Please create an account to participate in the Slashdot moderation system


Forgot your password?

Real-Time Power Monitoring Options? 172

tedpearson writes "I've wanted for quite a while to be able to look at my electricity usage in graphed form, both real-time and historical data. There seem to be a number of options for power monitoring in existence: some that hook into Google PowerMeter, others to Microsoft Hohm, and some that are standalone units. I've also seen DIY projects using Arduinos for reading the data and sending it to a computer. But I haven't found anything that is quite what I'm looking for, and I am hoping the Slashdot community can give me some advice. What I'm looking for currently: Some sort of device(s) that a) accurately measures power usage, b) allows me to access the data for storage in a database for my own graphing/analysis purposes, c) will work with MacOS (doesn't require Windows), and d) doesn't cost more than $150 or so. DIY is fine, though I don't understand circuit design, which is keeping me from designing something myself."
This discussion has been archived. No new comments can be posted.

Real-Time Power Monitoring Options?

Comments Filter:
  • Watt's Up Pro (Score:3, Insightful)

    by Robbat2 ( 148889 ) <> on Friday September 24, 2010 @04:57PM (#33692014) Homepage Journal

    I use the Watt's Up Pro, but it's for monitoring a single outlet.

    Do you intend to monitor your entire house, or just some devices?

    • And check how fast the dial in the electric meter is spinning.
      • +1... if you're too lazy to walk a few feet outside and read your meter, then you're definitely too dependent on electricity.

        • by lazlo ( 15906 ) on Friday September 24, 2010 @06:46PM (#33693198) Homepage

          Are you saying you're un-lazy enough to walk a few feet outside and read your meter? And write down the reading? Every five minutes? For a month?

          Yes, it's very easy to track your average monthly power usage, it's right there on your bill. It's also easy to check your instantaneous usage by looking at the meter. What the OP wanted to know wasn't just a point measurement, but a running graph to see how it varied from hour to hour throughout the day.

          • You could aim a webcam at it and have it snap a picture every 5 minutes.

            I actually suggested a friend of mine use that technique for wanting to be able to remotely check the reading from a weather station on his boat while he was at home.

            I'm not really joking. If you're good with software, but not hardware, this might be the way to go. Then have some image processing software read the data you want from a series of images, or in realtime, and pop it into a spreadsheet. You could even do with no internet con

        • That's fine if you want to check your power usage at whatever interval you go out there. What if you want to measure usage every 2 minutes 24/7 including while you're away at work?
          It's not laziness at all.

          However, my advice to the writer would be to reevaluate your purpose here. Saying "a) accurately measures power usage" is more than a little vague. I assume that you're vague because you don't know what you want to do. No offense intended, but you need to define the problem before you start generating
        • by smithmc ( 451373 ) *

          Um... this is Slashdot. Some people around here do geeky things for fun. Of course, as you've just demonstrated, other people around here post snarky adolescent comments that add nothing to the conversation. Takes all kinds, I suppose.

      • Reading the meter (Score:4, Informative)

        by SIGBUS ( 8236 ) on Friday September 24, 2010 @05:09PM (#33692194) Homepage

        At the meter, you can calculate the power draw. Look for the Kh value on the meter, and count the number of seconds it takes for the disc to make one full rotation. Then, use this formula:

        W = Kh / (Seconds / 3600)

        to get the power draw in watts.

        Of course, this assumes you're still using an old-school spinning-disc meter.

        • Re: (Score:3, Interesting)

          by markdavis ( 642305 )

          Why are you all assuming there is a turning disc in the meter? I haven't had one in mine for many years. There is only a digital readout with FULL KW/h and no current load reading at all. It is completely useless for real-time monitoring. Thanks for the "improved" meter, Virginia Dominion Power!

          • Re: (Score:1, Redundant)

            by blair1q ( 305137 )

            Here, you dropped this:

            you insensitive clods!

          • I do have a spinning disc. ;)
          • Re:Reading the meter (Score:4, Interesting)

            by jjhall ( 555562 ) <slashdot AT mail4geeks DOT com> on Friday September 24, 2010 @05:31PM (#33692478) Homepage

            The digital meters used in the Idaho Power area anyway has a scrolling line on the bottom of the digital display. This represents the old turning wheel and uses in fact the same calculations.


