Device and Method for calibration of an electricity meter
TECHNICAL FIELD
The present invention relates to an electricity meter for measuring the electrical energy output in a power network. More specifically, the invention relates to a device and a method for calibration of an electricity meter. In particular, the invention relates to an electricity meter com- prising electronic circuits for adjusting the calibration. In the following text, the term electronic circuit shall mean an electric circuit comprising at least one semiconductor and comprising a current path with one or more impedances .
BACKGROUND ART
A plurality of electricity meters exists for measuring the electrical energy output. In addition to the common so- called Ferraris meter, which is based on the induction principle, a plurality of electronic electricity meters are available. Signals from current sensors and voltage sensors, respectively, provide values to the electronic electricity meter. These sensors are usually integrated in the electri- city meter which also comprises a multiplier and an output member. The principle of measurement according to which the multiplier is arranged may be the so-called TDM (Time Division Multiplication) method, methods involving microprocessors and analog-to-digital (A/D) converters, or accor- ding to the so-called HALL effect.
Common to the principles of measurement is that calibration is required to achieve a desired accuracy. This is required because of variation and imperfections in the components included. It may be a question of current and voltage sensors as well as active and passive components included in the electronics which attend to the signal processing. These include resistors, capacitances, inductances etc. and, among
the active components, for example amplifiers, A/D converters, microprocessors, etc.
When measuring active electrical energy, at least one current and one voltage are always measured, the product of which is integrated over the time that energy is to be measured according to the formula:
where u(t) constitutes the voltage and i(t) the current. Both current and voltage are dependent on the time (designated (t) in the formula) . Since the magnitude of the signal from the sensors is different from the magnitude of the measured signal, these have to be compensated for by multiplication by an amplification constant K, such that the absolute value of the energy attains the correct level . The above-mentioned formula is then changed to read:
where U(t) and I(t) constitute the output signals from the voltage and current sensors, respectively.
Because of imperfections in voltage and current sensors, a certain distortion is always obtained, such that U(t) and I(t) are not constantly proportional to u(t) and i(t) . An example of a distortion which usually has to be considered during calibration is a phase-angle error.
When assuming an alternating current and an ac voltage without harmonics with the frequency f and with a phase- angle error δ between current and voltage, the following energy formula is obtained:
jK(U ) *U sm(wt) * I sm(wt + φ- δ(U, It f,T)) *dt
where φ is the phase angle between current and voltage and w is the angular frequency (2*π*f ) . Because of distortion and non-linearity in sensors and other components, the amplification constant K and the phase-angle error δ are dependent on the voltage, the current, the frequency and the temperature (T) . Depending on the magnitude of the distortion and the requirements for accuracy, one or more of these dependencies may normally be neglected.
During calibration, an electricity meter is normally placed in a test bench where it is influenced by given currents and voltages and is calibrated such that the measured electrical energy lies within an accuracy interval specified in advance. The calibration is either carried out manually, normally by adjustment of one or a plurality of potentiometers, or with a varying degree of automation. During such automatic calibration, a computer is usually connected to the electricity meter, whereby a program is executed which determines the correct values of the amplification constant, the phase-angle error and other constants. Its correct values are referred to as calibration constants. The computer program attends to the calibration constants being stored in the electricity meter, usually in non-volatile memories which may be of the EEPROM type.
From US 4,682,102, a so-called WH (WattHour) meter is previously known. The known meter comprises a switchable capacitor integrator. Another electricity meter is previously known from DE 3,529,472. This is a so-called
Etalon wattmeter with a double-layer current transformer. The meter utilizes potentiometers for eliminating an offset of the amplifier in an output stage.
SUMMARY OF THE INVENTION
The object of the present invention is to suggest ways and means of providing an electricity meter which is inexpen- sive and simple to manufacture. The electricity mete shall measure energy output with a high accuracy and allow a simple and time-saving calibration. During calibration, the amplification constant as well as phase-angle errors and imperfections in the components included are adjusted in an automated way with a minimum of manual operations .
This object is achieved according to the invention by an electricity meter according to the characteristic features described in the characterizing portion of the independent claim 1 and by a method according to the characteristic features described in the characterizing portions of the independent method claim 9. Advantageous embodiments are described in the characterizing portions of the dependent claims .
An electricity meter comprises a plurality of electronic circuits, through which current flows for influencing the properties of the meter. The current is varied by introducing various impedances in the current path. The impe- dances thus usually arranged are static in the sense that, at each moment, the impedance limits the current.
By instead arranging a switching device in the current path, a "virtual" impedance is obtained over time, which is varied by closing and opening the switching device. Within the scope of the invention, the terms "impedance" and "resistance" are used synonymously, and in the following text the term "resistance" will preferably be used. When placing such a switching device in series with a resistor R, a mean resistance is obtained which is dependent on the time during which the switching device carries current and the time during which the switching device prevents current from flowing in the current path. The ratio of conducting time to the sum of conducting time and non-
conducting time is called pulse ratio (or duty cycle) . At a duty cycle of 50%, the mean resistance in the described connection will be 2R during integration over time.
