IoT community sprinkler system using Raspberry Pi | The MagPi issue 83
Saving water, several thousand lawns at a time: The MagPi magazine takes a look at the award-winning IoT sprinkler system of Coolest Projects USA participant Adarsh Ambati.
At any Coolest Projects event, you’re bound to see incredible things built by young makers. At Coolest Projects USA, we had the chance to talk to Adarsh Ambati about his community sprinkler and we were, frankly, amazed.
“The extreme, record-breaking drought in California inspired me to think of innovative ways to save water,” Adarsh tells us. “While going to school in the rain one day, I saw one of my neighbours with their sprinklers on, creating run-offs. Through research, I found that 25% of the water used in an average American household is wasted each day due to overwatering and inefficient watering methods. Thus, I developed a sprinkler system that is compliant with water regulations, to cost-effectively save water for entire neighbourhoods using a Raspberry Pi, moisture sensors, PyOWM (weather database), and by utilising free social media networks like Twitter.”
In California, it’s very hot year round, so if you want a lush, green lawn you need to keep the grass watered. The record-breaking drought Adarsh was referring to resulted in extreme limitations on how much you could water your grass. The problem is, unless you have a very expensive sprinkler system, it’s easy to water the grass when it doesn’t need to be.
“The goal of my project is to save water wasted during general-purpose landscape irrigation of an entire neighbourhood by building a moisture sensor-based smart sprinkler system that integrates real-time weather forecast data to provide only optimum levels of water required,” Adarsh explains. “It will also have Twitter capabilities that will be able to publish information about when and how long to turn on the sprinklers, through the social networks. The residents in the community will subscribe to this information by following an account on Twitter, and utilise it to prevent water wasted during general-purpose landscaping and stay compliant with water regulations imposed in each area.”
Using the Raspberry Pi, Adarsh was able to build a prototype for about $50 — a lot cheaper than smart sprinklers you can currently buy on the market.
“I piloted it with ten homes, so the cost per home is around $5,” he reveals. “But since it has the potential to serve an entire community, the cost per home can be a few cents. For example, there are about 37000 residents in Almaden Valley, San Jose (where I live). If there is an average of two to four residents per home, there should be 9250 to 18500 homes. If I strategically place ten such prototypes, the cost per house would be five cents or less.”
Adarsh continues, “Based on two months of data, 83% of the water used for outdoor landscape watering can be saved. The average household in northern California uses 100 gallons of water for outdoor landscaping on a daily basis. The ten homes in my pilot had the potential to save roughly 50000 gallons over a two-month period, or 2500 gallons per month per home. At $0.007 per gallon, the savings equate to $209 per year, per home. For Almaden Valley alone, we have the potential to save around $2m to $4m per year!”
The results from Adarsh’s test were presented to the San Jose City Council, and they were so impressed they’re now considering putting similar systems in their public grass areas. Oh, and he also won the Hardware project category at Coolest Projects USA.
The MagPi magazine #83
This article is from today’s brand-new issue of The MagPi, the official Raspberry Pi magazine. Buy it from all good newsagents, subscribe to pay less per issue and support our work, or download the free PDF to give it a try first.
This is what I love about the Raspberry Pi, the way it gives creators the power to innovate. Such a versatile and easy to use SBC. Using it as a desktop is almost a waste.
It looks like the link to “download the free PDF to give it a try first.” is broken. …/issues/83/ is …/issues/issue-83/ these days.
Thanks for the excellent publications.
This worked for me: