XPost: sci.electronics.equipment   
   From: dmichel@prosound.de   
      
   Hi,   
      
   > I just bought an amp clamp meter, and it states the error is "+/- 1.9% +   
   > 3 digits".  What does the "3 digits" part mean?   
      
   I'll try to explain that with a simplified model of   
   a digital meter (please everybody correct me if it's   
   oversimplified and wrong):   
      
   The typical digital meter consist of some kind of   
   processing of the signal to be measured and an   
   A/D-converter that converts it's analog input   
   signal to a number that is displayed.   
      
   The input processing serves to transform the quantity   
   to be measured into an analog signal that is properly   
   adapted to cover the range of possible input signals   
   of the A/D-converter.   
      
   Let's say we want to measure an AC current of 10A with   
   a clamp meter like yours. Let the display of the meter   
   have 3 1/3 digits, so the range of displayable numbers   
   goes from 0000 to 1999 with an additional decimal point   
   somewhere.   
      
   The A/D-converter will not be able to directly convert   
   a 10A current, so we pickup the current to be measured   
   with a transformer, the wire carrying your 10A current   
   being the primary and a coil internal to the clamp   
   assembly being the secondary winding.   
      
   An AC current flow through the wire will induce an AC voltage   
   in the secondary winding. Since the A/D-converter may not   
   directly accept AC voltages, further processing may be   
   required, such as amplification or voltage division and   
   e.g. True-RMS detection of the AC voltage. All this   
   processing will end up in a voltage that is suitable for   
   the A/D-converter - say, 1V DC for 10A of AC current.   
      
   All the (analog) signal processing described here will   
   not be free of unwanted influences and processing errors.   
   The transformer at the input could e.g. pick up unwanted   
   magnetic fields, the amplifier could exhibit noise and   
   nonlinearities, the TRMS detection could exhibit some errors.   
      
   All these error sources or influences may be described in   
   the meter's specification as a percentage - e.g. the +/- 1.9%   
   you mentioned.   
      
   Now, the A/D converter converts the analog input voltage   
   into a number. One method to do this, when speed is not   
   a critical factor, is (dual) slope integration.   
      
   Let's assume for a moment that the input voltage is static,   
   i.e. the 1V DC mentioned before.   
      
   Basically, the conversion works by comparing the input   
   voltage (to be measured) to a linearly rising voltage   
   (ramp). Similar to a stopwatch, a counter starts when   
   the reference voltage begins to rise and a comparator   
   stops it when the ramp voltage is equal to the input voltage.   
      
   In our example with 1V input, the counter may stop at a   
   count of 1000. With the knowledge that, by means of the   
   input processing and the calibration of the meter, this   
   corresponds to 10A AC current. The meter would probably   
   display 10.00 (A).   
      
   But: At some time in the process, the counter will switch   
   from 999 to 1000 in a very short (almost zero) time.   
   That means, that the input voltage may just be a tiny little bit   
   less and the counter is stopped at 999, not at 1000.   
      
   That means, that for any input signal, you always have +/-1 digit   
   display uncertainty because you cannot know whether the   
   counter maybe was just before switching to the next count.   
      
   With a specification of +/-3 digits, the A/D converter has   
   a greater uncertainty when counting. For example, even at   
   a constant input of 1V, the internal counter may be less   
   precise and stop at 997, 998, 999, 1000, 1001, 1002 or 1003,   
   even if the input signal doesn't change. You can think of   
   this as a stopwatch that may be off some counts each time   
   you make a measurement.   
      
   This type of error is not related to the input signal   
   processing, so it is not very meaningful to express the   
   error as a percentage of the measured value. It is usually   
   expressed as a number of digits, because the error is mainly   
   caused by the process of converting input signals to numbers.   
      
   Of course, I know that this very simple single slope integration   
   is not used in meters, dual slope is the least you can do.   
   Also, the A/D conversion may contribute to the percentage error spec.   
   The (over)simplification is just a means to explain why there   
   are two numbers in the specification.   
      
   Just my two cents,   
      
   Dieter   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)