1. Limited time only! Sign up for a free 30min personal tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Compounding Accuracies

  1. Oct 9, 2014 #1
    Hey Guys,
    I'm not sure the correct procedure for comounding errors using some test equipment. I'm trying to measure electrical current using a fluke meter and a fluke current clamp. Here are the two devices I'm using specifically:

    Fluke Power Analyzer Model 437 Series 2
    Fluke i5sPQ3 AC Current Clamp

    Here is where it is confusing. The Fluke meter has an accuracy spec as well as the current clamp. For instance, the meter says in the range 0.5 - 200A, the accuracy is 0.5% + 5 counts. The clamp says the accuracy is 1% + 5mA in the range of 10mA - 1A.

    What is the appropriate way to combine these two accuracies to establish a single accuracy? I would think to add them, but I've seen some people that seem to take an average, using a RMS-type formula.

    Any suggestions?
  2. jcsd
  3. Oct 10, 2014 #2


    User Avatar
    Science Advisor

    Meter ±0.5% ±5 counts.
    Clamp ±1% ±5mA, in the range of 10mA to 1A.

    Combined ±1.5%, ±5 counts, ±5mA, over the range of 10mA to 1A.
  4. Oct 11, 2014 #3

    jack action

    User Avatar
    Science Advisor
    Gold Member

    Say you have 1.0 A on your meter. The meter accuracy says that the actual current measured is within the largest range between ±0.5% (0.995-1.005 A) and ±5 counts (0.5-1.5 A); 0.5-1.5 A it is.

    Then the current sent by the clamp (which is anything between 0.5 A and 1.5 A) could have been within the largest range between ±1% (0.495-1.515 A) and ±5mA (0.495-1.505 A); Hence the actual current of your 1.0 A measurement could be anywhere between 0.495-1.515 A.

    Doing the same procedure with a 0.5 A measurement yields:

    • ±0.5%: 0.4975-0.5025 A
    • ±5 counts: 0.0-1.0 A (largest)
    • ±1%: 0.0-1.01 A
    • ±5mA: (-0.005)-1.005 A
    Possible actual current measured: (-0.005) - 1.01 A

    If you used your clamp with the 1-5 A range with a 1% accuracy, then with a 5.0 A measurement, the actual current range is:

    • ±0.5%: 4.975-5.025 A
    • ±5 counts: 4.5-5.5 A (largest)
    • ±1%: 4.455-5.555 A
    Basically, the ±5 counts on the 0.1 A resolution of your meter is killing the accuracy of your clamp.
  5. Oct 11, 2014 #4


    User Avatar
    Science Advisor
    Gold Member

    When using these accuracies to determine an uncertainty of a measurement, only comparison with a standard will provide a way of eliminate or compensate for any calibration errors.
  6. Oct 11, 2014 #5


    User Avatar
    Science Advisor

    I note that the “Fluke i5sPQ3” AC Current Clamp appears not to be a current ratio transformer, but is specified to produce a voltage output of 400 mV/A. I expect that requires the “Fluke Model 437 Series 2” Power Analyzer then be used as a volt meter on the range up to 5A * 0.4V = 2V FSD.

    But the 437-II meter range is specified as Vrms 1V to 1000V phase to neutral, with resolution 0.01 V ± 0.1% of nominal voltage.
    for 0.5A * 0.4V/A = 0.2V, 0.01V is 20:1 = 5%
    for 1.0A * 0.4V/A = 0.4V, 0.01V is 40:1 = 2.5%
    for 2.0A * 0.4V/A = 0.8V, 0.01V is 80:1 = 1.25%
    for 5.0A * 0.4V/A = 2.0V, 0.01V is 200:1 = 0.5%
    So the limitation is not always the 1% accuracy of the clamp, but is the resolution of the voltmeter.

    To get more accurate results it would seem necessary to use a higher resolution voltmeter, change to a current transformer clamp, or use a voltage amplifier between the clamp and analyser.
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Similar Discussions: Compounding Accuracies
  1. Accuracy of sensors (Replies: 6)