Let there be light / Making Progress with the PFC v.2.0


Our collaborative PFC (Personal Food Computer) project with Oregon State University engineering students is coming along great! The ESE (Energy Systems Engineering) students gave a fantastic Capstone presentation, we’re ahead of schedule, and getting ready to test the electronics. We plan to have a fully operational PFC ready for receiving our transplants by January 2. From there, we’re planning to use our data gathered from an energy consumption analysis of the system, and build in a model so that we can monitor the energy usage during each growing cycle. Our hypothesis is that, even in a small chamber such as this, the energy usage will be the greatest with the heating and cooling systems.

Requirements for controlling the LED Panel

Looks great!

I’ve gotta ask because the title of your post, what lights are you using???

Keep us posted and let us know if you have any problems with the build. I would love to see that capstone presentation!

Where are you at with the software build?

What are you using for your cooling system?

Lastly, do you have an intended recipe you plan to use? Any experiments we can try to run in parallel?


Thanks so much @Webb.Peter. After a bit of consideration, I decided to use a great GE Arize Life lighting system that I purchased from Hort Americas. The modules weren’t very expensive at all, and are calibrated for growing leafy greens. Here’s the spec sheet on the lighting system we’re using: GE-LED-HorticultureArizeLife-DataSheet.pdf_tcm201-113499.pdf. The lights will output around 12 moles / photoperiod of 18 hours, which is plenty for growing leafy greens.

I’ll definitely do my best to provide updates of the build process. I’ll talk to the engineering team for sure, and get their permission to perhaps generate a Youtube video of the Capstone presentation that they gave last week.

The software installations have been going great. Our next step is a week of testing to make sure we can control the high and low-level devices. We’ve had considerable support from our partners at E-Space Labs here in Bend. They’ve provided build space and wiring tools, and they connected us with a technician who built the Signal Board. The ESE students also received a lot of support from an EE who has been instrumental in helping them with programming the Raspberry Pi, assisting with the Couch software installation, running the Python scripts, and installing the UI & ROS.

With respect to cooling, we’ve decided to try our hand at a passive cooling system for the air temp, and we’re integrating a DC aquarium water conditioner for controlling the water temp. We’ve discussed the benefits of having a compressor-driven cooling unit, but certain factors influenced our decisions to go a different route. Plus, we just may find that a compressor-based AC unit isn’t needed in a chamber this small. The game really changes when you increase the volume of space you’re using, and pack it with higher density crops and lots of electronic systems that generate a bit of heat. We’re confident that we can get by with a passive cooling system, and we’re keeping fingers crossed that we can do it.

We haven’t decided on a recipe that we’ll use, yet I imagine we’ll be getting there soon. I’ll be getting back in touch with my contacts at Hort Americas to glean some climate suggestions, and there are plenty of resources for determining optimal temp and humidity ranges. I’ve seen the climate recipes in Github, the engineering students we’re working with are aware of the recipes, and we all will to try to figure out the best “run-module” for our first growth cycle.

I’d love to discuss possibly doing a parallel experiment! I’ll definitely let the engineering students on our team know, and I think they’ll be able to come up with a more definitive experiment that they’d like to try out. I’ll be sure to let you (@Webb.Peter) know via e-mail as soon as I know more.

Finally, I am very impressed with our Oregon State University ESE (Energy Systems Engineering) student team. The associates in our organization are also quite delighted to be working on this project, and we are grateful to the OpenAg Initiative for providing a great open-source platform that has become quite a success. I am optimistic that we’ll continue to do our part to help bring community-focused agriculture, nutrition, and sustainable food systems into the fold. Be well & please keep in touch!


Looks great,

Is your team donating code changes back to the Openag code base?

BTW: I’ve forked the Arduino code and produced a version that allows low level debugging. My results are here (https://github.com/ferguman/openag-firmware). If you are interested in using it let me know and I’ll provide some support to your team.


Greetings @ferguman, Thanks for commenting on the post! With respect to donating code back to the OpenAg repo, I imagine our engineering team will definitely follow that protocol if they tweak any of the existing code. Our engineers are not software developers, so there’s quite a learning curve for them in terms of even utilizing the existing code that’s available on Github. The steepness of the learning curve will increase for them as they learn to adjust code for specified parameters.

We fully support the spirit of open-source coding, and will be sure to follow suit once we get a good handle on understanding how the scripts work with all the components of the PFC system. Once any changes are made to any of the open-source scripts, we’ll be sure to add them to the corresponding repo.

Thank you so much for sharing the low-level debugging for the Arduino and offering support. I’ll definitely be sure to pass this information along to the engineers on our team, and I’ll direct them to you if they find a use for the Arduino debugging firmware you provided.


Very cool! @ATBFarms, how are you categorizing energy consumption? Besides heating and cooling systems, what else are you independently tracking? Are you including the lights as a source of heat? Would love to see data visualizations or any info you glean that correlates specific energy usage to specific plant growth characteristics. Looking forward to updates.


Hello @Drew, The OSU engineering students on our team will be the ones to formulate the energy consumption model. As I know more about it, I’ll be sure to provide an update for the analysis on the forum. The lighting system we’re using isn’t a significant source of heat, as the low-wattage LED modules are industrial grade GE fixtures. I like your idea of correlating energy usage to growth characteristics. I suppose that’s the idea behind “climate recipes”.

One area of interest that we’re working on is correlating climate parameters with antioxidant presence at different stages of the growth cycle. That takes a bit more of an analytical set-up than we’re able to afford at this time, but we’re hoping to eventually conduct an antioxidant assay at some point next year. We have access to a Food Composition Lab in Corvallis that can perform the ORAC assay that measures antioxidant capacity, and we’re already looking at the Synergy instrument that will perform the assay. I eventually want to be able to conduct these assays within our own organization.


Based on the output they specify, do you have a estimate of how many of these would be needed for an 18" deep system with lights mounted from 18" to 24" above?


Good afternoon @pspeth,

We currently are running 3 of the 24" GE Arize Life light fixtures. The customer service reps are really great at Hort Americas, and they use some pretty great software to prepare a “light plan” for the dimensions of your growing system. I think the height of the light modules above the raft system in our PFC v.2.0 is around 18" or so, and the output on an 18-hour photoperiod is around 16 mols of light / day. If you use two of the fixtures, you’ll get around 11 - 12 moles per photoperiod, and an average light intensity around 169 umol /m^2 /s. That is plenty of light for growing leafy greens (lettuces, kale, microgreens). We are getting a pretty good growth rate with the lighting system, as they are industrial grade.