What You’ll Learn ⚡

Grasp power quality fundamentals: voltage/frequency deviations, waveform distortion, and grid stability standards (IEC 61000‑4‑30 Class A)
Measure and interpret harmonics, interharmonics, total harmonic distortion (THD), total demand distortion (TDD), and grouped frequency bands up to 150 kHz
Analyze symmetrical components (positive, negative, and zero sequence) to assess system balance
Perform flicker and flicker emission analysis (Pst, Plt) per IEC 61000‑4‑15/‑21 and identify rapid voltage changes (RVC)

Set up DewesoftX’s Power Quality module: select channels, configure calculations, and use data displays like numeric, recorder, FFT/harmonic FFT, and waterfall plots
Capture raw data to study fault events and transient behavior while applying sensor compensation (amplitude/phase correction)
Export and visualize data in real-time or later—via built-in displays or external tools—ensuring compliance with power quality standards

Course overview

The Power Quality Analysis course provides a detailed examination of electrical power integrity and grid compliance using Dewesoft’s comprehensive hardware and DewesoftX software. It starts by defining power quality—covering waveform purity, voltage and frequency stability, and the impact of non‑ideal conditions on equipment and operations.

You’ll then explore key disturbances: from harmonics that distort ideal sinusoidal waves—up to the 3000th order and grouped per IEC 61000‑4‑7—to symmetrical component analysis for detecting unbalance, along with flicker and rapid voltage changes per IEC 61000‑4‑15 guidelines . The course also covers Total Demand Distortion (TDD), interharmonics, and frequency‑dependent groupings that allow deep examination of power quality dynamics up to high-frequency bands.

In hands-on modules, you’ll configure DewesoftX’s Power Quality module: selecting analog channels, choosing FFT vs. time-based calculations, and enabling inline sensor compensation to ensure high accuracy . You’ll engage in real-time monitoring using numeric displays, recorders, harmonic FFTs, waterfalls, and scope views—then learn how to record raw waveforms during fault conditions for detailed post-analysis.

Finally, the course teaches effective data management: exporting recorded data, applying calibration corrections, and generating visual reports that comply with industry standards. By completion, you’ll be able to deliver compliant power quality tests, troubleshoot disturbances, and integrate power quality modules into broader electrical diagnostics (including motor efficiency, power analysis, and PQ solutions across renewable, industrial, and grid applications).

Page 1 of 14

What are harmonics?

Harmonics are integer multiples of the fundamental frequency (e.g. 50 Hz for the grid in Europe) and cause a distortion in voltage and current of the original waveform. Harmonic voltages and currents caused by non-sinusoidal loads can affect the operation and lifetime of electrical equipment and devices. Harmonic frequencies in motors and generators can increase heating (iron & copper losses), can affect torque (pulsating or reduced torque), can create mechanical oscillations, and higher audible noise, it also causes aging of the shaft, insulation and mechanical parts and reduce the efficiency of the motor.

Harmonic summation of sinusoidal waveforms

The image depicts such a case. In the first recorder, the fundamental frequency with the 3rd, 5th and 7th harmonic orders overlapping are depicted. So, what is the problem, they are all sinusoidal waveforms with different amplitudes, right? Well no, look at the second recorder this is the fundamental frequency, the perfect sinusoidal waveform would be perfect as all electrical equipment prefers receiving signals in this form as it is the ideal waveform to work with.

The problem with the harmonic orders is that they sum together with the fundamental frequency, causing non-sinusoidal waves to form. The recorders following the fundamental frequency depict what happens to the sinusoidal waveform as more and more harmonic orders are added. It is clearly visible that the higher the order goes the less sinusoidal the wave becomes. Already at the 25th harmonic order, it is visible how the waveform is changing onto a square wave. The higher the harmonic orders that are added to the waveform the squarer it will get. If there were infinitely many harmonics added to the waveform it would become a perfect square wave.

