The question of sampling rates for Power Quality metering is an important one to consider.

Most power quality monitoring is reliably processed at sampling rates of 256 or 512 samples per cycle. However, fast transients can cause damage to sensitive equipment and malfunctions in manufacturing processes. Lower-end power quality meters cannot reliably detect these fast transients as they only sample dozens of times per cycle. A fast transient power problem can do damage in a few millionths of a second. In order to capture these transients, sampling rates in the order of millionth of a second would be required.

The European standard for Electrical Fast Transient (EFT) testing is EN-61000-4-4. The U.S. equivalent is IEEE C37.90. Both of these standards are similar.

Identifying very fast transients requires the use of specialized, portable monitoring equipment that costs up to $20,000—and frequently requires an outside consultant (plus travel & subsistence). Newer 'next-gen' power quality meters boast sampling rates up to 6MHz. But is this practical considering the fact that these fast transients can only be measured directly - typically on 400v networks?

With fixed installations, the bandwidth of Voltage Transformers (VT's) and Current Transformers (CT’s) is way too low to correctly pass through fast transients. Unless broadband VT's and CT's are used, what is the use of a 6 MHz sampling rate, if the VT & CT only accurately passes through 2kHz data?

The other challenge with such high resolution data in fixed installations, is where does one store it - pushing that much data over a digital network is very resource intensive and costly.

The practical application for such high measurement frequencies is therefore very limited.

However; the way this functionality has been implemented on the OTELLO platform, the user can clearly identify the root cause of fast transients. The OTELLO device captures enough data to identify whether a fast transient originated from a switching event or from lightning. One can even detect whether it was a direct lightning strike or an indirect one (induced). This solves the high sampling rate question, as the primary goal of monitoring fast transients is the identification of the source.

Compliance with IEC 61000-4-30 Class-A, dips, swells and interruptions must be measured on a full cycle and updated every half cycle, enabling the instrument to combine the high resolution of half-cycle sampled data points with the accuracy of full-cycle RMS calculations.

Aggregation windows are when a power quality instrument compresses measured data at specified periods. A Class-A instrument must provide data in the following aggregation windows:

1. The basic measurement time interval shall be a 10/12 cycle (~200 msec) at 50/60 Hz, note that the interval time varies with actual frequency.
2. 150/180 cycles (~3 sec) at 50/60 Hz, note that the interval time varies with actual frequency.
3. 10-min interval, synchronized with coordinated universal time (UTC) 2-hour interval for Plt flicker

OTELLO samples at 500 kHz on primary input signals. 50 kHz data is digitally retained and subjected to the IEC 61000-4-30 ED3.0 Class-A signal processing requirements.