Optimize Your Fault Location System: Traveling Waves Part 2 of 2

We’ve highlighted the growing need for traveling wave (TW) fault location systems in smart grids in the first part of this blog. We’ve also reviewed how to locate faults using the parameters of traveling waves—that is, the nature of fault signals as fast transient impulse signals that come and go within a few microseconds. This post will demonstrate how to capture and enable an optimized TW fault location system using a fast slew rate amplifier and high throughput rate analog-to-digital converter (ADC). 

Traveling waves propagate from faults in power transmission lines at close to the speed of light, so efficient TW fault location requires a selective and responsive system. Understanding the behavior of the data that needs to be captured—that is, a fast signal—will inform our choice of front-end sensor, filter, ADC, and amplifier. 

Figure 1: Traveling wave fault system signal chain
 

Front-End Sensor 

Two types of common front-end sensors are ideal for TW fault location.¹

Wide-band current transformers (CTs) translate high-current-induced signals to lower and safer levels. Accuracy at low currents is one of the advantages of CTs. But they are considerably large and heavy, so they are best for locating faults within a substation like DTU/FTU. 

Rogowski Coils (RCs) measure alternating current (AC) and high-speed current pulses. They do not saturate, so measuring a wide current range is not a problem. RCs are smaller, lighter, and more flexible than CTs,i making RCs optimal for locating faults in overhead transmission lines.

 
Figure 2: Graph of traveling wave and utility frequency readings at the sensor
 

Filter  

Traveling wave fault location is concerned with frequencies beyond our utility frequency of 50 or 60 Hz. To determine where a fault happened, our utility frequency must be eliminated. A bandpass filter is valuable for capturing a specific frequency range and can be made by cascading a high-pass filter to a low-pass filter. The high-pass filter eliminates frequencies in the 50-60 Hz range, while the low pass filter removes unnecessary frequency resonances. 

TWs mostly travel within the 2 kHz – 10 MHz frequency range. Valuable information may reach the fault location system at the substation up to 5 MHz, so we must capture data up to this frequency to estimate where the fault occurred.
 

Amplifier 

We can select an amplifier based on two specifications: Bandwidth and slew rate. The bandwidth must be wide enough and the slew rate must be responsive enough to optimize a fast and responsive signal chain.  

A wide -3 dB bandwidth has a wide frequency range that helps us ensure signal capture within the frequency range without degradation.   

Slew rate ensures that the output of the operational amplifier matches the input so that fast signals can be captured as-is. The maximum slew rate is to suit an optimal performance.  

To find the required slew rate, we will consider the input voltage to our ADC and a maximum input frequency of 10 MHz from the amplifier. As an example, let’s use the AD7380-4, which has an input voltage of 3.3 V. We will see why this is a good choice of ADC for this application in the next section.

Figure 3: Slew rate calculations using AD7380-4, with an input voltage of 3.3 V.

ADA4807-2 offers a wider bandwidth of 28 MHz bandwidth and a fast slew rate of 225 V/us, which goes well beyond the required 207 V/us—making it a good fit for TW fault location.
 

Analog-to-Digital Converter (ADC) 

We need a fast sampling and high-throughput ADC to capture and convert fast signals up to 5 MHz. A fast throughput rate ensures our ability to capture high-speed transient data from faults. We achieve this by sampling the TW frequency range within the Nyquist bandwidth maximum frequency, which is twice the highest frequency of the signal under test.

If FS is the sampling frequency and Finmax is the maximum input frequency, then: 

FS = 2Finmax
 

Figure 4: Signal interpreted by the ADC (AD7380-4) after the bandpass filter (ADA4807-2)

The AD7380-4 is a 4-channel, 4 MSPS, 16-bit simultaneous sampling converter and could capture up to 2 MHz frequency. If the requirement is to capture all data, the AD4080 has a much faster sampling rate of 40 MSPS that can capture this information, enabling more precise fault distance measurement.
 

Summary

Smart grid systems are seen as an efficient solution to support the world’s rapidly growing energy needs. In this two-part blog, we learned the advantages of traveling wave fault location and how it can elevate the performance of smart grids, unleashing efficiency in troubleshooting by ensuring rapid diagnosis of faults within the line.
 

Relevant ADI Solutions

• ADA4807-2 Datasheet and Product Info
• AD7380-4 Datasheet and Product Info
• AD4080 Datasheet and Product Info
• ADAQ4380-4 Datasheet and Product Info
• LTC6269 Datasheet and Product Info | Analog Devices
   

References

1. "Current Transformers vs. Rogowski Coils: A Comparative Study." Smilics Technologies, 2024.
2. “Use Of Travelling Waves Principle In Protection System And Related Automations.” ENTSOE, April 2021.
3. Marx, Stephen et al. “Traveling Wave Fault Location in Protective Relays: Design, Testing, and Results.” Bonneville Power Administration, University of Idaho, & Schweitzer Engineering Laboratories, Inc., 2013.