Overview

Current harmonics in transformers increase copper and stray flux losses. Voltage harmonics increase iron losses. The losses are directly proportional to the frequency and, therefore, higher frequency harmonic components are more important than lower frequency components. Harmonics can also cause vibrations and higher noise. The effects on other electrical equipment and devices are very similar and are mainly: reduced efficiency and lifetime, increased heating, malfunction or even unpredictable behavior.

Dewesoft measures harmonics for voltage and current as well as active and reactive power up to the 3000th order. All calculations are implemented according to IEC 61000-4-7 and can be selected in the power module according to the following image. In order to calculate higher harmonics, the sampling rate has to be adjusted accordingly, for instance at a sampling rate of 500 kS/sec or higher DewesoftX can calculate up to the 3000th order.

Harmonic measurement options in the power module

Harmonics

Up to 500 harmonics can be calculated, in addition there is the option to choose all harmonics or just even or odd ones. If there are current channels used in the power module it is also possible to calculate phase angles, P, Q and the impedance.

Harmonic measurement calculation options in the power module

Number of sidebands

The basic idea of sidebands is that a certain frequency range is considered as one harmonic.

Example: 1 full sideband (equals +/-5Hz) at a frequency of 50 Hz means that a frequency range from 45-55 Hz is considered to be the first harmonic (it's the same for all other harmonics). If you select 2 sidebands the first harmonic will cover the frequency range 40 to 60 Hz.

Number of halfbands

The IEC 61000-4-7 (page 22) requires for the grouping of the harmonic sidebands where only the square root of the quadratic half should be added. This is required for the lowest and highest line and is defined as halfbands in Dewesoft.

Example I: 1 sideband and 1 halfband at a frequency of 50 Hz means that a frequency ranges from 45 to 55 Hz and the square root of the quadratic half of the 40 Hz to 60 Hz lines are considered to be the first harmonic.

Example II: 2 sidebands and 1 halfband at a frequency of 50 Hz mean that the lines from 40 Hz to 60 Hz have the full amplitude, while the lines at 35 Hz and 65 Hz are only considered with the square root of the quadratic half.

Interharmonics

Interharmonics cover all lines not covered by the harmonics

Please refer to page 26 of IEC 61000-4-7

Group FFT lines

The higher frequency parts can be grouped in 200 Hz and in 2kHz bands up to 150 kHz.

Depending on which grouping the measurement might require. Dewesoft offers the possibility to select one or both of these harmonic groupings.

Please be aware that according to page 29 of IEC 61000-4-7 these groups start at -95 Hz to +100 Hz around the middle frequency.

For 2.1 kHz, lines from 2005 Hz to 2200 Hz are considered to be one group.

Harmonic grouping according to the IEC 61000-4-7 standard

Full FFT

This option calculates a Full FFT which can then be exported to the database and displayed via a 2D-graph.

Harmonics smoothing filter

This option enables the low-pass filter which is required according to the IEC 61000-4-7 standard, page 23.

Background harmonics

With this option it is possible to subtract an existing and known harmonic pattern (magnitude and phase) from measured values. This is a typical application for the commissioning of a powerful power converter in order to ascertain the noise of the converter.

This function is available for both voltage and/or current, and it can be selected from the background harmonics editor in DewesoftX.

The only values that need to be entered for this calculation are the magnitude and the phase angle of the harmonic pattern as illustrated in the image of the input mask below.

Example

The following images depict a certain harmonic pattern that was measured using Dewesoft DAQ devices. A harmonic filter was then applied to the same measurement in Dewesoft X, and it yielded an adjusted harmonic by subtracting the background harmonics.

Measurement screen without a harmonic filter applied

Measurement screen with a harmonic filter applied

Page 1 of 14

How to perform a measurement with DewesoftX?

DewesoftX offers an array of display options for the following measurements: voltage, current, active and reactive power, phase angle, impedance, interharmonics, and higher frequencies, these can be displayed as Numeric Displays, on Recorders or as 2D-Graphs as the following image depicts. The user is absolutely free to configure the display as required.

Display options for the measurement screen in the power module

There are two possibilities for displaying harmonics in DewesoftX. The available choices are the Harmonic FFT and the 2D graph. The following image depicts the icons that are used for the two choices, left is the harmonic 2D Graph and right is the Harmonic FFT.