            Our power meters use a slightly different digital method, it has a bar that "fills up" at the bottom, and it is measured from the moment it resets to the next reset as the equivalent to one wheel revolution.

            Your power meter should have a way to see current usage, give your utility a call if you can't figure it out, and if there is in fact no way to read it, I'd get in touch with the public utilities commission and see if it is a requirement.

            • by jpapon ( 1877296 )
              You'd think that with all this digitizing of the meters, they would put a simple serial port (or even just some pins) on it so you could monitor it remotely.
              • they don't want a port that makes free power easy. No they want to brake the seal to get at that.

                • It would be pretty hard to hack the meter if the pins just output a blip (or change polarity, or any other binary indicator) for every X units of energy are consumed, which is all the information the meter really provides in the first place. That sort of signal could be read directly by a serial port if you really didn't want to build an interface, or converted to a real digital signal with a pretty trivial circuit.

                  That being said, it's not that difficult to build a circuit that optically reads the LCD "whe

              • I got radio controlled meter readers on my gas and water counters a couple of years ago, electricity still lags behind.

            • Our meters have a single set of digits showing whole number of kwatt hours. There is no bar chart, no line, nothing else except a tiny blinking triangle (which when I looked at it now is blinking about once per second). No indication as to what the triangle means. It is pretty crappy.
        • Before I tackled a similar project, I calculated how much this "real time monitor" would cost, figured out how much I'm likely to save (a few pennies per day), and then decided to buy a $20 gadget on amazon called "Kill A Watt"

          It tracks how many KWh a single device (like a TV or freezer) is using. Once a month I look at it, multiply by 9.5 cents, and figure out how much energy it used (typically 1-2 dollars - trivial).

          I discovered the appliances use very little energy. Most of the $300 bill is caused by t

        • by 2.7182 ( 819680 )
          And if you want to automate it, but a little wireless camera on it, stream it into matlab, track the image of the spinning disk and count its rotation rate.
        • PG&E was very close to rolling out the capability for nearly everyone in California to look at their power usage in near-real time online, but public hysteria over the "RF" generated by smart meters has halted rollout in many locations.

          The recent gas line explosion really has called out the pitchfork and torches types, so I except most other locations in the US to have this a lot sooner than California.

          Of course PG&E really wants the capablity to charge us double on the hottest days, but they would

      • Really? What if one wants to know that they are being billed correctly? BGE is big on overbilling. Heck, we went off grid, and got $2900 in bills for usage we could not have made (meter disconnected - by THEM no less). It took NINE months for them to admit it was some sort of "error" - a fucking "error" that they were claiming the smart meter, disconnected from everything, was reporting such power usage? My ass. Baltimore City was charged a quarter of a million in usage charges for a building that had been

        • Our current LIPA bill here in NY has DOUBLED for some reason, with no new appliances, and no new usage patterns.

          Is your neighbor running a grow house? Maybe you should look for a vampire feed ...

          • LoL, no such thing. We're still trying to figure this out and even had an electrician come in and meter peak usage. No changes since the last time (when the bill was less). There was a recent rate increase (but not double or even close), so next we are trying to get an idea of whether there's some new idiotic method (of billing) they are doing things that may be causing this. Our power usageis generally consistent every hour of the day (we run a stack of always on servers - same stack for years, same server

      • Re: (Score:3, Insightful)

        by Idarubicin ( 579475 )

        Go round the side of your house...And check how fast the dial in the electric meter is spinning.

        Which works brilliantly, as long as he's only interested in knowing his electrical consumption while he's outside the house -- and not when he's busy cooking, watching television, ironing, using power tools, or having a hot electrically-heated shower. And as long as he doesn't care about electrical usage while he's asleep at night or when he's at work during the day. And as long as he doesn't mind getting funny numbers because the baseboard heaters in the front hall cycle on for a few minutes every time

    • Re: (Score:3, Informative)

      by onebadmutha ( 785592 )
      Stick a webcam on it, do ocr to text on the numbers. Sheesh!
      • Re: (Score:3, Insightful)

        by jpapon ( 1877296 )
        You wouldn't even need OCR. Just watch the lowest digit, and count every time it moves!
      • Re: (Score:3, Funny)

