Overview of protection relays and the use of high accuracy AFEs for voltage and current measurement

Now that we are familiar with AC analog input module architecture, part four will focus on selection of key products such as an analog to digital converters, that ADC amplifiers and references used in the AC analog input module.

Voltage and current waveforms in real world are always not a simple sinusoidal. Waveforms could be anything from simple sine wave to complex periodic waveform consisting of multiple frequency component or a periodic waveform with noise or transient. The signal can be represented in either time domain or frequency domain.

Time domain analysis is a real world way of representing signals. Whereas in frequency domain, voltage and the current signals are represented by power spectrum showing of magnitude and phase as a function of frequency.

Depending on the type of application, algorithms can be implemented in time and frequency domains. For example, in case of analysis of a periodic signal such as impulsive transient signal, time domain analysis is preferred.

On contrary, if the signal is periodic with repetitive patterns and contains multiple frequency components, then it will be feasible to perform frequency domain analysis rather than time domain. It is important to note that choosing the right domain for analysis helps in obtaining reasonable answer, simpler and faster.

Fundamental to processing of signals in either time domain or frequency domain is sampling. RMS value of the voltage and current signal is computed from samples obtained by ADC or the predetermined time period of the signal.

If the sample is on the time record, do not start and stop feeding the same value in the time domain frame. The FFT interprets this as a discontinuity in the waveform.

If the sample time window is different from the signal time period, than FFT analysis of the signal produces frequency components which are not present in the original sinusoidal signal, as shown in the picture. This is termed as noncoherent sampling.

For a frequency domain analysis of the signal, it's very essential to perform coherent sampling, which describes a rational relationship between the input signal frequency and ADC sampling frequency. Coherent sampling describes the sampling of a periodic signal where an integer number of its cycles fit into a predefined sampling window. Since the end points of the data record match, the resulting FFT shows the proper frequency spectrum of the sinusoidal signal with minimum spectral leakage.

Another important requirement considered during selection of the ADC is capability to oversample. If the signal is sampled at much higher rate than the Nyquist rate of signal, then it's called as oversampling. By oversampling the signal, the average noise flow can be reduced as shown in figure one and two. Digital filter can be used to achieve further noise reduction. Oversampling improves the resolution, reduces noise, and helps in aliasing by relaxing performance of antialiasing filter.

This slide lists out the various parameters that play major role in the selection of ADC. In analog domain, performance parameters such as linearity, resolution, and the noise play a very important role. Selection of ADC also depends on the dynamics of the signal like frequency content and amplitude that's being measured.

In digital domain, parameters such as interface, multichannel sequencing such as MUX or simultaneous sampling, programmability, and calibration play key part in the selection. Some of the other considerations include power, support, pin, package, and size of the part.

The last and most critical consideration is the architecture of the ADC. This slide illustrates the difference between conversion of input signal while using SAR and delta sigma converter. SAR captures instantaneous or snapshot of the signal waveform at the time of sample and hold, as shown in the left side picture. Whereas delta sigma provides average value of the signal by continuously sampling the input waveform, as shown in the right side picture.

When it comes to sampling of multiple analog inputs, this can be achieved in two ways, either multiplexed or simultaneous. Most of the multichannel ADCs are multiplexed where a single A to D converter is used to convert signals from all the channels. This could be a cheaper solution. However, it does have a number of disadvantages.

Perhaps the most important of these is that this configuration does not sample all channels at the same instant of time. The time differential between samples is typically referred to as the samples skew. This, in many cases if not most applications, the time skew between samples on different channels is not problematic. , However, in some applications, these skews or phase shifts between signals cannot be tolerated and the standard multiplexed configuration cannot be used.

All the sample inputs can be captured at the same time by using sample and hold on each input. These sample and hold are triggered by a common input to achieve simultaneous capturing of the multiple channels. In this case, single ADC can be used to convert all the input channels.

Alternatively, it's possible by providing independent A to D converter for each channel, as shown, which results in two simultaneous sampling. Both the simultaneous sampling methods listed here should provide good results.

These are some of the key specifications of ADC. Sampling specs include timing details such as conversion time, acquisition time, throughput rate, et cetera. DC specs are generally derived from ADC tests obtained from DC or low frequency test signals. Similarly there are AC specs obtained from a sine wave test signals.

This table lists out some of the potential ADC solutions from Texas Instruments suitable for the AC analog input module for protection delay. Alias 131 series is a 24-bit delta sigma type ADC with simultaneous sampling. This has four or eight channels along with the programmable gains and the fault detection features. Alias 8588 is a 16-bit SAR ADC with simultaneous four, six, or eight channels, which can take bipolar input on a single power supply.

In this session, simultaneous amplitude ADC has been focused. This presentation focuses on TIDA-00834, which uses ADS8588S for sampling multiple analog inputs simultaneously.

The analog input signals are conditioned using amplifiers, which could be operational amplifier or instrumentation amplifier or fully differential amplifier. Op-amps are the simplest and most commonly used type of conditioning, which provides flexibility for configuring input and output signals. There are a wide variety of Op-amps available with dual and quad amplifiers, which optimizes the design and cost. Op-amps pumps have limited common mode rejection capability and inherent offset.

Instrumentation amplifier have very high common mode rejection ratio, which makes it affective to amplify low level signals obtained from the transducers. These amplifiers have a fixed, stable, and accurate gains, which can be adjusted by a single resistor. These are good against common mode noises with very high input impedance.

Fully differential amplifiers are used for conditioning inputs from sensors for ADCs with differential inputs.

In addition to amplifiers, external reference, host, interface, and isolation are also key in signal processing.

Here are some of the amplifiers and reference solutions from TI that could be used in AC analog input modules, such as operational amplifiers, instrumentation amplifiers, differential amplifiers, CDs, voltage references, isolation amplifiers, and digital isolator with integrated power converter.