Display options for harmonics in the power module

Harmonic FFT

In the Harmonic FFT the harmonics of the voltage, current, power and reactive power can be displayed. The following image depicts the voltage harmonics of a three-phase system.

2D Graph

With the 2D graph, it is possible to display voltages and currents of different phases in one graph. In addition to this there is a wide array of display options that the user can configure as needed for the specific application. The following figure displays the harmonics for the phase voltage of L1. On the right-hand side are the display options that are available for 2D graphs. Here the graph type can be chosen, whether line or histogram as well as the graph scale which can be either linear or logarithmic. The scaling of the graph axes can also be set independently.

Persistence

There is also the option available to depict the Persistence of the harmonics in the 2D Graph. This means that when the harmonics change during a measurement, the changes are displayed blurred, as illustrated in the following figure.

Higher frequencies

Example: higher frequencies from 2 Hz up to 20 kHz shown in a 2D graph as histogram (application: HVDC converter station).

Interharmonics

Example: Interharmonics shown in 2D-Graph as Histogram. Peak at 900 Hz which is the switching frequency of a HVDC converter operated in the public grid.

Interharmonic measurement screen in a 2D graph and a FFT graph

Page 1 of 14

What are the symmetrical components?

Fundamental symmetrical components

Normally an electric power system operates in a balanced three-phase sinusoidal steady-state mode. Disturbances, for example a fault or short circuit, lead to an unbalanced condition. As the following image depicts on the left-hand side is a balanced system with a symmetrical phase shift and equal vector distances. On the right-hand side is an unbalanced system with an unsymmetrical phase shift and uneven vector lengths.

Balanced system and an unbalanced system

By using the method of the symmetrical components, it is possible to transform any unbalanced 3-phase system into 3 separated sets of balanced three-phase components, the positive, negative and zero sequence.

Unbalanced system to balanced system transformation

The advantage of the symmetrical balanced system lies in that the calculations are simplified. Should a fault arise or there is a short circuit in the system, an unbalanced system can be transformed into a balanced system with symmetrical components, where the system calculations can be done with the normal formulas that would be used in a balanced system. The calculated values are then transformed back to the unsymmetrical system (real-scenario) phase voltages and currents. In general, a 3-phase system can be depicted and mathematically described as follows:

A balanced 3-phase system looks like the image below with the same RMS-value for all line voltages and currents, and a 120° phase shift between each of them.

In order to explain the basic idea of the symmetrical components, the first step would be to define the operator as a unit vector with an phase angle of 120° or 2*pi/3.

a=e\frac{j2 \pi}{e^3}

The voltages can be described mathematically in different ways as the table below shows:

Symmetrical components complex and angle calculations

Calculation of zero-sequence system

In an symmetrical system the following equation is valid:

U_{L1}+U_{L2}+U_{L3}=0

In a real system the sum won't be zero. There will be a voltage difference as the following equation shows:

U_{L1}+U_{L2}+U_{L3}= \underline{\Delta} u\]

This voltage difference divided through 3 represents the so called zero-sequence system:

U_0 =\frac{1}{3} \cdot \underline{\Delta}u =u_{10}= u_{20}= u_{30} \]

The zero-sequence systems for the three phases (u10, u20, u30) all have the same amplitude and phase. Therefore, only the value for the zero-sequence system „U_0“ will be shown.

The calculation of the zero-sequence current is analog to that of the voltage equation.

When the currents of the zero-sequence system are multiplied by three (=I_0 * 3) it will yield he current that is flowing in the neutral line U_N

Calculation of positive-sequence system

The positive sequence system has the same rotating direction as the original system (right). This means it will have the same rotating direction of an electrical machine connected to the grid.

