely something MARSfarm is considering making. I’m curious if anyone has heard of a similar product, I’d love to see a thread started on this topic, I know @Juan1 was working with them too:
Plant_Pillows.pdf (1.9 MB)

Chieri Kubota’s talk on Substrate is phenomenal. Her speciality is with strawberry production which is really picky with regards to medium. This series of talks from CEAC is one of the best resources I’ve ever found on growing:


Initially I used a wet paper towel, but later I tried just adding them directly to the cups and that worked. This was in an outdoor greenhouse. I used a 2 tier PVC rack for water distribution, solo cups and rocks for the plant zones, and a garbage can reservoir with a pump for water. I also had airstones in the reservoir and at the start of each PVC run. It was a pretty dumb system, I added nutrients manually and never tested for PH. Plants grew fine though, I even started a avocado tree from seed to 18".

I’m just learning about this project and is there anything preventing the Pi3 being switched out for a Pi zero W? $300 still seems really high for everything, and I’m certainly curious how to bring it down further.

This is around the stage 3D printing was at in the mid 2000s. It would be great to see these grow at a similar rate.


A Pi 0 should work, but I have not tested it with one (yet). They are slower, but the MVP is not pushing the processor. Be aware that most Pi 0s don’t come with header pins, which you will need to solder on yourself (or pay for a pre-soldered version). I am using the B (and looking at the B+) because I am testing a lot more sensors and running computer vision software on top of the MVP. It is the problem of buying for what you need now, or trying to plan for future expansion.
The $300 price is if you go to the hardware store and Amazon and buy everything new. If you have scrap PVC and can scrounge a lot of the parts, you can bring the price down. We want to be honest with people about what they are getting into, rather than promising a cheap system then nickle and dimeing them with a lot of add-on costs. If you just want to try growing plants (and not collect data) you can get started with little more than some lights, then grow the system over time as your budget allows. On the other hand, you can spend a lot more if you collect multiple Raspberry Pis and start experimenting with trying different lights!!


I’m curious how 3dPonics is going for you @rbarta. It looks like a fascinating project and I love your approach of KISS. @BioLumo may be especially interested in this too.


Hey hey! Now that is cool.

Give me a day and I’ll be back with a pic of how my aero/hydro seed germination tray worked out and has progressed. I’ll be back to edit this post soon.

EDIT: okay! Here is how my Aero/Hydro (i dont like the term fogponics so i don’t use it) seed germination system has progressed. It actually works quite well. The two pepper plants got huge rather quickly and I’ve already had to move them to a larger Deep Water Culture (if i get the terminology right). I’ve eaten peppers off of one already! The other is growing bell peppers now in the windowsill. Several of the indeterminate tomatoes had to be moved out as well. And eventually even the determinate tomato (LA1996) as my shelf was just a little too short for it too (not by much). I potted that one up.

Despite this working well as a germination tray i don’t really think i really like hydroponics for long term growing. I think Soil is the better way to go for me long term maybe. But i think i will keep this system for germinating tomato seedlings and such. I did eventually have terrible algae problems so i cleaned it out and covered it with reflective tape and that has fixed it. Left some small openings for water level diagnostics. Never had to add that internal fan as i just removed one of the two piezo ultrasonic misters and that reduced the heat enough to acceptable levels. The light, the airstone, and the mister all are still on an arduino timer that turns off at night. Sometimes i put a small battery powered fan on top to stiffen the stems of the seedlings to harden them off before planting. Don’t want them weak.

I currently have Solanum galapagense, Solanum cheesmaniae, S. lycopersicoides, S. sitiens, Domestic tomato ‘Anasazi’ (one of the ones that tastes good and does well in my climate), etc. seedlings currently growing and sprouting for the upcoming summer. I have more i will want/need to germinate.



I haven’t gotten to printing any of the 3Dponics stuff to test out. Probably a summer project :wink:

Lately I’ve been modifying some of the MVP code to create a distributed MVP environment so that I can separate the parts into modules and potentially run multiple MVPs with a single head end. See my MVP build for more details

One late night coding, and just to be clear I’m no artist, I was thinking of an icon for the MVP. Now if we have an artist in our midst, then please improve on this.