        You joke, but I've done that: put a handheld digital multimeter face-down on a flatbed scanner, and used gocr to get data. It was *very* awkward, and once I got gpib running, I've never looked back at that without shivering.
      • Better yet, something like this [].
    • I'd like to monitor the entire house. I've thought about the whole shebang... monitoring every circuit - but that would be more complex and expensive.
  • IObridge (Score:5, Informative)

    by rodrigo1979 ( 255519 ) on Friday September 24, 2010 @05:02PM (#33692090)
    This is about as cheap as it gets for a DIY project. If I were to give you a quote for a commercial grade version you'd shoot me in the eye. [] []
    • Re: (Score:3, Informative)

      by rodrigo1979 ( 255519 )
      Here's the kind of meter we typically provide for commercial building applications: [] That's just the retail cost of the meter.. add the cost of the web appliance (Honeywell/Tridium) with I/O module or Lonworks/modbus interface, plus labor to the electrician for wiring/installation, graphics design, programming the appliance and commissioning the whole enchilada. Not cheap.
  • Check out the Tweet-a-watt from AdaFruit.

  • 1. Point web cam at power meter.
    2. Hack software to read pictures of meter
    3. ???
    4. Profit !!!
    • That is not a useful solution for real-time power monitoring when your digital-only house power meter only displays full kilowatt-hours used. (Hint, my stupid meter is that way).

    • by arivanov ( 12034 )

      If the meter is modern enough it is likely to have mBus over IR or some other IR based interface for automated reading.

      As far as the webcam, that may also work. It will not be very precise though.

  • The Energy Detective (Score:4, Informative)

    by swillden ( 191260 ) <> on Friday September 24, 2010 @05:12PM (#33692228) Homepage Journal

    It's a little more expensive that what you want -- $200 rather than $150 -- but other than that, I think it's exactly what you're looking for. The gateway device itself stores sufficient data to allow you to look at short-term detailed usage and long-term trends via its web interface, but if you want more than that, you can set up something to periodically poll the device, downloading detailed, per-second, usage in an XML format. You can then store that data however you like, and mine it however you want.

    There may be other solutions out there, and I'm interested to see what others suggest, but I have a TED unit and I couldn't be happier with it. It also uploads to Google PowerMeter.

    • Re: (Score:3, Informative)

      by Stele ( 9443 )

      Yep, absolutely fantastic system and not too hard to install yourself. The wireless receiver actually has a built-in web server you can log into from any computer, and "there's an app for that" too. You log into the server, tell it what your energy rates are, and it'll tell you instantly what the electricity is costing you. I have mine sending everything to Google PowerMeter (this feature is built in) which provides very accurate persistent usage data.

      I sprung for the optional remote which sits in our kitch

      • by swillden ( 191260 ) <> on Friday September 24, 2010 @06:23PM (#33692996) Homepage Journal

        First thing I learned after installing mine: the clothes dryer uses the most electricity by far, and leaving my computers on 24/7 doesn't use as much energy as I thought it did.

        I learned the same things. The clothes dryer, stove/oven and dishwasher dominate my power consumption. A microwave is an extremely efficient way to heat food. Computers are small users, even my dual-processor Opteron file server with eight hard drives only draws about 120W. The cool "multi-can" lighting systems in my kitchen, living and family room suck a lot of juice -- each room is about 800W with the lights on. My swamp cooler uses more juice than I thought it did.

        One thing I discovered the first day I installed the device was a "phantom" 400W draw that was pretty much always on. By shutting off all the circuit breakers one by one and watching the draw I was able to narrow it down and eventually discover that it was a large vent fan in my attic on a thermostat. It may have been necessary originally, but about five years ago I installed those spinning "hurricane" vents so my attic has good passive cooling -- but with that fan's thermostat set to turn the fan on at about 100 degrees, it was on nearly full-time during the summer. I turned the thermostat up to 120 and I don't think the fan has come on since. Turning it up hasn't appreciably affected the amount of time my swamp cooler runs.

        So far, I think I'm saving about $20 per month since installing the TED. It should pay for itself quite handily in a year's time.

        • and dishwasher dominate my power consumption

          I've tried to get organized with my dishwasher such that it doesn't dry the dishes electrically any more - I just use it to wash the dishes, then I open the door and pull out the racks and let them air-dry overnight. Works most of the time :)

        • The cool "multi-can" lighting systems in my kitchen, living and family room suck a lot of juice -- each room is about 800W with the lights on.

