ITPWC is a phytoremediation device based on the Eco-Machine systems developed by Dr. John Todd* and research done by the ITPWC team on aquaponics systems. Built in the men’s bathroom on the ITP floor, our inspiration for this project was to create a system that made a positive impact on air quality while also improving the overall ambiance of the space.

Aquaponics is the combination of aquaculture and hydroponics. Aquaponic systems are designed to grow plants and fish symbiotically in one integrated system. The fish waste provides a food source for the growing plants and the plants provide a natural filter for the water that the fish live in.

A large scale aquaponic system build on the Eco-Machine paradigm has been designed to filter human-created wastes from water, so that water can be recycled for human usage. This system lives at the Omega Center in upstate New York. Two members of ITPWC team had an opportunity to visit Omega, and get a tour of this new inspirational creation.

To track the impact that this system had on the air quality in the men’s room, we installed sensors that measure changes in the level of methane and solvent vapors. These sensors were attached to an Arduino that uploaded average readings every five minutes to data feed onPachube.

Here is an overview of how the different components in this system work together to create a symbiotic ecosystem and to remediate the water:

1. Aquaculture: The fish living in this component excrete waste and ammonia into the water, which will act as fertilizer for the plants. Water is pumped up to the grow bed and then gravity does the rest to keep things moving.
2. Phytoremediation: These plants remove chemical vapors from the air, as well as help rid the water of pollutants. They absorb nutrients from the nitrate-rich water and produce oxygen at a high rate. Plants used include dresden, boston fern, golden pathos, english ivy, peace lily.
3. Aquaticremediation: Aquatic plant roots provide habitat for waste eating microbes. Filter feeders, like the clams in this system, help remove algae and other toxic particulates from the water. Plants used include penny worts, banana plant, freshwater plant, freshwater flat clams, and duckweed.
4. Bio-Filter: The bio-filter is an anaerobic environment (no oxygen) where micro-organisms and bacteria strip any excess ammonia, nitrates, nitrites and phosphorus that remain in the water. The mud used to inoculate the system comes from Prospect Park lake in Brooklyn.

I want to thank the ITP WC team for all the hard that they devoted in building this system. We all wore many hats on this project but I want to give credit where it is due. Macaulay took the lead on the physical design and production of this living system, while Adib became our resident aquaponic/eco-machine expert, and Marko led the way in creating the communications for to system.

Here are some pictures from the installation itself:

STEP 2: Building the Installation and Fixing the Code

During this past week we continued to make good progress on the ITP WC project. Our main objectives were to continue working on building the physical structure for the aquaponic installation and to fix issues with the sensors to ensure that we are able to get valid readings from the bathroom.

The Physical Build
After working on developing detailed plans for the bathroom installation, Adib, Marko and Macaulay set out to construct the structure for the aquaponic system. The final design leverages the large Poland Springs bottles that are common to the bathroom as the containers for the living system. To create an appropriate structure, Adib designed a set of clear plexi stands that will serve as displays for the repurposed bottles. Here are a few design sketches:

The Bathroom Layout – Overhead View

The Aquaponic System – Close-Up View

Now here are a few pictures of Adib, Marko and Macaulay working on building the installation earlier today.

Air Quality Sensors
On the air quality sensing front we continued to encounter some issues. For our Carbon Monoxide and Methane sensors to work properly we had to re-write the Arduino code. While for our Pachube feed to work continuously we had to upgrade our account to Pro level (luckily, for students the pro membership level is free).

Here is a brief overview of the changes that we incorporated into the code for the Methane and Carbon Monoxide sensors. Working with gas sensors is usually tough. One of the main reasons is that gas sensors need to run for several days before they are able to provide accurate readings (usually they need to run for 4 to 5 days). Since we set-up our sensors we had been encountering problems with the readings readings and the feed.

Based on information that I found out from conversations with Melissa Clarke last week I became aware that we needed to set-up our sensors differently. Carbon Monoxide and Methane sensors require that their heating element receive alternating amounts of voltage every 20 seconds. Furthermore, in order to get accurate readings it is crucial that the microprocessor capture data during a very small timeframe that lasts approximately 5 milliseconds.

This data was not clearly spelled out in the short datasheet that I received along with the sensors. It was only after finding the the detailed documentation online that I was able to accurately understand how to write the code. Here is a link to the code I wrote, available on github.

STEP 3: Local Sensing, Worldwide Sharing

Over the past week we have been making a lot of progress on our bathroom intervention project, ITP WC. Macaulay and Adib have been taking the lead on the physical design and planning for our installation, including the aquaponic system. They have completed some initial prototyping work and have begun to procure the materials for the final build. More updates to come on that front from the two of them.

Marko and I have been focusing our efforts on creating the circuits that feature the gas sensors, and writing the code to share the data using Pachube. I’m happy to report that we have come a long way. We have installed our two first sensors in the ITP men’s room and we have started to feed data to Pachube (feed 11193).

Our first step was to set up the sensors on a breadboard and hook them up to an Arduino with an Ethernet shield. We started with a sensor the captures Methane and Carbon Monoxide (TGS 3870), and a second one that captures solvent vapors (TGS 2600). For building the simple circuits we got some help from an instructable article, since the datasheets from Figaro left much to be desired. We also looked at some tutorials on the arduino site to get up and running with the ethernet shield.

Getting started on Pachube required a good bit of learning. We had to get up to speed onPachube’s API, which is not very complex but does require some time and effort. We used one of their tutorials as the basis for our code.

The last step was getting the sensors installed in the bathroom and tweaking the code to make sure we were getting reliable readings. The main change we incorporated into the code was to add functionality that takes several readings from the bathroom during a one minute period and averages those readings together uploading data to Pachube (which is done once a minute).