The highest possible precision in power and energy calibration is a matter of great interest, in particular to national metrology institutes (NMIs). As a result of technological progress in accuracy of power and energy measurement, an energy meter better than class 0.1 is available.

There has also been a change in metrology systems to ensure traceability of reference standards at the meter manufacturer and utility end. To maintain the proper metrological traceability, utilities and manufacturers are using high precision power and energy standards for internal intercomparison, rather than sending each meter testing standard to an NMI (see Figure 1). This saves time and money at the user end, but raises questions about the calibration of these units at the NMI end.

Earlier published principles reported achieving uncertainties of the order of 10x10-6 to 30x10-6 based on digitally synthesised voltage and current, sampling voltmeter measuring methods and computerised evaluation of measurement data. Other methods such as the three voltmeter method have not yielded uncertainties below 15x10-6 for the unity power factor and 50x10-6 for other power factors. [1]


A team of scientists from PTB Braunschweig (National Metrology Institute of Germany) has developed a new system for generating and measuring active, reactive and apparent power for power frequencies of up to 400 Hz, combined with generation and measurement of single phase sinusoidal voltage, current with phase shifts adjustable between 0° and ± 180°. The uncertainty of the active, reactive and apparent power (related to the nominal value of apparent power of 600VA) for any value of power factor is 2.5x10-6 (k = 1) for frequencies from 45 to 65 Hz. [1]

Fig. 1: Metrology system at manufacturing or utility end

ZERA GmbH acquired the right to use this technology and added value in the area of measuring range expansion, conversion from the laboratory model to easy use for all users around the world, with software for easy calibration and control.


The operating principle is based on synthesised ac voltage and current generation, using a single sampling voltmeter for synchronisation and computerised evaluation by means of discrete Fourier transform (DFT).

PC software communicates with the frequency generator (FG115) which is a dual voltage programmable amplifier. A signal passes to the voltage and current amplifier (based on linear amplifier technology) through an antialising filter (FES101). Accurate voltage and current transformers are used for galvanic isolation before the power calibrator is put under test. Voltage output is again measured through an inductive voltage divider, with feedback to FG115. Current output is measured using a two-stage current transformer and high precision resistors, and the measured value is fed to FG115 for corrective action. PC software also communicates to the device under test and collects the value measured, performs the comparison and records the error.

The traceability to the SI units dc voltage and dc resistance is ensured by the RMS voltmeter and ac resistor with small and well-known frequency characteristics. [2] The reduction of the measurement uncertainty as compared with other methods is due to the use of a single clock for both generation of the required voltage and for evaluation of these signals. This clock signal fclock is taken from the sampling voltmeter. These measures lead to a significant reduction of synchronisation errors with sampling method, and unavoidable differences between sampling voltmeters. [1]


  • Ultra precision voltage, current and power calibration.
  • Measurement uncertainty in power is in the range of 10 ppm at all power factors.
  • Simply traceable to dc standards, which have low uncertainty and are easily available in many national laboratories.
  • Fast, highly stable and repeatable measurement.
  • Covers all power frequency ranges 40 Hz to 60 Hz.
  • Voltage ranges 480V/240/120/60 V.
  • Current range 0.1 A to 10 A (further extension up to 100 A is possible).
  • RMS value of voltage, current, power factors and power (active, reactive and apparent) displayed on the PC screen.
  • Automatic calibration of power calibrator and calculation of measurement uncertainty is possible.
  • Industrial design (19" cabinet with rear lockable door) and modularity of system offer various advantages.

Fig.2: Operating principle


  1. G. Ramm, A. Braun, and H. Moser : A new scheme for generating and measuring active, reactive and apparent power at power frequencies with uncertainties of 2.5 x10-6 – IEEE IM Vol. 48, no 2, April 1999.
  2. G. Ramm and R. Vollmert : Development and setting up of an AC resistance measuring system – PTB-Bericht E-41, Nov. 1991,
    pp. 1-32.