          I hate those can lighting systems, but if you have them then you should consider dropping some CFLs into them. Incandescent can lighting is basically as inefficient as it gets. There was also that article about the really efficient directional LED bulbs recently, but they're pretty expensive so I don't see how they'd be worth it.

          • I have considered it, but as of yet I can't find any CFLs that fit in the cans. I know they're really inefficient. They do light up the room really nicely, but the energy!
    • This is remarkable... exactly the sort of thing I hoped to find in the comments (I am not the OP). Thanks for pointing it out!

  • by SheeEttin ( 899897 ) <> on Friday September 24, 2010 @05:12PM (#33692238) Homepage
    If all you want is graphing, then Google Power Meter is probably the best way.

    That said, there have been a few articles on Hackaday recently [] concerning methods of interfacing meters with Google's API. I assume that once you submit it, you can get it back out.

    Or, if that doesn't do it for you, I'm sure you could adapt one of the projects on Hackaday to your own ends.
  • by mspohr ( 589790 ) on Friday September 24, 2010 @05:13PM (#33692242)
    This product is designed for whole house real time monitoring with the results available on the web... looks interesting. []

  • by dave562 ( 969951 ) on Friday September 24, 2010 @05:13PM (#33692246) Journal

    Take a look at what a monitored PDU costs for a server rack. APC offers them, as do a few other vendors. You're easily looking at $450 per PDU. It will do everything that you want though, including output to SNMP so that you can trap it.

  • Although designing something like this would be trivial with even a small amount of electronics knowledge, if you have none I'd first go toward one of the turnkey systems earlier commenters have suggested, then be a good nerd and pick up some books on electronics so next time you think of a project like this you'll be well armed.

    • Although designing something like this would be trivial ...

      Designing a circuit to measure power consumption, especially in as-billed dollars, can be surprisingly complicated. It really depends on the application. For many applications, measuring current consumption will correlate sufficiently well with power consumption, that many consumers will never know the difference.

      True power measurement is surprisingly complicated. There are 4 wires going into the average household, 2 lines, one neutral, and on

    • by necro81 ( 917438 )
      While the measurement circuits for what this guy wants are fairly straightforward, interfacing these with LIVE power at the breaker panel is very serious and dangerous stuff. A qualified electrician might be able to do some work in the breaker panel itself, though it isn't exactly standard work, so finding someone willing to give it a go with home-spun electronics is dicey. To do it safely would require the power to be cut to the house at the utility pole - before it even reaches the breakers. That's som
      • by mcgrew ( 92797 ) *

        You don't have to access any bare wires to measure current and voltage. You can loop a coil of insulated wire around the insulated current carrying lead and measure it by induction. This is how high voltage ammeters work.

  • by NevarMore ( 248971 ) on Friday September 24, 2010 @05:14PM (#33692262) Homepage Journal

    1) Wander around house, see if lights, appliances, devices are on/plugged in.
    2) Make arbitrary decision about power usage.
    3) Turn off/unplug device.

    There. Now go play outside.

  • by smellsofbikes ( 890263 ) on Friday September 24, 2010 @05:17PM (#33692296) Journal
    The Analog Designs ADE7763 is a pretty awesome chip for doing this sort of stuff. Here's the appnote in a pdf [], and here's the chip itself []. It's quite easily interfaced to an Arduino using SPI. I just laid out a board interfacing this to an ATMEGA1284 for doing power quality monitoring and logging, but it's for an internal project so I can't just hand out the code or layout, but it was a dead simple chip to work with: one crystal and two caps were all it required for support, and if it were interfaced to an Arduino, that could handle all the I/O to a computer or write to an SD card.
  • by Anonymous Coward

    See here for a commercial product that exploits this:

  • by drachenstern ( 160456 ) <> on Friday September 24, 2010 @05:18PM (#33692310) Journal

    Check with your power company, especially if you are with a smaller co-op. I write software that does the analysis and historical reporting on modern (aka, "smart", the kind that can phone home with readings on usage, peaks, etc, and all over the powerline itself) meters, and we have all that data like you're describing. More complex systems allow for complete home monitoring, but they do require some specialized devices inside the house.

