Liquid System Update
Last month we briefly introduced the idea of the “user-day equivalent”. You might be wondering what that means, exactly?
As we develop, deploy, and test our systems, we need a way to compare data we collect in different phases of testing and at different sites in a way that makes sense. In particular, we need a way to compare data that we collect in the laboratory — where we control how much urine and feces are flushed through our system, and measure it all very precisely – and data that we collect from a field site, where we simply log the number of times that users visit the toilet. We have to constantly evaluate (and re-evaluate!) whether the assumptions we make in the laboratory bear out in the real world.
When we look at our field site at CEPT University, the math is much easier: we estimate that a typical user visits the toilet 3 times a day. So we just divide the total number of uses logged by 3.
Here’s the fun part: once we do this we can look at data from the lab in North Carolina and the field site and India, and see how they compare. Here are a couple of examples:
Starting on the left, conductivity (you may remember from last time) is a measure of the salts in our process liquid. Most of the salts in our process liquid come from urine. As you can see, the early data points are in really good agreement between the sites, but over time, we see much less conductivity in the liquid from the field site than we do in the lab. The same trend emerges with COD (chemical oxygen demand, right graph). COD is a measure of all the chemical species that consume oxidants, which here are generally nitrogen compounds that again, primarily come from urine.
So what’s going on, here?
Well, the honest answer is that we don’t know – yet. But we do have some ideas. One possibility is that we’ve underestimated the amount of hand wash water that is going into the system at CEPT, which would dilute everything. Another possibility is that our usage assumptions aren’t panning out at this test site: specifically, that users are defecating in the toilet much more often than they are urinating in it. After all, we don’t follow users into the toilet—even we have limits about what we’re willing to do for science! It’s also possible that both of these things are true.
One of the challenges for us in the laboratory going forward is to re-visit our usage assumptions and see if we can reproduce what we see in the field in the lab. Once we are able to do that, we’ll have a better idea about how to improve our system for use in the “real world”.