Being at home over Christmas, followed by a week of cold weather has meant a rather poor start to my aim to achieve 10:10.
Tesco has for some time allowed people to convert Clubcard points into Airmiles, so on one hand, this story seems a bit of a non-event… Except for the fact Tesco are actively suggesting people convert the bonus miles earned when buying CFL bulbs into Airmiles.
That’s right – suggesting you get in a jet and burn a load of fuel as a reward for ‘greening’ your domestic lighting.
I’m all for encouraging people to reduce their consumption, but this needs to be by educating people on the reasons why, not by offering discounts on massive energy consumption as a reward for minor reductions.
After a few weeks away, I’ve had another fiddle with the Python scripts I run to graph data from my Current Cost meter. You can see the results here.
Of note, I’ve switched from using the Google Graphs API to RRDTool. This is for two reasons – the Google graphs API just seemed a bit fiddly, even using PyGoogleChart, and rrdtool graphs just seem pretty easy to output. In addition, it means I get averaging for free and don’t have to manage the database.
I’m still using the Python I tweaked from Dale Lane, but I created an RRD based on Paul Mutton’s guidance and update it using a Python module for RRD from Corey Goldberg. Feel free to ask for a copy of the scripts.
For some reason my sqlite db update script liked to bomb out after a couple of hours, but this seems to be working fine and has been gathering data reliably for the last few days.
Of course the whole point of this isn’t just to make pretty graphs, but to try and do something about our consumption. I’m already obsessing about the graphs produced, and they’re making it easy to see when something’s left on – our ‘base load’ seems to be about 200W, so anything more than this means something’s on somewhere.
Our most obvious heavy power usage is the oven, at about 3kW, but there’s not much I can do about that other than cook more with gas – or eat more salad! It’s generally not on for long, so not the end of the world. Interestingly, the oven seems to heat up in under 10 minutes (solid power use, followed by intermittent bursts to keep it hot). I’d always assumed this to be longer, so that’ll save some time and money!
I’m also quite conscious of the fact we sit in the living room with a few 60W lights on in the evening for quite some time. They’re on dimmer switches so I’m now hunting for dimmable CFLs. The Megaman bulbs seem to come quite highly rated, but I’m not really sure how good they are. If anyone’s readig this and has some insight, feel free to leave a comment!
My next plan with the graphs is to work on aggregating daily/weekly data to show trending totals and ignore the short-term fluctuations, so that I can see how things are improving. When I get some time at home to myself I’m going to try and work out exactly where the 200W base load is going (NAS, router and Sky are the key candidates) and see if I can get it lower.
Since installing my Current Cost meter, I managed to get Dale‘s graphing app up and running fairly quickly. Since then, I’ve progressed a bit with my own Python.
The scripts are fairly thrown together as I’ve waded my way through various Python documentation, and the sqlite and Google Chart API modules, but I’ve managed to get some initial graphing up and running, which you can find on the Current Cost tab (or click here).
The graphs should update every couple of minutes, and so far show power usage & temperature readings for the last 15 minutes, and last hour. To get these working, I have a script which listens continually for the output from the CC128 and stores it into a sqlite database. There’s a second script which reads this database and makes calls to the Google charting API, generates PNG files and then pushes those up to my blog host.
I’d like to get 6-hourly, daily and weekly graphs, but these will require a bit more fiddling as the data needs aggregating before passing to the chart API. I suspect it’ll be a few weeks before I have the time to get that working… Likewise the graphs need a bit of tidying up (better axes, labels, etc).
In the unlikely event that anyone’s interested in the Python scripts to do this, either leave a comment in this post or drop me a mail.
…almost. I’ve mentioned before how I’ve been reading my electricity and gas meters every four weeks or so, which gives a rough approximation of how much we’re consuming in our house.
I’ve got my Current Cost meter up and running with Dale Lane‘s Python app. This was ever-so-slightly fiddly, so I thought I’d add my voice to those already out there with some step-by-step instructions for OS X users.
