Andrew's Custom T-slot Frame (Food Cube?)


Not sure what category to post this in. Is there a way to have multiple categories?

I guess this is a build diary of sorts, it’s also an alternate hardware frame prototype and on-the-fly and subject to change. This is sort of an offshoot to my original thread titled “Alternative Hardware” or something like that.

Anyway, here is the current progress on my custom T-slot cube that i hope to turn into an indoor grow box at minimum and a fully functional PFC potentially in time. Who knows, perhaps some ideas and/or innovation may happen in this thread or with this design that could improve or change future PFC designs. PFC Version 2 hardware just came out and it looks like its a big improvement over 1.0. My choice of trying to build an alternate frame out of commercially available T-slot instead is based on two factors. 1. I really like t-slot and i think if it’s used well it can really simplify things. 2. I already had most of the t-slot around from a previous unfinished project which was to be a large custom 3d printer.

Here are a few photos. More photos to come as i continue to work through this design. I’ve gone through a few different ideas on 3d printed connectors for this, many of which i am currently using. Pardon the messy floor, i’ve been meaning to clean organize that for awhile, along with a few other areas too.

So what do you guys (or gals) think? Too simple? Too ugly? Just right? Is this a design or material you are interested in? Should i keep working on this? Other thoughts? Ideas to make it better?

Well that’s all i can think of to say for now. I wish it was more along, but hopefully we will get there soon. I also wish i had better pictures, but that’s all i’ve got right now.


Food Computer - Gipuzkoa style
3D printing viability

Yay! I’was thinking the same as you. I come from the 3D printer world too and the t-slot type is imho the optimal solution to build these kind of projects.

I’ve spent some time studing the FCv2 BOM, and the tube structure was the first part of the design I was thinking to change. Mainly for two reasons, the first one is because I’m from spain, and some of the material listed at BOM is not available here and the t-slot is easy to find, and the second one is about the flexibility that these kind of structure gives. In terms of costs, there’s no much difference at all.

I was studing to update the CAD project with t-slot… do you have a custom design for the FC?


I was studing to update the CAD project with t-slot… do you have a custom design for the FC?

If you mean for the whole FC, then no i don’t. I’m just making this up as i go. But if you mean have the CAD files for what i’ve already built, then yes. The connector parts are all 3D printed and designed by scratch by me to fit my 8020 brand 1010 and 1020 size T-slot. The CAD files are in Solidworks and a full Solidworks Assembly file could easily be made.

The connector pieces have already been uploaded to Thingiverse, and you are welcome to use them or adapt and design ideas into your own parts. I imagine you might not want to use 1010 and 1020 size t-slot as it is not metric. But the same ideas should work, they just would need to be resized.

I’m currently thinking of doing something different for the bottom corner connectors instead as they may add too much space height wise. So this design in it’s current state is subject to change.

Here are the appropriate thingiverse 3d printer files as they are so far:


Thank you so much!!! These kind of connectors are the ones I was thinking to use in my project. As you said, I need the metric ones, but since you uploaded the solidworks parts to thingiverse, I can update them easily.

I’ll try to document the evolution of my idea, hope to see yours soon, looks amazing.

Thnx again!!


No problem! Thanks for your interest and work on a similar project. I look forward to seeing what you come up with as well.

I think i may have the design for the new bottom bracket ready. This is what i’m thinking of using now.


bump. planning on resurrecting this thread. Finally printing (or reprinting) my t-slot corner brackets. They seem to work well. They help make a more accurate square compared to the last corner pieces. Expect more updates to follow soon i hope.

Oh, and here is the STL file for that bracket mentioned above. I improved it since then and might improve upon it again.



This is my recent Prototype Frame Design of t-slot material, using google sktetchup.

I heard this t-slot could assembling without any processing such as make a hole or tapping.
Using this parts.(I have not any experience making t-slot material)But it seems easy to assemble if i making accurate design.

I would using top of inner ceiling to placing brain modules, and right side used to placing PH/Nutrient/Water and peristaltic pumps.

But in my opinion, @BioLumo, if i print some joint feature using 3D printer, it is more strong!


@house, Wow! this is fantastic! I am really loving your design! I am VERY interested in seeing how your design continues to develop. I really like the look and ease of putting the t-slot frames together over the traditional Food Computers do date thus far. I kinda hope the t-slot becomes the standard frame element for PFC in the future. But maybe that’s just me.

My “Food Cube” is not necessarily meant to become a true PFC, but it could. For a real PFC i like your design much better than mine. But perhaps we can continue to help each other develop and improve each other and collaborate on this topic more! My 3D printed connectors are based off of 1" 10 Series 8020 brand t-lot size 1010 and 1020, but the designs could be minorly modified to fit a better t-slot metric standard. I dont know what size t-slot is commonly used in some 3d printer designs like the Lulzbot TAZ. 20mm size t-slot? Is that a good size to use for a Food Computer? I don’t currently have any of that size to play around with right now.

More updates to come from me soon on my Food Cube. I need to print a lot more of these angle connectors for mine tomorrow. I have integrated a standard 1/4"-20 thread into the plastic angle bracket that serves many functions on the cube. The 1/4"-20 bolts are standard size bolts for the 1" t-slot so it works out quite nicely. The coarse threads print out just fine and are usable without any “rethreading” or using a tap to clean them out. A smaller thread size would probably not work as well. If converted to metric it would be wise to stick to a larger coarser thread size if this feature were to try and be replicated / preserved.

EDIT: you are right that t-slot is easy to assemble and requires minimal machining or tool knowledge. On my Food cube i did however tap the end extruded holes with a 1/4-20 size tap to accept the screws in the corner cubes that hold the frame together. This was a minimal tapping effort though as i did not need to pre-drill the hole i was tapping. The pre-extruded hole was just about right to tap 1/4"-20 already. I suspect that was actually by design of the 8020 tslot engineers.