If a switching device is placed in parallel with the resistor R, a mean resistance of R/2 is obtained with the same reasoning as above. In this way, a plurality of impedances and switching devices may be arranged in the current path in order to achieve the intended effect. The method is primarily suitable to use for eliminating potentiometers when using principles of measurement which traditionally have required this, such as according to the above-mentioned TDM principle and the Hall-effect principle, thus reducing the component cost and eliminating the need of manual, time-demanding operations when calibrating such a meter.
One suitable switching device is a so-called analog switch. This comprises a semiconductor circuit with a possibility of carrying current in both directions. For most applications it may be assumed to function as an ideal switching device or circuit breaker (zero ohm when it is closed and infinite impedance when it is open) but contains in practice always a certain resistance. The switching of an analog switch is carried out in a simple manner by a control signal applied thereto. The switching device may thus, in a very simple manner, be brought to respectively open and close a current path based on an applied control signal in the form of a pulse train. Such a pulse train may be periodic or non-periodic. With a non- periodic pulse train, the risk of interference with other time-dependent functions within the meter or its connected units may be avoided.
To control the pulse train, the electricity meter is equipped with a calibration device comprising a member for generating a pulse train. The calibration device also comprises a device for controlling the pulse train. This device is usually a microprocessor which, on the basis of
calibrated constants stored in the calibration device, controls the switching device. These constants are values for the frequency and the previously mentioned duty cycle of the switching device in question.
To calibrate the meter comprising current paths with analog switches, the electricity meter is arranged in a test bench where it is connected to a network with a known consumption. The known consumption is fed into a computer connected to the electricity meter for the calibration. Usually, the computer is connected by an optical interface. Starting from a given calibration program, the ratio of conducting time to non-conducting time of the analog switches is varied until a desired accuracy has been achieved. The values obtained are stored in non-volatile memories in the electricity meter.
BRIEF DESCRIPTION OF THE DRAWINGS
The invention will be explained in greater detail by description of a few embodiments illustrated by means of figures, wherein
Figure 1 shows a current path with an analog switch in series with a resistor R according to the invention,
Figure 2 shows a current path with an analog switch in series with a resistor Rl and in parallel with a resistor R2 according to the invention,
Figure 3 shows a current path with a plurality of resistors and analog switches for calibration of amplification factor K and phase angle φ, and
Figure 4 shows a current path with an integrator, resistors and an analog switch for calibration of offset of the operational amplifier used in the integrator.
DESCRIPTION OF THE PREFERRED EMBODIMENTS
The invention is based on the fact that the impedance in a current path can be adjusted by bringing a switching device, connected to the current path, to close the current path, over an integration time, during a first period of time and to open the current path during a second period of time. By specifying the ratio of the two periods of time, the desired time-integrated impedance can be obtained.
The current path shown in Figure 1 comprises a resistor R in series with an analog switch 10, shown in the figure, which is controlled by a pulse train. During a first period of time, the switching device carries a current I, whereby the resistance in the current path is R. During a second period of time, the switching device breaks the current path, whereby the resistance is infinite and the current zero. The current integrated over time is obtained as the current I multiplied by the duty cycle.
The current path shown in Figure 2 comprises a resistor Rx in series with an analog switch 10, shown in the figure, which is controlled by a pulse train. The analog switch is in this embodiment arranged with a parallel resistor R, . During a first period of time, the switching device carries a current 1 , whereby the resistance in the current path is R . During a second period of time, the switching device breaks the current path, whereby the resistance is R^Rj and the current I2. The current integrated over time is obtained as the sum of Ix during the first period of time and I2 during the second period of time.
The connection shown in Figure 3 comprises a plurality of current paths of the type which is shown in Figures 1 and 2. The connection comprises a first analog switch 10a, by means of which the amplification constant K is adjusted, and a second switch 10b, by means of which the phase-angle error δ is adjusted. The values of the respective duly cycle obtained during the calibration are stored in non-volatile
memories in a calibration device contained in the electricity meter.
In meters of the intended type, errors often occur in the form of imperfections of the components included. Such components may, for example, be an operational amplifier which has a so-called voltage offset. In certain types of principles of measurement, such as, for example, the TDM method and when integrating a signal from a Hall element, an operational amplifier connected as an integrator is used. Figure 4 shows such a connection. The embodiment shows how an analog switch 10 is used to compensate for offset in an operational amplifier.
It is not uncommon that electronic equipment is sensitive to temperature variations in the surroundings . In the same way as shown in the embodiments, it is possible to arrange, within the scope of the invention, a current path which compensates the measurement process for such changes in temperature. To this end, an adaptive circuit is arranged in the calibration device, the duty cycle of the adaptive circuit being controlled on the basis of the signal from a temperature sensor. Also other adaptive systems may be arranged in the same way.
The experiment has shown that the pulse frequency may be chosen relatively arbitrarily. A pulse frequency of between 1 Hz and several tens of thousands of Hz has proved to fulfil reasonable requirements for accuracy of measurement.