New BDes(Hons) Interaction Design

Counting the Cost


A couple of weeks ago I came across a bunch of blogs written by IBM folk at Hursley Park, in which they described how they’d been playing with the CurrentCost energy monitor so they could get the data out and send it to twitter feeds, produce comparative graphs of consumption (like those above), and other stuff.

I’ve been interested in this idea ever since I saw Wattson by DIY Kyoto being demonstrated at DesignersBlock a couple of years ago… the CurrentCost is pretty fugly compared to wattson, but then again it’s far cheaper. I just bought one for £28 from eco gadget shop.


The system comprises a clamp that you fit in your meter cupboard, which then measures the amount of electricity you’re using and transmits it via radio link to a small visual display that you can use to keep an eye on things.


There are other devices that do similar things, but the CurrentCost has a serial port through which it outputs its readings every 6 seconds or so, which means you can use the data for whatever you like. It’s even in XML.

The only downside is the connection — an undocumented RJ45 jack. There will eventually be a data cable available, but it looks like only IBM people have this at the moment. I spent some time today trying to fathom out the connection but the pinout doesn’t appear to match any of the conventions for RS232 over RJ45. I chopped up an ethernet cable since they use RJ45 connectors and I have loads.

Anyway after some experimentation and thanks to Roo Reynolds I now have the unit connected to my Mac. It turns out that the data cable has a converter chip in it, and that the signals are 3.3V. I already have a cable that does the same (made by FTDI up the road in Glasgow, and which I recently bought from Adafruit in NYC — oops), so I used that to make the connection and all was well.


My unit sends its data at a particularly underwhelming 2400 baud, but since the packet only arrives every 6 seconds or so it’s fast enough. (Just!)

The pic’s a bit fuzzy, so here’s the order of wires from the ethernet cable, Left to Right:

green/white (pin 7, data in to CurrentCost)
blue (pin 4, ground)
brown (pin 8, data out from CurrentCost)

I’m looking forward to connecting this up to an arduino. I have a couple of different ethernet shields to try it with: the adafruit one with an Xport that I’ve had for ages, plus the new supercheap one from nuelectronics (£12.99! they also sell freeduinos for the same price).

Both of these have code available to let them talk to pachube via EEML.

Right, easily enough gobbledygook for one posting.

Excellent article, thanks!

I have a Current Cost meter and the associated serial wire, but since my macbook doesn’t have a serial port I’ve been attempting to use my Arduino to read the serial data and send it over USB to the arduino serial monitor. I’ve not been having much luck.

Do you have any notes on using Currentcost with SoftwareSerial library on the arduino?

I now have one of these units, and the cable too. Is there any information on the software people have used to grab the XML file and do anything with it?

For the record, Alexis Birkhill has documented connecting and coding the Arduino (and I see Martin already found this the same day he rolled his own).

For further info I would keep an eye on the new CurrentCost wiki even newer wiki hosted on wetpaint.

And another small update: Dale Lane has written some windows software for logging and viewing the data.

Also Nick O’Leary and Roo Reynolds did a presentation on CurrentCost hacking at Opentech 2008.


My data lead arrived today from Current Cost but I have found out I have an old 2400 unit. When I spoke to CC they told me they were selling “Display” only units on eBay. These do not have any transmitters or clamps.

I have bought one straight away for £15 with £2.99 P&P.

[…] Power On September 6, 2008 On 1st September 2008 I moved into a new flat.  As I’d already been playing around with the Current Cost energy monitor, I’ve decided to use this new blog to document what I’m doing with it in the new place. […]

I’m down under here in 230-240V 50Hz NZ & have just been testing a CC for accuracy againts more professional meters! Yikes -on even simple resistive loads (such as a 100W lamp & 2kW kettle), it was miles out,giving instantaneous power readings ~30% too low. Surely a calibration routine/tweak exists? I’ve also found the clamp position & orientation when straddling the phase mad a significant reading difference.

My unit has a USB B type connector instead of the RJ45. I found that the data is transmitted from pin1 (red wire, usually 5v) and the pin 4 black for ground.
Has anyone managed to access the data that is stored on the ubit?

Regarding accuracy, I just read this on the Ebay description of a CC sensor ‘jaw’, being sold there by current cost:

“Best results will be obtained if the sensor is perpendicular to the cable and (if practical) with the cable nearest the hinge end.”

I am a Mac user, but have been testing the CurrentCost monitor with RJ45-USB link to a Toshiba Windows XP Netbook using the open source Dale Lane GUI which I can recommend. The current version (0.9.15) includes a number of helpful graphical analyses and the facility to export data to Excel spreadsheets. An earlier version (0.9.11) has an interesting 3 phase real-time display.
It is important to note that it is essential to use the C2 (433MHz enabled) version of the CC monitor – earlier versions will not work.

From our tests CurrentCost is aware of inaccuracy if the sensing claw is not attached correctly. This needs to be mounted so the power cable is as close as possible to the claw’s hinge. CurrentCost intends to issue a sponge collar to assist with this.

The task now is make this all work on a Mac with OS X.


People using the Current Cost and similar monitor types (Electrisave/Owl, Efergy, Wattson, etc.) – which pick up their signals using current transformer clamps around the live mains feed – should be aware that these have a fundamental inaccuracy if the electrical load is not simply resistive. This is because they work on current, not power, and over-read if the voltage and current are out of phase. This is the case with some motors (e.g. fridges and freezers), and with low-energy and fluorescent lights. For example, some of my lighting shows up on the Current Cost at roughly twice the correct wattage! There are also problems if you have solar photovoltaic panels connected to the grid, since these monitors can’t tell the difference between outgoing and incoming power. These monitors are useful, but this limitation should always be remembered.

[…] the other blog I had a comment recently from Eric Eisenhandler, a physicist who’s involved with the Blewbury Energy […]

Just hacked mine, opened up the unit to and had a mosey about in it.

Its pic based and all very basic stuff. re-writing firmware for it should be pretty easy if you wanted to do that.

Among other trivial hacks are the serial port itself, running mine on a home brew FTD232BM board is use for PIC development. Oh and this one runs at 9600 not 2400. There looks to be support for squirting other things into that port too.

Wiring an X-port in in place of that RJ45 should be pretty easy too. I have one kicking about someplace so thats on my to-do list. Just gonna write some software now :)

[…] some more detailed uses check out: and also for some tips on […]

Have just got my CurrentCost cc128. Very excited to get the data off and start playing.

[…] either a home built serial cable ( or a cable from Current Cost themselves you simply connect it to the PC and open the serial port. […]