\underline{v_{1m}} = \frac{1}{3} (\underline{v_1} + \underline{av_2} + \underline{a^2v_3})\]

\underline{v_{2m}} = \underline{a^2v_{1m }} = \frac{1}{3} (\underline{v_1} + \underline{av_2} + \underline{a^2v_3}) \cdot \underline{a^2}\]

\underline{v_{3m}} = \underline{av_{1m }} = \frac{1}{3} (\underline{v_1} + \underline{av_2} + \underline{a^2v_3}) \cdot \underline{a}\]

As the values of the positive-sequence system for all three phases have the same amplitude (now that they are symmetrical) and has a phase shift of exactly 120°, it's adequate to show only one value. The value for the positive-sequence system in DewesoftX is called „U_1“.

Calculation of negative-sequence system

The negative sequence system has the opposite rotating direction as the original system (left). This means it will rotate in opposite direction of an electrical machine connected to the grid.

\underline{v_{1g}} = \frac{1}{3} (\underline{v_1} + \underline{a^2v_2} + \underline{av_3})\]

\underline{v_{2g}} = \underline{av_{1g }} = \frac{1}{3} (\underline{v_1} + \underline{v_2} + \underline{a^2v_3})\]

\underline{v_{3g}} = \underline{a^2v_{1g }} = \frac{1}{3} (\underline{a^2v_1} + \underline{av_2} + \underline{v_3})\]

As the values of the negative-sequence system for all three phases have the same amplitude (now they are symmetrical) and has a phase shift of exactly 120°, it's adequate to show one value. The value for the negative-sequence system in DewesoftX is called „U_2“.

Matrix of zero, positive and negative-sequence system

According to the following equations the phase voltages and currents are transformed into the symmetrical components. The result are three balanced 3-phase systems, the positive (U1, I1), negative (U2, I2) and zero sequence (U0, I0).

Matrices of the zero-,positive-, and negative sequence system

The basic values of symmetrical components (U_0, U_1 U_2, I_0, I_1, I_2) are calculated for each harmonic and addd up geometrically.

As illustrated in the following images, an unbalanced system can be rectified using the positive, negative and zero symmetrical components. The image below depicts an unsymmetrical system as a screen shot taken from DewesoftX.

Unbalanced system vectorscope in Dewesoft X This screen was provided by Kurt STRANNER (KNG Netz GmbH)

The following image depicts a screen-shot showing the three systems (positive, negative and zero) of the symmetrical components in DewesoftX:

nbalanced system to balanced system transformation in Dewesoft X This screen was provided by Kurt STRANNER (KNG Netz GmbH)

Out of the parameters of the symmetrical components (positive-, negative- and zero- sequence) the original system can be rebuilt easily, e.g.:

U_{L1}=U_0 + U_1 + U_2

The following variables are calculated in DewesoftX and show the components of the zero- and negative-sequence system compared to the positive-sequence system (for the total and the fundamental harmonic).

Zero-,positive-, and negative sequence system calculations in the power module

Extended positive sequence parameters (according to IEC 614000)

The following calculations are based on Annex C of IEC 61400-21.

Based on the measured phase voltages and currents, the fundamental's Fourier coefficients are calculated over one fundamental cycle T as first step.

u_a cos= \frac{2}{T}\int_{t-T} ^{t} u_a (t) cos (2 \pi f_1 t) dt

u_a sin= \frac{2}{T}\int_{t-T} ^{t} u_a (t) sin (2 \pi f_1 t) dt

It is important to mention that the index a stands for the line voltage L1. The coefficients for L2 (ub) and L 3 (uc) as well as the coefficients for the currents (ia, ib, ic) are calculated exactly the same. Furthermore f1 is the frequency of the fundamental. The RMS value of the fundamental line voltage is:

U_ {a1}= \sqrt{\frac{u^2_{a,cos}+u^2_{a,sin}}{2}}

Extended positive sequence calculations in the power module

Extended negative sequence parameters (according to IEC 614000)

Extended negative sequence calculations in the power module

Extended zero sequence parameters (according to IEC 614000)

Extended zero sequence calculations in the power module

Page 1 of 14

What is a flicker?

Flicker is a term used to describe fluctuations (repetitive variations) of voltage. Flashing light bulbs are indicators of high flicker exposure. Flicker is especially present in grids with a low short-circuit resistance and is caused by the frequent connection and disconnection (e.g. heat pumps, rolling mills, etc.) of loads which affect the voltage. A high level of flicker is perceived as psychologically irritating and can be harmful to humans.

