Optimizing signal chain cost and accuracy for isolated current sensing in energy metering

1 Introduction

As we move toward a more sustainable energy grid with sources such as solar and wind power, the need arises for energy metering devices to obtain detailed insights into individual power consumption in order to identify areas for improvement, optimize usage, and reduce costs. The types of electronic end equipment that require an energy metering subsystem include smart electrical meters, electric vehicles (EVs) charging stations, power supplies and power distribution units, smart appliances, street lightning, and building automation components. The sheer number of these products drives the need for the lowest possible cost for energy metering solutions, while regional metering standards such as American National Standards Institute C12 (in the U.S.) or Measuring Instruments Directive (in Europe) impose strict accuracy and safety requirements [1], [2].

Figure 1 illustrates a typical signal chain inside an energy metering application, showing only one phase for simplicity. An analog-to-digital converter (ADC) simultaneously measures and digitizes voltage and current for each phase. Subsequently, digital signal processing extracts metrology parameters such as active and reactive power and energy, line-to-line voltages, fundamental power and energy, and harmonics [3].

The fundamental buildings blocks of the signal chain are:

Line voltage sensing front end (A in Figure 1).
Current measurement sensor (B).
Front-end and signal conditioner between the current sensor and the ADC (C).
ADC (D).
Digital signal processing hardware (E).
Galvanic isolation (F).

While the line voltage sensing front end is in most cases implemented using simple resistor-dividers [3], various options exist for the choice of every other building block. For each of these signal-chain components, performance, size and cost trade-offs exist. This article focuses on the performance vs. cost trade-offs for the current measurement sensor and signal conditioning, as well as the ADC.

Figure 1 Energy metering subsystem signal chain.

2 Comparing current-sensor elements used for energy metering

Table 1 summarizes the performance benefits and challenges vs. cost for three current-sensing technologies used in energy metering applications. Current transformers are the most popular sensors given their wide dynamic range, durability and low insertion impedance (noninvasive current measurement) [4]. Their cost is most likely the highest of all current-sensing technologies, however. Shunt resistors (shunts) are very attractive given their magnetic immunity, smaller size and reduced cost, but lack isolation and provide less accuracy at higher currents because of thermal self-heating [5].

Rogowski coils are an interesting alternative to the other two sensors and are the lowest-cost option, especially when considering printed circuit board (PCB) coils vs. bulk Rogowski coils [6].

Table 1 Current-sensor comparison.

Sensor Type	Current Transformer	Rogowski Coil	Shunt

Block diagram
Transfer function	$V_{o u t} = \frac{I_{l i n e} R_{b u r d e n}}{N_{t u r n s}}$	$V_{o u t} \propto \frac{d I_{l i n e}}{d t}$	$V_{o u t} = I_{l i n e} R_{s h u n t}$
Benefits	Provides isolation High linearity High durability High accuracy	Provides isolation Saturation not possible Small form factor (PCB) High linearity Fast response time Low power-loss	Not isolated Antimagnetic Small form factor
Challenges	Saturation possible Power loss in burden resistors Phase calibration required Size and weight	Integration required Cannot measure DC PCB coils can have lower sensitivity	Resistance may vary when self-heated Less reliability for overload signals
Cost	≥$0.21 (no shield) ≥$0.26 (with shield)	<$0.10 for some PCB coils ≥$0.10 for bulk coils	≥$0.10 (manganese)

Because of their low cost and flexibility of installation, PCB Rogowski coils are attractive for low-cost energy metering applications [7], [8]. Let’s analyze the benefits and challenges of a PCB Rogowski-based metering design and how to optimize the signal chain for the lowest possible cost while complying with regional metering standards.