- Get a Current Cost meter
- Order the USB serial lead
- Install the drivers for said lead – this creates a new device
- Install the Apple Developer Tools
- Download the Python source for Dale’s app
- Follow the prerequisite list
- Install MacPorts
- Rebuild SQLite (run sudo port install sqlite3 +loadable_extensions from an XTerm)
- Export the DYLD_LIBRARY_PATH (for me this was export DYLD_LIBRARY_PATH=/opt/local/lib)
- Run the app – python currentcost.py
Not too bad, and you should get a nice pretty graph, like so:
Next up, I’m going to be taking the real-time data into a database and trying my own bit of graphing.
My Current Cost CC128 arrived yesterday, an eagerly awaited treat after my day at Home Camp ’08.
It’s already producing surprising data – when I got up this morning, the house seemed to have an ‘ambient’ power consumption of 278 watts, which seems pretty high, but I think the boiler was whirring away at the time.
As I sit here, with some lights on and various other bits and pieces, it’s reporting 674W, or £2.26/day at this rate of consumption.
The real value of the Current Cost though is the cable for attaching to a PC, so you can capture and report on the data. Not being too much of a coder, it’s going to be a bit of an adventure – the CC just outputs chunks of XML every 6 seconds, but handily there’s tons of good work out there to ‘borrow’.
I came across a great post at ecogeek.org (hmm, that makes me sound even more nerdy than I am) today on a topic we discussed at Home Camp. It talks of encouraging people to cut home energy consumption by turning it into a game, referencing the Behaviour, Energy & Climate Change Conference.
It seems that the idea’s gaining some traction – I see more and more references to it. I really like it as a way to get consumers as a whole to talk about their energy usage, something which most people see as inherently fairly dull.
Whilst products like the Current Cost or Onzo are great for getting individual households to reduce consumption by comparing against their own historic use – or even more simply watching how much money it costs to boil a kettle – encouraging people to be competitive seems a great way forward.
As well as the encouraging news that British users will all have a smart meter by 2020, we need to make sure the data provided by these meters is standardised and easily ‘mashable’ to allow games like this to evolve on their own. I can imagine energy providers or other organisations offering ‘free stuff’ or discounts to the winners of ‘biggest loser’ competitions if all of us can share our energy data safely and securely – though of course there’s some irony in giving out products to people who are best at cutting their energy consumption. It’d be great to see, for example, the government offering discounts on your council tax bill as a reward for consistently reducing your consumption. I suspect, though, that would be a step too far given how most councils still live in the dark ages.
So far, a very basic form of this exists – my Carbon Account posts my carbon footprint to Facebook each time I add a reading, and I can compare with friends, but this is an entirely manual process reliant on my submission of meter readings once a month. I’m sure wider availability of Wattsons, Onzos and Current Costs will get this moving – I can see me handing them out to my family and friends as Christmas presents this year.
Meanwhile, take a look at this video showing a great example of how it might all come together.
For some time I’ve been taking monthly meter readings of my gas and electricity bill – I started a little over a year ago when we moved into our new house. The new house is quite a lot bigger than our old flat, so I was concerned that the bills would be a lot more… Also, I’m keen on doing my bit to become more efficient.
This graph shows 4-weekly gas & electricity readings, divided by the number of days between readings, to give an average daily reading. The most striking is our gas usage. Obviously usage declined as the weather got better, but we had a condensing combination boiler installed to replace a 20 year old conventional one at the end of May.
Following the boiler install, our gas usage dropped to basically zero – helped in part by three weeks in the summer when we had our new kitchen installed, so the only use was hot water. Since the weather’s got colder again and we’ve had the heating on, you can still see a significant difference to our gas usage.
For electricity, we’re obviously using less than we were last year. We’ve been consciously trying to be more efficient – I’ve been gradually replacing light bulbs with CFLs, and trying to turn things off when not in use. I’m hoping that getting a Current Cost meter (or hopefully an Onzo) will trigger some further change in behaviour.
I also sold my car in the summer, and instead rely on Streetcar. This is saving me a lot of money, but also is an environmental benefit to a degree. There’s the debatable benefit of reselling my car theoretically avoiding someone else buying a new one, but also the fact that it’s encouraged me to think twice before driving.
What’s next? I’ve noticed our loft insulation isn’t up to much, and there’s a big hole where our cold water tank used to be, so I’m thinking of installing some Thermafleece.
I’d really like to double glaze all our windows, but large double-glazed sash windows don’t come cheap, so I think the best I’ll manage is to install draught strips around the doors and windows.
Any additional suggestions welcome!