    Here's a link (ok, the first on google I came across on the terms you need) but still, this will get your foot in the door. HTH. []

    If you're in with a bigger firm, sorry charlie, not much to suggest there.

  • CurrentCost Envi (Score:3, Informative)

    by Anonymous Coward on Friday September 24, 2010 @05:18PM (#33692314)

    I have an older version of the CurrentCost monitor..

    When I get some extra $ together will likely upgrade.

    • I've read that the CurrentCost monitor isn't very accurate as it has a SET power factor and doesn't calculate the power factor, producing results that can be off as much as 10% or more.
  • What I'm looking for currently: Some sort of device(s) that a) accurately measures power usage, b) allows me to access the data for storage in a database for my own graphing/analysis purposes, c) will work with MacOS (doesn't require Windows), and d) doesn't cost more than $150 or so.

    The old saying, "Cheap, fast, good. Pick two." applies here. You have too many requirements.

    • Re: (Score:3, Insightful)

      by geekoid ( 135745 )

      No he doesn't.

      He only needs two: Cheap and good. His requirement doesn't have time. Based on you post it will be perfect, cost him nothing and be done in infinity years... he might was to adjust a little.

  • by marcmerlin ( 48598 ) on Friday September 24, 2010 @05:22PM (#33692364) Homepage

    See []
    and see what you can output with one of those guys: []

    There is one caveat: you need windows for the initial setup, although I did it in vmware, maybe it works in wine too, but since then it's been running fine on linux (and it would work just the same on MacOS since it's a python script).


  • TED (Score:5, Informative)

    by Yossarian45793 ( 617611 ) on Friday September 24, 2010 @05:22PM (#33692374)
    I use TED []. It's right around your price range. It monitors whole-house power usage in real time and has a USB-Serial interface which you can easily suck data out of with Python script. I personally do all the data logging on a Linux box and export it through a web interface.
    • Re:TED (Score:4, Informative)

      by GruntMan ( 669188 ) on Friday September 24, 2010 @06:49PM (#33693230) Homepage

      I have a TED-5000. Very happy with it. 15-minute install in the main panel; the bigger hassle was resetting all the clocks in the house afterwards. Connected the gateway device to my home network, now any device that has a web browser can see power usage. Easily accessible from the outside world by web browser, with the right router settings. Monitoring is down to the second, with a claimed accuracy of +/-2%

      Nothing need be installed on the PC, and it doesn't rely on a PC to store data; the gateway device records the data and is the web server.

      The manufacturer seems pretty open; they publish the XML format and there are plenty of people reading the device with PHP scripts and logging to SQL databases for more flexible & permanent data storage. There are a few iPhone apps and I think there is a Android app, or talk about one. You can export the data from the gateway in second, minute, hour, daily, or montly format, with the follow capacities:

      ~2 days of per-minute data
      ~66 minutes of per-second data
      ~58 days of per-hour data (likely longer... I've only had mine for 58 days!)

      One caveat: the device that connects to the power panel (a pair of current clamps and a pair of voltage taps) communicates with the gateway via power line. Seems like many of the problems people have are related to power line communications, either due to electical noise or other power line communications devices (e.g. X-10) in the house. Some people have success with filters (extra cost), others never seem to solve these problems.

      I think it meets the poster's requirements for a), b), and c). It cost me $243 Canadian delivered to my door in 3 days from a Canadian supplier []

      No connection to either company here. Just a very happy customer.

  • It's beyond your budget, but a friend of mine was "showing" me his egauge []...

    About $500

  • I go to CMU and I did my capstone in Embedded Devices last year. One of the other groups was doing something that was almost exactly what the OP asked for. The project was called Jeules and you can probably still contact the team members to get some more information about it.
    Their wiki unfortunately is locked down now, but it used to have the exact parts list and some of the circuit diagrams to build the system. []

  • Maybe currentcost (Score:3, Informative)

    by Booker ( 6173 ) on Friday September 24, 2010 @06:06PM (#33692840) Homepage
    The currentcost meters are fairly cheap, OSX-capable I think, and very popular in Europe so there are lots of little scripts for them. In the us you can find them at []

    The DIY rig at [] is fairly straightforward, even if you're not that technically inclined....

