Home  5  Books  5  GBEzine  5  News  5  HelpDesk  5  Register  5  GreenBuilding.co.uk
Not signed in (Sign In)

Categories



Green Building Bible, Fourth Edition
Green Building Bible, fourth edition (both books)
These two books are the perfect starting place to help you get to grips with one of the most vitally important aspects of our society - our homes and living environment.

PLEASE NOTE: A download link for Volume 1 will be sent to you by email and Volume 2 will be sent to you by post as a book.
Buy individually or both books together. Delivery is free!


powered by Surfing Waves




Vanilla 1.0.3 is a product of Lussumo. More Information: Documentation, Community Support.

Welcome to new Forum Visitors
 Join the forum now and benefit from discussions with thousands of other green building fans and discounts on Green Building Press publications: Apply now.




Data and monitoring: Vampire hunting with OWL

Bottom of Page

31 to 42 of 42

  1. <
  2. 1
  3. 2
  4. >
    •  
      CommentAuthorDamonHD
    • CommentTimeJun 5th 2012
     
    Though I did notice an EU directive or somesuch stating that new energy efficient lighting had to have a PF very close to 1, so that may help in the longer term...

    Rgds

    Damon
    • CommentAuthorborpin
    • CommentTimeJun 5th 2012 edited
     
    Not sure if this is relevant as I understand about 1/3 of what is going on, but the Open Energy Monitor posted this blog recently http://openenergymonitor.blogspot.co.uk/2012/06/high-accuracy-current-measurement-over.html
    •  
      CommentAuthorJSHarris
    • CommentTimeJun 5th 2012
     
    <blockquote><cite>Posted By: borpin</cite>Not sure if this is relevant as I understand about 1/3 of what is going on, but the Open Energy Monitor posted this blog recently<a rel="nofollow" href="http://openenergymonitor.blogspot.co.uk/2012/06/high-accuracy-current-measurement-over.html">http://openenergymonitor.blogspot.co.uk/2012/06/high-accuracy-current-measurement-over.html</a></blockquote>

    That looks to be a good approach to improving accuracy, certainly a far better method than is used by most (maybe all) of the commercial clamp on type sensor systems I'm aware of. Perhaps it will filter through to the commercially available devices before too long.
    • CommentAuthorEd Davies
    • CommentTimeJun 6th 2012
     
    Umm, well, yes, I did say earlier “It's also possible they have a programmable gain amplifier on the front of the ADC so the resolution at low power is better than at high power”. I was thinking of something like this:

    http://jeelabs.org/ap2

    (follow the links - I've used the URL which should be persistent here). It's an up to 18-bit ADC with a 1x to 8x programmable preamp all on one chip. Not quite as wide a range as in that Atmel circuit but the extra bits of ADC resolution will compensate. I have one for a battery monitoring application but haven't got round to playing with it yet.
    • CommentAuthorJoiner
    • CommentTimeJun 6th 2012
     
    • CommentAuthorrhamdu
    • CommentTimeJun 7th 2012
     
    Posted By: borpinNot sure if this is relevant as I understand about 1/3 of what is going on, but the Open Energy Monitor posted this blog recentlyhttp://openenergymonitor.blogspot.co.uk/2012/06/high-accuracy-current-measurement-over.html" >http://openenergymonitor.blogspot.co.uk/2012/06/high-accuracy-current-measurement-over.html


    I don't know much about electronics but I don't think there is any big technical problem in designing an analog-digital converter which operates over a wide range. At a price, of course.

    I am beginning to wonder if there is a problem with the clip-on current transformers themselves. Do they give a true zero reading? It seems unlikely. Even when no current is flowing through the meter tail, there is an alternating electric field around it, and the sensor would have to be very clever to ignore that altogether.

    To relate this to something of which I do have a bit of experience: what would be the chances of making a hum-free audio recording, with the microphone a few millimetres from a 25mm2 cable energised with 230V AC?
    •  
      CommentAuthorDamonHD
    • CommentTimeJun 7th 2012
     
    As stated above: only reading the current introduces several sources of (varying) errors, whatever accuracy/precision the current is measured at.

    Rgds

    Damon
    • CommentAuthorrhamdu
    • CommentTimeJun 7th 2012
     
    Sorry, Damon, I didn't make myself clear.

    We all agree that only reading the current introduces errors. It is also established that there are inaccuracies in analog-digital conversion.

    My (new?) point was that the physical sensors - clip-on current transformers - may not even be all that good at the limited task they are designed to perform - which is to read the current.

    The sensor should output zero current when zero current flows through its primary (i.e. the meter tail). I'm saying that is unlikely to be achieved, because if you take any piece of wire connected to an amplifier and poke it that close to mains voltage, it is going to pick up some 'hum' electrostatically. The devices are presumably designed to minimise this, but they are unlikely to eliminate it altogether. The resulting non-zero reading could be interpreted as a 'vampire' power drain when actually no power is being used at all.
    •  
      CommentAuthorJSHarris
    • CommentTimeJun 7th 2012
     
    <blockquote><cite>Posted By: rhamdu</cite><blockquote><cite>Posted By: borpin</cite>Not sure if this is relevant as I understand about 1/3 of what is going on, but the Open Energy Monitor posted this blog recently<a href="<a href=" rel="nofollow">http://openenergymonitor.blogspot.co.uk/2012/06/high-accuracy-current-measurement-over.html</a>" >http://openenergymonitor.blogspot.co.uk/2012/06/high-accuracy-current-measurement-over.html</blockquote>

    I don't know much about electronics but I don't think there is any big technical problem in designing an analog-digital converter which operates over a wide range. At a price, of course.