The Dewesoft Power Analyzer can measure all the Flicker parameters according to the IEC 61000-4-15 standard. The Flicker emission calculation is implemented according to the IEC 61400-21 standard and allows for the evaluation of flicker emissions that are fed into the grid by wind power plants and other power generation units.

The flicker-meter architecture is depicted as a block diagram in the next image. It is divided into two parts, simulation of the response to the lamp-eye-brain chain and the on-line statistical analysis of the flicker signal leading to the known parameters. The blocks within the block diagram will be discussed briefly.

Block 1

The first block contains a voltage adapting circuit that scales the input mains frequency voltage to an internal reference level. This method permits flicker measurements to be made, independently of the actual input carrier voltage level and may be expressed as a percent ratio.

Block 2

The second block has the function of recovering the voltage fluctuation by squaring the input voltage scaled to the reference level, thus simulating the behavior of a lamp.

Block 3

The third block is composed of a cascade of two filters, which can precede or follow the selective filter circuit. The first low-pass filter eliminates the double mains frequency ripple components of the demodulated output.

The high pass filter can then be used to eliminate any DC voltage component. The second filter is a weighting filter block that simulates the frequency response of the human visual system to sinusoidal voltage fluctuations of a coiled filament gas-filled lamp (60 W/230 V and/or 60 W/120 V).

Block 4

The fourth block consists of a squaring multiplier and a first order low-pass filter. The human flicker perception, with an eye and brain combination, to voltage fluctuations applied to the reference lamp, is simulated by the combined non-linear response of the blocks 2,3 and 4.

Block 5

The last block of the chain performs an on-line analysis of the flicker level, thus allowing direct calculation of significant evaluation parameters.

The following image is an example of a rectangular voltage flicker.

Page 1 of 14

How the channel list is defined in DewesoftX?

DEWESoft has created a special FFT algorithm (software PLL) to determine the periodic time (frequency). The algorithm determines the periodic time of the signal via a special FFT algorithm at a sampling window of multiple periods (typically 10 periods, definable in power module). The calculated frequency is highly accurate (mHz) and works for all applications (motor, inverter, grid, etc.).

How Dewesoft X calculates the power of an AC system

While other power analyses calculate the power in the time domain, in DewesoftX calculates in the frequency domain. With the predetermined period time, an FFT analysis for voltage and current is done for a definable number of periods (typically 10, with electrical applications) and a definable sampling rate. The FFT yields an amplitude for the voltage, current, and the cos phi for each harmonic. One major benefit of this FFT transformation is that the behavior of amplifiers, current or voltage transducers in amplitude and phase for the whole frequency range (using the Sensor XML) can now be corrected. This way of power analysis has the highest possible accuracy. Another benefit is that harmonic analysis and other power quality analysis can be done completely synchronized to the fundamental frequency.

With the FFT corrected values, the RMS voltages and currents are calculated out of the RMS values of each harmonic.

U_{rms_{total}}=\sqrt{U^2_0 + U^2_1 +U^2_2 + ...+U^2_n}

I_{rms_{total}}=\sqrt{I^2_0 + I^2_1 +I^2_2 + ...+I^2_n}

The rectified mean is the average of the rectified signal. In terms of an AC signal, it's the average of the absolute value of voltage or current.

U_{RECT}= \frac{1}{T}\int_{t=0} ^{T} \left\vert u \right\vert (t)dt

I_{RECT}= \frac{1}{T}\int_{t=0} ^{T} \left\vert i \right\vert (t)dt

The rectified mean is used e.g. for transformer testing as the rectified mean is proportional to the magnetic flux.

The power values for each harmonic and the total values are calculated with the following formulas:

There are two definitions for reactive power included, because there is up to date no official definition that has been standardized. The formula of Q1_L1 loses sign, while QH_L1 does not. If there is a need to calculate the reactive power form only the harmonics (no fundamental included) the following simple math formula can be used: ‘QH_L1’ – ‘Q_L1_H1’