    Otherwise I'd just echo the suggestion to suck it up for the extra $50 and get the Ted 5000

    My recent time-waster is finding a way to make all these different gadgets able to talk to all the various websites [] ...

  • What you want is a Zigbee enabled monitor in the home and a smart meter outside the home. The two can then communicate with each other. This shows entire home usage though, not just for a single device. Zigbee Alliance and WiFi Alliance are cooperating&#65279;, so I presume some sort geek friendly device exists.
  • I've built my own (Score:3, Informative)

    by baileydau ( 1037622 ) on Friday September 24, 2010 @08:03PM (#33693686)

    In conjunction with getting solar power at home, I've also set up real time usage monitoring.

    I've had a stand alone power monitor for a while (our state Government offered them + a bunch of CFLs and other stuff for $50). However it doesn't have any PC connectivity. One day I was looking in the meter box, and I realised that the sensor was just a clamp meter around the input wires.

    I already had a clamp current meter attachment for my multimeter (which also has RS232 out), so I put the clamp around the same incomming wires, connected it to my multimeter and then to my PC (via an RS232 -> USB cable). From there I have some scripts to take readings and enter them into a database as well as a web interface for output. Fortunately for me, the meter box is just outside of the room where the PCs are, so it was easy to wire up.

    I actually did this setup in a number of stages. Initially, I used some software for my multimeter to plot / save to text file the raw (amperage ) data. I then started batch importing it into the database and calculating wattages etc from there. Now it all happens automatically. Readings are taken at 1 minute intervals.

    Even though I already had all the parts, they cost well under your budget. From memory their original costs were:

        * clamp meter attachment (Digitek QM1565) $25 (see [])
        * multimeter with RS232 (Digitek QM1538) $50 (NB. this model is no longer available, don't know what an equivalent would be)
        * RS232 -> USB $6 (from eBay)

    Now I live in Australia, so your meter box setup may be different to ours. In mine, the meter and circuit breakers etc are mounted on a board in the box. This board has hinges on one side, so you can swing it out to get behind it. That's where the wires are that you need to put the clamp around. Obviously you want to be very careful back there, but there *shouldn't* be any bare wires etc. If in doubt, you could get an electrician to do this for you.

    I've put a sample of our median usage and production on Imageshack []

    Here is the usage and production for a single day []

    Having this type of data is great for tracking down where your usage is going.

  • They'll call you about once a day, and you divide what they demand right now by your cost per kilowatt/hour.
  • by dj.delorie ( 3368 ) on Friday September 24, 2010 @08:17PM (#33693768) Homepage

    I had that desire too, but my electronics skills were up to an overkill DIY solution... []
    I record watt-seconds for each of 64 circuits once per second to a linux server.

    • That's a nice board. Checkout our light dimmer [] that includes current sensing and 6lowpan radio. We used a shunt design which has much lower cost.

      There are similar chips to the ADE7753 out now with costs close to $1. You can probably source the CTs in China for less than $1.

    • by ( 463190 ) on Friday September 24, 2010 @10:47PM (#33694558) Homepage
      This is very cool! But you have a big problem unless I'm missing something. I only briefly skimmed your code so maybe I missed this, but it doesn't look like you are accounting for power factor. In order to do that you need to measure the voltage on each phase, ideally at a few KHz, and generate CT samples at the same rate which are multiplied by this measurement. This way you properly deal not only with low PF loads, but also variance and distortion in the AC line voltage supplied by the utility.

      Some devices can have very low PFs, for example insteon switches and other small loads, and lightly loaded switching power supplies, it can be as little as 0.1x. A ceiling fan running at low speed, or a CFL might be something like 0.4. So the number you are calculating is properly called VA (volt-amps) and is not the same as watts, which is what you're actually consuming and being billed for.

      I see you have put calibration factors in for each circuit. You may find that the reason you're needing these at all is because those loads are low PF and are reading higher than they should.

  • Plenty of solutions that will work if you are willing to poll serial links for mod bus registers at a reasonable cost.

  • What's the cheapest Zigbee/controller/sensor that I could cook up to do this? Then I could write Linux apps to manage the data from the Zigbee network.

At work, the authority of a person is inversely proportional to the number of pens that person is carrying.