I started Saturday morning huddled with an old friend and one of his colleagues in an Imperial College corridor, wondering what I’d let myself in for. I usually spend most of my time in meetings debating the minutiae of corporate security policy, and relatively little doing ‘academic’ thinking, so coming to an event designed to encourage open debate on emerging technologies with a bunch of self-professed geeks was an interesting departure.
I’d come to HomeCamp, an unconference on home automation and energy efficiency. Despite being initially sceptical, I’ve spent much of the weekend mulling over the day, which consisted of a number of sessions that behaved more like one big group discussion.
There were a number of topics that stood out for me.
Data Gathering and Visualisation
The basic assumption of much of the discussion was that people need to understand their consumption, and this is achieved through monitoring devices. Ideally, you’d monitor the usage of every device consuming power in your home (of course minimising the energy used to do so!). Andy Stanford-Clark gave a great talk on his automated, twittering home – with lots about his Current Cost meter, and impressive use of the IBM message brokers that he’s developed, to gather data from everywhere… Even down to his mouse traps!
In reality, I believe the average (non-geek) user needs an easy way to capture this data and see it represented. I had a great chat with Ben from Onzo, who look to be producing some nicely designed hardware and have already done a deal with a major electricity supplier to distribute these to customers. I’m looking forward to finding out more about their products.
I’d hoped that someone from AlertMe might be there too – I’ve followed their security products with interest, and was recently excited to see they’ll be expanding into energy products. A logical step given their network of Zigbee sensors, and low-powered, always-on base station.
Behaviour Change – Demand Shifting, Demand Reduction
Having gathered the data, you need to do things with it. Firstly, smart meters (or better, smart appliances) will allow your utility company to communicate with your devices in order to smooth the demand curve, and operate at cheaper times. This is important given the inefficiencies that result from our varied demand, which results in power stations that have to start up at short notice, or run at less than 50% capacity. Tom Taylor wrote a good overview post on dynamic demand, based on what he learned yesterday.
I’d never really thought too much about “Economy 7″ and how it came about, beyond my grandad running his dishwasher on a timeswitch so it started at 2am to save him money. If this concept were extended into much more advanced variable pricing to encourage us to smooth the power demand curve, electricity could be generated much more efficiently, reducing costs and pollution – but also allowing use of more ‘variable’ renewables, like wind.
In addition to smoothing the curve, there’s clearly an incentive for us to all use less power, full stop. Dan Hill eloquently describes a concept for visualising this sort of data, and this sort of concept was discussed at length at Home Camp.
Of note, there was a debate on how to encourage people to act – between the simple financial savings and leveraging peer pressure to encourage reductions in consumption. There were some great suggestions around something like ‘Xbox Achievement points’ for challenges, or using Facebook, etc., to give context to your consumption. What’s the best way for me to show my friends I pay £50 for electricity a month, and my fridge represents 6% of that, for example? Could we reward people for being the ‘biggest loser’ somehow, for making tangible changes to consumption rather than just advertising cheap CFL lightbulbs and hoping for the best?
I’d love my electricity bill to give me that context. Don’t just tell me I need to pay £50 a month, tell me that other similar-sized houses with a working couple pay £75 a month, and that I could reduce my bill to £40 by getting a smarter fridge that switched itself off whilst the rest of the country made tea after Eastenders.
For businesses, we got to hear about Pachube (pronounced patch-oo-bay, I think), a startup service enabling people to capture real-time data and aggregate/share it. It’s designed for any sort of data from anywhere, but we talked a lot about companies (or even cities) submitting building data (lighting, heating) which can be reused. A fascinating example showed Pachube data being fed into SketchUp for visualisation – not far from Dan Hill’s idea of hovering sparklines of consumption data above buildings.
In summary, then, it was a great day. I learned about a lot of new technologies, and new ways to apply existing ones to help us reduce energy consumption. It was a good start to what will no doubt be a successful series of events in the future. I’ve have had Home Camp thoughts running around my head all weekend – to the point that I think it’s reaffirmed my belief that I’d like to take a career in this field when a suitable opportunity arises, nudged even further by a recent post by a friend. Thanks to Chris Dalby for organising, and I look forward to the next Home Camp in March. Until then, I’m hoping to spend more time thinking about these topics, and ideally writing about them here.