This is my first week at Nourish the Planet. I’ve been focusing all of my attention to creating an automated water testing robot. I’ve calculated that at Nourish the Planet about 14-18 man-hours of time are used to test and record water quality each week. If we assume that those hours are approximately worth $10 per hour, that’s $140-$180 each week. If there was some way to automate this process NTP would save $7280 – $9360 per year, at least in terms of the value of the labor.
I’ve been researching and coming up with a lot of different ways to have a robot automatically test the water. So far I’ve needed to learn about the different types of analytical chemistry and how to best approach the problem. I’ve also learned of the current methods of testing water. The closest I’ve seen to what I want is the Seneye sensor. The only problem with the Seneye is it only measures ammonia and pH. I want to make something that does all the testing a human does, including nitrates, nitrites, etc. What I essentially want to do is make a robot that will do exactly what a person does with the API test kit. I’ve looked into a couple of different ways of doing this.
There’s this thing called the Hanna checker which uses a colorimeter to measure the change in color of a water sample after reagent is added. This could work as basically the “seeing” part of the robot that would make a judgment of water quality based on color. The other part of the robot is making something that will combine the test reagent and the sample water in the right ratio, then dispose of it and rinse it. I think this will probably be the hard part, although from what I’ve found there’s a couple of different ways to solve this problem already. I’ll have to do more research to find out what’s the best method. For now, this is my initial design: