Building, growing and designing in Riggi's "garden"


Hi everyone,

Some of you may or may not already know me. My name is Thimon Rotstejn, also known as Riggi by some of you already (you may call me either really) and I’m an Industrial Product Design student. I’m currently working during my internship on a Food Computer. I’d like to make use of this thread to post updates of my work (whatever I’m allowed to share of my work), to gather feedback and work together with you guys (and ofcourse document my work in the meantime).

Now to start this off, as I still am missing some of the parts however, I couldn’t really just wait around and do nothing so I decided to just write my own Arduino code with the components that I have. This way I don’t waste any time and I can get started with growing and designing as soon as possible.

The Arduino code that I wrote for this temporary Food Comuter tracks the time, gathers information from the water temperature sensor, ambient temperature and humidity sensor and from two water level sensors. The code also is able to control both the heat bed (to adjust the water temperature whenever necessary) and the light panel. Both the time and all the sensory data is currently displayed on an LCD with a very basic UI. Some footage of the LCD and of the water level system.

Now for some more pictures of everything!

A general overview of the electronics and the sensors.

The heated bed I’m using to heat up the water and the lighting.

Now it’s time to just put this all in a frame and start growing. The first thing I’ll try to grow is corn salad/mâche/etc. (scientific name: Valerianella Locusta). If any of you have grown this kind of lettuce and have any advice or if you have some general advice or if want to give me feedback or ask me any questions, they are all very welcome!


It’s been a while already but it’s time for an update! Last couple of weeks, I’ve been busy with assembly, cable management, coding and ofcourse also some troubleshooting and at long last, it’s finally up and running.

The frame is a simple wooden frame built by the company itself. In terms of measurements, it’s very similar to the original frame. The only key difference is that the doors are mounted on the sides instead of on the front to be able to mount the doors easier.

Now for the lighting I’m using two Erligpowht 45 Led panels from the PFC1 who have been slightly modified. The sides have been cut off, internal power wiring connected to each other and the backplate has been replaced with one solid piece of MDF. In terms of lighting, I’ve read here earlier that they aren’t that great so this is a component that probably will be replaced by something else in the near future. What kind of lighting solutions do you guys have? I’ve read about your COB LED lighting solution earlier @Webb.Peter, how is it working out?

Now for the electronics department… Well let’s say cable management could still be improved massively, but it’s a start. The relay board however seemed not to be working for some reason. I’ve noticed that there’s a capacitor a bit loose, perhaps that has to do with it. I’ll do some more testing later today or one of the following days to verify if it’s actually broken or not and it wasn’t just a bad connection somewhere.

Earlier today I did ran into a small issue concering our power supply as I currently had it mounted on the side with some double sided sticky tape, but as you can see, it came loose and took some cables with it unfortunately. I’ll have to fix this later today.

Currently the Food Computer is germinating lettuce using a heated bed and some humidity covers. So far the water is being refreshed twice daily to avoid still water, however with this heatbed, this is not really the ideal solution. I’ve read on the OpenAg wiki about this mini seedling germinator from Rob Baynes and I’ll probably try to alter his design and integrate it inside the Food Computer itself as there’s no need for me to germinate a large amount of seedlings beforehand. Do some of you already germinate his/her seedlings inside the Food Computer itself? And if you do, how do you do it currently?

As I’m still missing some components, the integration of the Raspberry PI with the openag_brain software will have to wait a little longer. In the meantime I’ll look to make better use of all the space inside of the Food Computer and I will most likely hack this heating bed a bit for a larger grow platform.


Great documentation, looking forward to seeing where this goes

  1. I’m curious why you chose to run an LCD screen? Is this just because you have no other UI for output?
  2. It seems like you have experience with Arduino’s, do you have a plan to integrate a Pi/OpenAg signal board?
  3. Do you have a way of testing the lights? @ferguman Do you know what the V1 fixture light output is?
  4. @Riggi I am a big fan of full spectrum COBs. Here’s what you’ve got to keep in mind: horticultural LEDs are designed for large scale applications. This means they are focused on energy efficiency. Most of the companies producing them are startups, or spending loads on R&D and therefore the lights are very expensive. They are also in much smaller batches than other forms of commercial/residential LEDs. I’ve found since my goals aren’t energy efficiency and are more focused on initial capital and output athat full spectrum COBs are a no brainer. Downsides are you can’t control for spectrum (I don’t think the V1 light can either). 4 full spectrum residential E26 LEDs using full spectrum COBs. We’re getting 270-300 umol m2s at 18" with them. The COBs I show in my first forum post put off about 3x that.
  5. @jimbell has also been working on a germinator. I’ll be honest, I germinate inside a plastic bag almost always. It’s just too easy, plus it keeps all the seedlings at an even height. Then I’ll keep them in a seedling tray till about 10 days, then I move them into little seedling cups, then at 21 days I transplant into the main grow system. This method is most efficient, but you can also just do entire growth cycle inside of the PFC.


