Update December 2010: This effort is moving forward. See the product data sheet here: Grid Insight AMRUSB-1 Receiver for Itron ERT utility meters.
----
With all the hype about Google PowerMeter this past week, and with my recent discovery of Tendril's upcoming efforts, I think it's time to go public with my current project. In layman's terms, what I've done is simple: I'm reading my house's electric meter. Wirelessly. Every four seconds. And piping the data into my computer where I can do interesting analysis on it. It's basically the equivalent of Google PowerMeter, without Google and without a fancy "smart meter". It's a 100% DIY high-tech energy conservation solution. (And it was really fun to build.)
Until now, reading your electric meter took some bit of physical add-on apparatus, like the little gadget Black and Decker provides to time the revolution of the spinning horizontal disc inside your meter. Clever, but clumsy. A digital radio solution would be smaller, simpler, and could be used, in theory, from inside the house. But how would that work?
This idea is not new. I didn't invent it, nor did Google. In a previous post, I described ways in which at least two companies have solved the home power monitoring problem. But the new part, the part I've been working on, is going to be revolutionary. Why? Because until now the automated meter reading (AMR) technology embedded in many electric meters has been held as a trade secret by its manufacturer, Itron Inc. As a Washington State resident, I'm happy to see Itron, a local company, doing so well in their market. But as a consumer, I'm ready to see this technology open up a bit. After enjoying two decades of patent protection, its time for Itron to let the rest of us see under the covers of at least their older products. To be direct: the original patent on Itron's method of transmission and encoding, called ERT, has expired. According to my novice understanding of US patent law, that makes it fair game for the rest of us to explore and use.
We should exploit this technology because several studies (including one from the University of Oxford) have shown that direct consumption feedback (informing people about how much power they are using) can lead to simple changes in behavior that save power without reducing quility of life. Let's use some guesstimate numbers to see what the impact could be. If we assume that even 10% of households can be provided with cheap ERT-driven direct consumption feedback devices, resulting in 10% savings on 10 MWh annual consumption, that's a savings of 1 MWh times 1/10 of around 113 million US households. The result:
11,300,000,000,000 watt-hours of savings annually
(Someone more in the know on carbon equivalency can convert that to global warming figures for us.)
Millions of deployed Itron AMR-capable meters, and meters sold by partners (e.g. Landis+Gyr), use Itron's encoder-receiver-transmitter, or ERT, technology to transmit your electric consumption in 10 watt-hour increments every few seconds using a simple 900 MHz band digital radio signal. (The same band is used by baby monitors and older cordless phones.) Though the information is not really granular enough to support non-intrusive load monitoring (NILM) on small devices, the data provided is perfectly suitable for informing consumers about their energy usage, both in terms of immediate load and usage trends.
During January, in my dark and drafty basement using second-hand electronics gear (thanks to Paul in Lake Forest, CA for the awesome vintage Tek oscilloscope), I built a radio receiver and designed a digital decoder that listens to and decodes Itron ERT signals. After I receive the signals and convert them into digital data, I feed them into some software on my PC that updates an ongoing log of my usage. From there, I can easily massage the data into some simple graphs (see below). The visual are primitive right now, and I cut some corners in the elegance of the implemention. But it works. (Here are the live charts.)
As far as I can tell, this is the same concept that Google PowerMeter is based on, except people seem to have the idea that a new-fangled "smart meter" is required.
Maybe not. If you have an older meter with an embedded Itron ERT, chances are that your meter is chattering away every few seconds at about 915 MHz without anyone noticing. You could be using a receiver/decoder, like the one I've built, to get most of the benefit of the Google PowerMeter idea now without waiting for a fancy new meter.
I'm considering releasing under Creative Commons license my design and all the knowledge I've gained through my work building this prototype. It wasn't easy. But I'm torn on whether that is the right thing to do for the industry. Tendril, Inc. is preparing to market a product as part of their TREE system that does, as far as I can tell, more or less the same thing as what I have built.
I'm asking myself: is it right for me to potentially undermine Tendril's pricing power by releasing my engineering work for free? I mean, I admire the work that Tendril has done, but I don't want to see the public get snowballed into thinking that they need to buy some expensive piece of hardware in order to get value out of their existing electric meter. Moreover, I think other innovators need to take a close look at Itron's technology to see if there are ways this huge deployed base of semi-smart meters can be used to economically achieve "smart grid" goals in the short term with minimal additional expense and delay.
So, what do you think? Should I keep my work under wraps, or should I set it free? And if I do set it free, what terms should I put on its use?
This is very cool, and I love the site.
My thoughts are that this would be a great grass-roots effort to help get grid info into the larger web. Like we talked about before, at some point the ability to visually manage this data is the key - your power graph is priceless in many ways.
So, can you get this data into a format that Google's Power graphs can display? Can an open-source or subscription based portal be built that adds the visualization?
Since the device can be dumb, and widely built or distributed, then make that part as easy and cheap as possible. Add real value by then creating an aggregation portal for anyone that wants, or a gateway to someone else's (Google).
If you can put the graph in people's hands, or their iPhones (shocking that I would suggest this, I know) then they can make better power decisions.
Sasha
release it! we released ours on 1/5/2009 and dozens of companies and people are either using our designs or going to it seems...
http://www.ladyada.net/make/wattcher/index.html
http://wattcher.appspot.com/history?user=adawattz@gmail.com
http://wattcher.appspot.com/
http://twitter.com/tweetawatt
Fifty out of fifty exTendril employees agree... Let's see what you've got. Some pictures or gear a SourceForge project?
PS: You're not the only one...
http://davestech.blogspot.com/2008/02/itron-remote-read-electric-meter.html
Nice post.
How can I tell if my meter would work with your device? It seems we need some sort of database of meter features that can help home owners figure out what features their meter supports.