Thank you, @BioLumo
I understanding what we are trying through this thread. I agree with your intention when you are started!

Previous my first PFC based of “PVC PIPE”. It is good to making frame and easily assemble. But after cutting PVC PIPE they have a some gap and weak to support weighting. I considering various usages.(Like a aquaponics or aeroponics). And i prefer assembling without tapping because i want redesign and recycle it.

Anyway, I purchase t-slot 3030 size(30mm). In South Korea t-slot divided two types.(light and weight types). I ordered light types 3030 t-slot, bracket and t-bolts. (to making previous posted design, and i paid total cost approximate 110US dollars.)

I expect i can take t-slot on tomorrow and i will post how to making new design using t-slot!

And i really like your strategy using 3d print to assemble t-slot. But in my personal environment i couldn’t sourcing 3d printer : (

Keep sharing your “Food Cube”! I like your idea too!

2017.12.7 udpated.
Today I can take a t-slots! I consume 2 hours to assemble it alone. But it is really strong and no gap between t-slot more than PVC pipe!


thanks. more updates to come soon i hope. :slight_smile:


Brief update. Got some more threaded angle brackets 3d-printed. Now i can start drilling some holes in this tile/drywall stuff. I got one piece (what is presumably the top) bolted on. I don’t like the tile that much, so i may want to see about the stuff the other designs are using for walls…

But, hey some progress is progress anyway. Hope to keep the updates coming. Maybe i’ll also mount this LED grow light array on the bottomside of this panel in the next day or so and get more of the sides bolted on. I may need to order more t-slot nuts and 1/4"-20 bolts because i think i’m almost out again.

But my 3d-printed brackets with 3d-printed threaded holes works quite well for mounting. No tapping required. Easy peasy.

Happy Tinkering!


Okay. So it seems i went a little bit down the rabbit hole. Turns out those huge brackets were not needed and infact made it so my t-slot frame was not square in the middle to mount the tile panels. I guess at some point i knew the frame would not be perfectly cubical and for some stupid reason i thought this meant the middle would be the same as well. This is not the case since for the bottom i am using double wide 1020 size t-slot which is thicker and heavier but this also adds an inch to the bottom. If i was using all 1010 size t-slot all i would need would be the small cube connectors like on the top. So i have gone back to using my modified custom corner brackets that i had designed earlier. And as a bonus i already had them printed out. So yay! Lucky me. i had a sneaking suspicion i needed those!

More photos to come soon. P.s. i kinda am hating these tile panels right now. I need something better for the walls. What are those reflective walls people are using in the other PFC designs? Some sort of insulation with reflective coating?

P.s. mounted the light. Hope to have all the side wall on tomorrow and take a photo. Good or bad.

and i am starting to experiment with hydroponic / aeroponic growing.


Reflective mylar is what we are using. You can buy it in rolls at most hardware stores or on Amazon under the name “Reflectix”.

You might also want to paint/cover the exterior of your reservoir. I know it’s really tempting to use a transparent tub, but you’ll find it ends up being bad for your roots. This will often lead to root rot, or just lots of algae and other growth in the reservoir. Best to use something opaque, paint it, or cover the sides (aluminium foil/mylar works great).


yeah, i will. Algae will be of concern. I’m not doing anything with it yet.


Quick preview:
I got rid of the paneling from HomeDepot that was basically drywall i tried using before. I ordered new custom 26" X 26" paneling, in this case PVC panels 1/8" thick (was going to order 1/8" white acrylic that didn’t let any light through, but my main plexiglass supplier was slow, so switched companies and PVC was cheaper from them.) Looking good so far. Think I’m gonna try to buy some easy removable hinges for the door. More updates soon. Plan on using this for upcoming winter tomato breeding projects.

Was planning to get back to this for winter wild tomato species breeding, but was inspired to pick up the pace by Food Computer - Gipuzkoa style


Thank you for the comment @BioLumo

Happy to see you recovering the pace.

About harvesting in winter … I will try with pepper, a picture of that will be epic ! :sunglasses:



Another preview: i got the LED light mounted on the new paneling. Will have to wait until i receive my hinges and can rig a door up. I got both another white panel and amber plexiglass for UV filtering. Not Sure which i will like better.


Finally got the hinge to work the way i wanted. Took much longer than ideal, but it’s done now. I ended up designing my own custom 3d printed hinge as none of the standard hinges were exactly what i needed or wanted because of several different factors. I plan on publishing them on thingiverse in the near future.

Next to figure out a magnetic latch and a handle.
My 1/8" amber transparent plexiglass broke when i was trying to drill for hinges as i was not careful and forgot how brittle it can be at that thickness. I might either try again for a clearish door that looks nice or think about seeing if 1/4" plexiglass could work.

But for now i have a door out of the white pvc panels like the walls. And actually a solid door will reflect more light back in which is better for plant growth.

But making progress.

I Also need to think about adding some in/out vents with either AC or DC fans. I’m thinking of some sort of rubber strips on one side of each fan to help reduce airflow when the fans are off and not in use for climate control.


I’ve already used my food cube to help extend the season for a late planted squash. The squash has now been harvested. I am now using it to try and grow out some seeds for my red podded peas.and maybe some tomatoes.

I have designed and 3d printed a handle for the door. It has been published on Thingiverse. I will install it soon, along with what i hope will be a magnetic door latch that will work. I have the lights on a timer and also plugged in a USB powered fan that also turns on with the light. Since i do not have vent / fan holes yet i have to currently keep the door cracked a bit to have good airflow for the plants.


I’ve drafted up a rough copy of my food cube in Solidworks for those who are interested. :slight_smile: