Building Prototype Planters

Hello Rooftop Farmers everywhere!

This is the first real update to what will probably not be the most-often updated blog in the world, but will provide a general account of the 2009 growing season’s effort to pull together a Philadelphia Roofotp Farm, which, because the acronym works, will heretofore be called “PRooF.”

So far this season several of us including Shy, Jade, Steph, Ryan and Andrew, have gathered at my house in West Philly to brainstorm and conceive everything from planters that will be great for rooftop farming to structural systems that will span rowhouse roofs and elevate planters off of roof surfaces to pulley systems that will conquer gravity and lift farm materials up into the sky.  We’ve also talked up the project at an urban agriculture-oriented Urban Sustainability Forum and made a lot of great contacts.  There has been much progress.

Over the last couple weeks Steph, Ryan, Andrew and I put our minds together to design a prototype rooftop farming self-watering planter.  I recently put the finishing touches on one and planted a bunch of tomatoes.  Here are details.

The goal:
To develop a light, inexpensive, self-watering planter than can be easily built and will function well in rooftop farming applications.  I built two 8′ x 2′ planters.

The process:


wood for PRooF's self-watering rooftop planter 2009 prototype

wood for PRooF's self-watering rooftop planter 2009 prototype

The planting tubs I used last season in the garden on my roof were made out of plastic Rubbermaid tubs.  They worked well, but from the beginning we decided we’d like to avoid using plastic.

We first thought about building our prototype planters out of wood salvaged from shipping pallets, but then abandoned the idea when we realized that the pallets were heavy — really really heavy — and difficult to work with.  So we decided to try thin, light wood to frame out the planters.  In the photo you’ll see enough wood for two 8′ x 2′ planters — four 8′ 1×10″s and one 8′ 1×12″. I bought relatively cheapo wood, not pressure-treated. It will rot over time, but won’t run chemicals into the soil.


Next, I cut the 1×12″ into four 2 ft long pieces to use as the short ends of the planters.  (why 1×12″?  aha!  You’ll see later):

8ft 1x12" cut into four pieces to use as prototype planter ends.

8ft 1x12" cut into four pieces to use as prototype planter ends.

Next, using my highly advanced carpentry skills, I screwed two planters together using exterior screws:

Frames of PRooF prototype planters, 2009

Frames of PRooF prototype planters, 2009


The next thing I did was very clumsily apply stain to the wood, inside and out, with the idea that having some stain on the wood will make it last longer. We’ll see. I forgot about the staining operation until I had the planters framed so I didn’t have proper stain-applying tools or even proper stain. I used the leftover remnants of the stain I put on my backyard fence several years ago. It came out splotchy and pretty ruddy-looking, but no one’s looking (except you).

Clumsily stainted PRooF prototype planters, 2009

Clumsily stainted PRooF prototype planters, 2009

Fancy schmancy.

At this point the sun was shining and I got hot and took off my shirt. Note to self: next time wear sun block.

shirtless farmer


Next, I took an 7′ x 10′ pond liner (epdm rubber) I bought at Lowe’s, cut it in half and stapled each half inside one of the planter frames to serve as the water reservoir.  We chose epdm because it’s strong and impermeable and is supposed to not leach any chemicals.  (we used pond liner rather than the epdm used for roofing because that epdm has UV-coating that could come off — thanks to our Milwaukee Rooftop Farmer friend for the tip.)  After I took the photo below I stapled the pond liner up along the insides of the frames.

Pond liner stapled into PRooF 2009 prototype planters

Pond liner stapled into PRooF 2009 prototype planters


The next step was to staple permeable fabric (I used weed-blocking fabric from Home Depot) around the frame to hold the potting mix. I hung the fabric over the top of the frame and stapled it on the outside of the frame so the weight of the mix would hang on the frame.

The intent was for the fabric to hang down about 7″ into the frame. Later  I drilled a hole about 4″ up the side of the frame. Logic has it, a couple inches of the mix will hang down into the water and interact with it, allowing the plants to wick the water up when they’re they’re thirsty. We’ll see if that happens.

A  likely scenario is that the constant connection between too much water and the mix, without enough aeration, will lead to root rot.   In fact, I’m kind of sure that will happen.  So what to do?  Right now, with this first generation prototype design, cross our fingers.  In the future, we’ll factor good aeration into the design and actually make things work well.  Admittedly, this is a beta version….

Why did I break the fabric into two sections in each planter? Read on, curious farmers.

