A challenge: Home power consumption monitor project.

Discussion in 'The Projects Forum' started by hithere, Nov 7, 2009.

  1. hithere

    Thread Starter New Member

    Nov 7, 2009
    Hi everyone, I am new here although have used some of the good info of this forum before.

    I want to create this new project and although I have do tones of research I would like to see others opinions.

    The idea is to monitor the power consumption of a house.

    The method: From what I have research the most appropriate method would be using several Rogowski coils.

    Why this method:
    1. you dont need to cut or manipulate power wires of the house, just put the rogowski coil around it.
    2. shunts, amperimeters, etc involve risk of manipulation, heat creation, etc
    3. should be easy to build 4 or 5 and monitor different wires coming from the main power box of the house.

    How to do it.
    My idea is to create 4 or 5 circuits as shown above and connect them to a pic with an equal amount of A2D converters. the pic will be in charge of reading the value and then sending it via rs232 to a computer i have 24/7 on. the computer will receive the values and store/create graphs/statistics etc with the data. Also an lcd will show useful information as instantaneous consumption of each power line, totals for days/weeks/month etc etc.

    I have a wattmeter which i will use to "calibrate" as far as possible the values received by the pic.

    Similar commercial products
    (which do not fulfil my intentions)
    Whole project, really nice but expensive ==>http://www.kondra.com/circuit/circuit.html


    What I need:
    well the thing that worries me more is building the Rogowski coil.
    -Has anyone got experience with it?
    -Any suggestions / schematics about it?
    -about the op amp, I really dont know which one to use, I am completely new to them. Here is a list of ones I have available, but cant really tell differences:

    I am not an electronic expert so any help is welcome.

    Thanks for any help and hopefully we can use this post as my project progresses. Ill post info + pics of the project.

  2. t_n_k

    AAC Fanatic!

    Mar 6, 2009
    Hello hithere,

    You can buy split core current transducers which take all of the guesswork out of the current sensing side. So no breaking into conductors is required. A typical AC current tong tester falls into the same category.

    For example on the split core

    Building and calibrating a Rogowski coil is an interesting process - I've made a couple over the years. I was more interested in very fast pulsed currents often of high magnitude - or simply very high currents which would saturate a conventional current transformer.

    Good luck:)
  3. hithere

    Thread Starter New Member

    Nov 7, 2009
    thanks t n k , I had seen the option about buying them "made" but the problem is finding them. I would really prefer to at least buy 1 or 2 and develop the project with them. Once things are working then think about building a couple of them more (if it is cheaper than buying them already build.)

    As i say, the problem is finding them. and maybe the price if I do find them.

    So, with one of those you linked. Would be just a matter of feeding the transducer and reading the output value? just connect it to an AD input of the PIC and calibrate the readings

    thanks for your help.
  4. hithere

    Thread Starter New Member

    Nov 7, 2009
    t n k, I just found a website which I could buy from. its in spanish but it is easy to read.

    could you recommend me what type or characteristics should I be looking for?

    this is not for professional use, so cheap is important.


    oh, i forgot, I guess that it should be able to read between 0 to 20 Amps for the big power wires, or maybe 0 to 10 Amps for subcircuits.
    Last edited: Nov 8, 2009
  5. t_n_k

    AAC Fanatic!

    Mar 6, 2009
    Unfortunately I can't make recommendations concerning particular devices - this is a requirement of forum membership.

    If you simply want signals for linking through A-to-D & micro-controller interface then ideally you would select a current transducer which will give you (say) 0-5 or 0-10V output for 0-20A primary input current. These are going to be more expensive than a simple current transformer [CT], and they require a suitable DC power supply which adds to the overall cost. It looks like the split core transducers you found are quite expensive. The straight split core CT option, while cheaper (probably around $100 USD), will still require additional electronics to convert the secondary side current signal to a useful input signal for your measurement system. Also it may be harder to find a CT with a suitable range - although a 100:5 (20:1) option might do the trick with the right electronics.

    I agree you should start simple and establish the cost & feasibility of setting up a single measurement. Keep in mind also that simply measuring current won't necessarily give you the whole story on your power consumption - particularly if any of your loads have a low (or significantly less than unity) power factor. In principle you need to multiply the instantaneous phase voltage and current signals and take the average of this to give you the average or real power.
  6. hithere

    Thread Starter New Member

    Nov 7, 2009
    once again, thank you for your time.

    From what I have seen, prices are pretty high for off the shelf rogowski coils.

    So to keep this project cheap, at least initially, I will try to make my own. I have seen a couple of websites that people build their own to measure current in their projects so I guess its not that bad idea.

    hopefully I will manage to get it working and with a minimum of precision which can make the readings useful.

    so for example, if I want to send the Vout from the first schematic of this thread, to the A2D pin of my pic (16f690) I just plug it in? I dont know the max Voltage of the pic A2D. or if i need to protect it or something else.

    thanks again.
  7. n9352527

    AAC Fanatic!

    Oct 14, 2005
    It is maybe worth looking at Analog Device energy metering ICs. There are models that interface directly to rogowski coils. Also there are extensive application notes available at the site.

    One of the advantage using a rogowski is better accuracy across wide measurement range, extending for several decades. There are many commercial meters in use that are based on rogowski.