    I am beginning to wonder if there is a problem with the clip-on current transformers themselves. Do they give a true zero reading? It seems unlikely. Even when no current is flowing through the meter tail, there is an alternating electric field around it, and the sensor would have to be very clever to ignore that altogether.

    To relate this to something of which I do have a bit of experience: what would be the chances of making a hum-free audio recording, with the microphone a few millimetres from a 25mm2 cable energised with 230V AC?</blockquote>

    Getting a good A to D at better than 16 bit resolution and accuracy is only moderately challenging, but it's very difficult to mass produce something of this resolution and accuracy without individual calibration and trimming, which would price such a thing out of the domestic energy monitor market. For example, if you use a common 3.3 V power supply to the chip, then 1 bit at 10 bit resolution is just over 3.2 mV. This is just about OK for a circuit that doesn't need trimming or calibrating to be mass produced. Go to 12 bits and a single bit drops down to 800 µV, 16 bits and a single bit drops further down to just 50 µV. It's pretty difficult to mass produce circuits that will work repeatably and accurately down at these low signal levels without them needing some sort of trimming or calibration process after manufacture.

    The clip on current transformers are reasonably good at only picking up the current flowing through the core of the toroid, and are fairly immune to stray flux I believe. As Damon has already said, the big errors come from elsewhere, like not actually measuring the supply voltage.
    •  
      CommentAuthorJSHarris
    • CommentTimeJun 7th 2012
     
    <blockquote><cite>Posted By: rhamdu</cite>Sorry, Damon, I didn't make myself clear.

    We all agree that only reading the current introduces errors. It is also established that there are inaccuracies in analog-digital conversion.

    My (new?) point was that the physical sensors - clip-on current transformers - may not even be all that good at the limited task they are designed to perform - which is to read the current.

    The sensor should output zero current when zero current flows through its primary (i.e. the meter tail). I'm saying that is unlikely to be achieved, because if you take any piece of wire connected to an amplifier and poke it that close to mains voltage, it is going to pick up some 'hum' electrostatically. The devices are presumably designed to minimise this, but they are unlikely to eliminate it altogether. The resulting non-zero reading could be interpreted as a 'vampire' power drain when actually no power is being used at all.</blockquote>

    Sorry, missed this.

    The impedance of the clip on current sensors is very low, so they are extremely unlikely to pick up anything other than the flux from the cable running through the core. They are quite unlike something like a microphone, that usually has an impedance that is pretty high, in the hundreds of ohms range.
    •  
      CommentAuthorSteamyTea
    • CommentTimeJun 7th 2012
     
    Does seem a shame that as most of them run if a small mains power supply that they cannot sense the mains voltage. Or just a 10 quid plug in power meter with some logging would be useful to calibrate things.
    •  
      CommentAuthorJSHarris
    • CommentTimeJun 7th 2012 edited
     
    <blockquote><cite>Posted By: SteamyTea</cite>Does seem a shame that as most of them run if a small mains power supply that they cannot sense the mains voltage. Or just a 10 quid plug in power meter with some logging would be useful to calibrate things.</blockquote>

    I suspect the reason they don't sense voltage using the plug in power supply has to do with cost and the need to maintain double insulation. The cheap way to sense the mains voltage would be to just stick a resistive potential divider on it and sample it at a high enough frequency to measure the true rms waveform. This would fall foul of the double insulation requirement, though, and mean using proper mains-rated cable and connectors from the power supply to the unit, as well as other stuff to keep it "safe" (from a certification viewpoint).

    The safe way to sample the incoming voltage would be via a small isolating transformer, but although that then gets around the double insulation problem it gives two other problems, cost and accuracy. The transformer will likely not be very accurate in terms of ratio (even 10 bits needs a transformer with a ratio accuracy of around 0.1%) plus transformers tend to add cost and these things are built right down to a price (my guess is that they have a total manufacturing cost of less than Ă‚ÂŁ10).

    If you have one of the newer digital electricity meters then you can pretty easily use that to measure consumption pretty accurately. They have an LED on the front that flashes once every 1 Wh (or at least mine does). A simple optical sensor stuck to the front of the meter will give you a measure of true power consumed, all you need to do is keep track of the pulses. I believe one or two power monitors already use this technique.

31 to 42 of 42

  1. <
  2. 1
  3. 2
  4. >
Add your comments

    Username Password
  • Format comments as
 
Back to Discussions Top of Page
   
The Ecobuilding Buzz
Site Map    |   Home    |   View Cart    |   Pressroom   |   Business   |   Links   
Logout    

© Green Building Press