Permeable fabric hanging inside PRooF prototype planters, 2009

Permeable fabric hanging inside PRooF prototype planters, 2009


For the potting mix I used the special formula concocted by master container gardener Ed Smith. Basically, it’s 1/2 compost and 1/2 a mix of peat moss, perlite and vermiculite, with some minerals thrown in to sweeten the deal (greensand, lime, phosphate, dried blood and axomite). [[There is debate over whether peat moss is an environmentally sound material to use. It really seems to be the best thing to use in self-watering containers because of the way it retains moisture. But it’s a very very slowly renewable resource — it renews over hundreds of years, if at all — and is therefore being depleted. I’ve read that less than 1 percent of available peat land is being used for commercial peat-ing, and that the peat-ers are trying to renew the resource faster than they use it — if the don’t call it “repeating” they’re sorely lacking pun abilities — but it’s still going away. I decided to use peat this year, though slightly less than Ed suggests in ratio to the compost, and try to see if there are alternatives for next year.]]

Each planter took about 1.9 cubic feet of peat moss, 160 or so pounds of compost and a bunch of perlite and vermiculite to create air pockets.

Anyway, here it is:

Potting mix in PRooF protoype planters, 2009 -- hurrah!

Potting mix in PRooF protoype planters, 2009 -- hurrah!


Covering the planters with plastic enables the mix to become a closed system — the plants drink from the reservoir when they’re thirsty but the water doesn’t evaporate. This can theoretically extend the growing season because the soil will warm up faster, but that may also kill things off because of extra heat. That’s all part of the experiment.

PRooF Prototype planters 2009 covered with plastic

PRooF Prototype planters 2009 covered with plastic

Why did I use a fancy brown tarp rather than plastic sheeting? To waste money, I guess. [At some point I’ll calculate how much this all cost, but not tonight.] Maybe one place we’ll do research is to find out for sure which permeable coverings will be best — black on top or white reflecting? plastic or something else? what covering will leach stuff and what won’t…?

And now, the answer to the question of why the 1×12″ boards and why the fabric was broken into two sections. With the 1x12s on the ends and 1×10″ sides, the plastic angles down into the middle. When rain comes, ideally the water will run into the middle, where I put the fill hole. Theoretically these self-watering planters have just become super self-watering. (There was a big rain the day after I build the planters and I think that worked.)

By this point I was super-exhausted. Luckily, planting was easy. I bought some late seasons starts from Greensgrow, cut holes in the tarp and put the transplants in. Voila — we have a farm:

Voila!  We have achieved farm.

Voila! We have achieved farm.

Thankfully, I remembered to drill an overflow hole about 4 inches up the side of the planters before going in for the night and having the mix and plants become completely waterlogged.

After all this effort, we’ll see if all the plants die.

1) Brainstorming and building stuff is fun. Thanks again to everyone who has participated so far.
2) The simplest design usually seems to be the best one.
3) In this case, the lightest design will probably be best too. These planters are almost as light as they come. The main weight is the compost.
4) We should do as much of everything on the ground as possible and haul it up rather than do it on roofs, like staining the wood and mixing the potting mix.

5) I think the big successes of this initial design were developing a way to avoid using too much plastic and keep the planter weight down, starting to conceiving a way to have rain fill the reservoir, and just generally beginning the design and building process.  There were some failures (er, not-yet-successes) too:

1) I couldn’t really figure out a way to suspend the mix high enough above the pond liner so it doesn’t weigh down and touch the bottom.

I tried and failed to come up with a way to clamp the fabric to the inside of the planter about 7″ down so we’d get a reliable 7″ hang, but I couldn’t do it. We’ll have to figure something out for mass applications of these things.

2) We’ll see how quickly root rot destroys the plants.  Actually, I didn’t completely plant one of the planters and still may have the opportunity to come up with an aeration solution.  This would give us beta version 1.1.  I plan to plant spinach and broccoli in the fall so the goal is to figure out a solution before that.
3) I didn’t work out all the costs yet but I kept the receipts. We can minimize costs by working with compost projects and growing from seed rather than buying and buying. Pond liner costs about 65 cents a square foot.
4) I didn’t weigh everything, but we should be able to do some tests and figure that out too. The finished planters are heavy.

Is there anything else? Not quite yet. The next step is to experiment with installing planters on boards that span rowhouse roofs, putting the weight on the party walls. There will be photos galore from that experiment, so stay tuned.


One response to “Building Prototype Planters

  1. Great site Jay! I loved Step 3 — Aloe should help 🙂

    – Emily

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s