Why not run each MVP as a separate unit (Raspberry Pi) and then just replicate the data somewhere (another Raspberry?) for a single data repository? That is the architecture I envisioned. The latest software/data version I am working on has a guid for each environment/MVP, that is on each data record. Trying to make the separation in the code (rather than by device) can be difficult.


@webbhm Technically that is the way it is being done. Technically each MVP will are self contained, but it gives up control to a master Processing Engine/Database while they are alive. If they go away for any reason they will continue to run independently. The main processing engine could be a full fledged MVP or just the processing engine, depending on what modules are enabled in the config. Once I get back to finishing up the code for the processing engine it make make more sense but for now see the [rough design]
(https://github.com/rbarta/OpenAg-MVP-II/blob/distributed/MVP/design/OpenMVP%20-%20Distributed.pdf). I’ll post the code once for the engine once I’m done.


Hey there, I noticed there’s only one of the led light packs left on the BOM link. Im in Canada and this is the same product on the .ca site: http://a.co/4o5WXyi It’s a lot more expensive so was wondering if there are any alternatives I could try instead? Thanks


We picked these because they were inexpensive (here in the USA) and easy to get. If you are not concerned with consistency with other MVPs and just want to get something up and running you can go with any light that gives equivalent spectral radiation. We have tried a lot of different lights, and if they have good specs, you can get about anything to work.
LEDs are changing regularly, with significant improvements. There are a lot more options these days than the short time ago when we made this choice.


I’m using Ecosmart 100W Daylight bulbs from Home Depot. They were $6.97 CDN for a 2 pack, maybe not as good as the GE but it works. Figured next time I’m in the U.S. I’ll pick up some supplies, but for now I’m good with the ones I have.



I have just built an MVP food computer in the body of an MIT food computer, and I couldn’t be happier with the result! Although I had no problem with the build, I am a complete novice when it comes to software and coding. I got everything set up by following the coding instructions, and the MVP dashboard displays all the data, but I have a few questions with the MVP dashboard;

  1. Can the color of the graph line be changed?
  2. The time seems to be off - Is there a way to change the time format when you hover over a point on the graph?
  3. Is there a way to hide old data? I have a few outlier points on the graph all the way at the bottom because a sensor was disconnected or something while I was testing the software

Thank you in advance to whoever can answer these questions, and thank you to everyone who made the MVP project a thing!

  1. The color can be changed, right now I think it is is just the pygal default. We are looking at moving all of the charting to a cloud UI, so once this chart was functional, little effort was applied to asthetics.
  2. The time is UTM (Greenwich), not local time. A decision made for consistency.
  3. Short of deleting data from the database (not recommended), you need to wait for more data and let the old data scroll off. I believe it keeps the last 250 data points.

Glad to hear you are up and running and didn’t have any difficulties with the software.


Ok, thank you for clarification!

If you do choose to improve aesthetics in the future, I think adding a desktop shortcut would be very convenient and professional. I made my own by following this guide: https://www.raspberrypi.org/forums/viewtopic.php?t=22235

Best of luck in continuing to develop this project!


I am ready to plant my seedlings and I am wondering if you want them submerged in water? The instructions said to put 3 gallons into the tubs but the 2 inch cups don’t make it to the water level at that point or as much as I have added. My common sense says to submerge, but as this is my first time I am looking for some guidance.
Thanks in advance.


I just finished putting together the MVP too. Thanks for all the help. I made a modification to the Relay.py code and submitted a PR. I noticed there was another PR waiting as well. Thanks for putting this together. My 3 year old daughter had a blast making it.


What we have found is that the airstone in the reservoir provides enough moisture to keep the cups damp (which the seeds need for germination). The action of the air bubbles breaking at the surface creates a mist that keeps things damp until the roots are long enough to reach the water.
Try it and see, as you can always raise the water level till the bottom of the cups are wet if the airstone does not provide enough moisture. A lot depends on the type of medium you plant your seeds in (rock-wool, etc.)


Thank you for the response.