Thanks! I’m curious to see where this goes as well honestly. My internships ends at the end of this week though so officially I continue with this project in februari for my bachelor thesis. I will however continue to do tests and grow as much as I can in the meantime so that I’ll have enough data to know where I have to work on. Upcoming updates/logs will most likely be concerning the Pi integration and the OpenAg signal board and growing in general. Maybe there will also be some design related updates in the meantime but I can’t promise anything.

  1. This is exactly the reason why I’m running an LCD screen. Keeping it connected to a computer all day with the serial monitor on wasn’t really an option. I thought about keeping it very basic by using RGB leds to show several possible states (for example temperature too hot = red, too cold = blue) but I just decided to use an LCD screen I still had laying around as this is more accurate than color coded data (although working with colors to display data could be very interesting in terms of user experience for unexperienced growers).
  2. Yeah I’m planning to use the OpenAg signal board along with a Raspberry Pi later on, but I’m still missing the OpenAg signal board. I’m going to make one myself at school by photo etching it with the equipement we have available. I’ve already got all the components of the board and I’m just waiting for the board itself to arrive so I can begin etching. How I’m going to solder everything, I’m not entirely sure yet but currently it seems like I’ll be soldering it by hand (which will probably be a big pain in the ass).
  3. No unfortunately I currently don’t have any way of testing the lights. Maybe I’ll be able to get my hands on something but currently it doesn’t seem like it. I did look around a lot to see if someone knew what kind of PAR it was giving, but I couldn’t find anything.
  4. Oh that sounds very promising then actually! The V1 lights can’t be controlled either although currently that doesn’t worry me that much actually. It’s mainly the PAR I’m concerned about. I am interested how big the difference is in energy efficiency. I’ve already looked at quite a few lighting solutions although that actually doesn’t include full spectrum COBs. Maybe I’ll try to DIY one myself and see how it works. The upcoming GrowBlade could also be an interesting option.
  5. That’s actually a pretty interesting way of germinating. If it works, it works really! Do you happen to have any images or footage regarding your germination setup? I’d be interested to see how you’re doing it!


The version one FC light panel was decent. I think it output about 170
umols/m^2/sec at the grow level using a PAR meter to do the measurement.


Oh thank you @ferguman! I actually expected less from the PFC V1 light panel. Maybe I’ll be able to stick around with two of those Erligpowht Led Panels longer than I expected. I guess I’ll just see how growing my lettuce goes with them.


It’s the last day of my internship here at the company today and now the question of course is what’s going to happen now, in what direction will this project go? Currently, it seems like we’ll be going for a mid-end system, somewhere in between the MVP and the PFC. Now there is a catch though as I’ll have to gather as much data as possible the coming months. This means I’d basically need to grow as much as possible and I highly doubt I’d be able to grow that much with only one Food Computer.

In terms of sensors, the Atlas Scientific pH and EC sensors are quite expensive on their own so grabbing multiple of these isn’t really an option. After having read this discussion in terms of growing efficiently and a talk with my mentor (it seems like I might be able to get an area at school to grow), I may have found a possible solution. The idea is to work with several growing areas and to grow different kinds of plants in each growing area. As plants differ in terms of needs like temperature, nutrients, … and as I won’t have the space inside of the current frame to grow this many plants, I’m thinking of designing a mobile sensor unit.

(Visuals are purely for explaining the concept)

The mobile sensor unit would be able to communicate with these growing areas and order them to adjust the nutrients/pH accordingly. This way I can reuse the same electronics/sensors for multiple growing areas, save on costs and grow more. Maybe it would be interesting to be able to easily customize the growing area along with the lights in terms of height and length by placing/sliding them in different slots.

The interesting aspect with this though is how it could be quite easy this way to enlarge your growing area. I’d only need to buy/make myself a new growing area and throw in some basic sensors like water level, water temperature (and maybe a camera as well) along with some pumps (unless it works based on the Kratky method).

Now what I am wondering though is if this model would work on a consumer level as well. I am aware that something like this isn’t really a valid option large scale as in that case, it would be much more profitable to work with one large growing area, but for small indoorn and personal farm this could be an interesting solution, especially for people that want to expand their growing area later on and grow a variety of vegetables. How often do you guys or the Food Computer have to adjust your pH and nutrient values? Is it every few hours, daily? Cause if it’s something that has to happen very frequently, I doubt this concept will work. Does anyone have a clearer picture of the viability of this concept or have some general input or feedback?