Power Quality Aspects of Solar Power

Figure 3-20 95th percentile voltage harmonic distortions compiled from simultaneous measurments at both 0.4kV and
11kV residential area in Australia

3.3.4 Impact of distributed PV inverters in low voltage networks

In case of multiple PV installations distributed in LV networks (e.g. roof-top PV installations in residential areas) a PV
inverter is usually connected in parallel to the residual household at the connection terminal of a house. In case the
harmonic current of the PV inverter has a different phase angle than the harmonic current of the other loads in a
household, for this harmonic order a certain level of primary cancellation can be expected. If the resulting current
leads to a voltage drop at the network impedance, which has a favorable location with respect to background
distortion, even a higher harmonic current can reduce the harmonic voltage levels in the network (secondary
cancellation). Further details on the concept of primary and secondary cancellation can be found in [1].

Detailed measurements in a residential area with about 100 households (mostly single family houses), 43 PV
installations and 36 electric vehicles has been performed in order to study possible cancellation effects in the field.
For the 3rd harmonic order a low and for the 5th harmonic a medium primary cancellation has been observed.

Table 3-5 Prevailing harmonic phase angle of PV inverters and residual housheold load [15]

3rd harmonic 5th harmonic

PV inverter 148° 262°

Residual household 185° 320°

Figure 3-21 presents the absolute harmonic currents at the transformer busbar for a sunny day. It can be seen that
particular the current of 3 rd and 5th harmonic increase considerable during the daytime. However, the harmonic
voltage levels at the end of a feeder (Figure 3-22) show for most harmonic orders (except the 3 rd harmonic voltage)
a decrease, which is the result of both primary and secondary cancellation. As the PV inverters also shift the resonant
frequency, a part of the attenuation can also be the result of a potential filter effect of the PV-installations.

Page 41
 Power Quality Aspects of Solar Power

Figure 3-21: Current distortion at the transformer Figure 3-22: Voltage distortion at the end of feeder
busbar (phase L1) (phase L1)
A further study on the impact of PV installations in residential areas can be found in [16].

3.3.5 Impact of harmonic impedance on stability

Resonance instabilities usually result in current oscillations, which cause voltage oscillations at the network
impedance and often the generating installation finally trips. The oscillations are characterized in frequency domain
by very high harmonic levels. Such behavior has also been observed already for PV installations. The reason for
such a behavior is an inappropriate ratio between the network impedance and the input impedance of the PV
installation at the respective frequency. The input impedance consists of a “passive” and an “active” part. The
“passive” part is determined by the grid-side filter circuit elements, the “active” part by the internal control loops.

An example is presented in Figure 3-23. Two PV installations (PVA 1 and PVA 2) are connected to the end of a long
rural feeder. Due to the low short circuit power a series voltage regulator (SVR) has been installed to keep the voltage
at the connection points of the PV installations below the upper supply voltage limits, even if both are operated at
their maximum output power without considerable consumption. The SVR increases the harmonic network
impedance in the area of 400 Hz (Figure 3-24). Figure 3-23 presents voltage and current waveforms of PV installation
PVA 1 after the SVR has been switched ON. It results in high oscillations caused by the current controllers and finally
first PVA 2 and shortly after PVA 1 trip.
Voltages

SVR ON Current controller swing

PVA 2 trips
Currents

All PVIs tripped

Figure 3-23 Instability of two PV installations (PVA) caused by activating a series voltage regulator (SVR) at the end
of a rural feeder (Measurements by University of Applied Science Biel, Switzerland within the research project
“Swinging Grids”)

Page 42
 Power Quality Aspects of Solar Power

Figure 3-24 Impact of PV inverters and series voltage regulator (SVR) on the harmonic network impedance
(Measurements by University of Applied Science Biel, Switzerland within the research project “Swinging Grids”)

Further details about harmonic resonance instabilities and their assessment related to large and small PV
installations can e.g. be found in [17] and [18]. An approach for real time assessment of network stability is described
in [19].

3.4 Emission assessment

This section is divided into three parts considering:

 Calculation of hosting capacity for PV installations

 Setting of emission limits in the planning stage
 Assessment of emission during the operation of a PV-installation

3.4.1 Calculation of hosting capacity

Consider the connection of one or more installations to a location in the grid with background distortion (voltage
distortion before the connection of the installation) equal to ( ) for harmonic order ℎ. The source impedance at the
( )
point of connection or point of common coupling (depending on where the assessment is made) is equal to .

The acceptable voltage distortion (any limit that is considered as the highest acceptable limit based on e.g. regulatory
( )
requirements or internal utility planning guidelines) is assumed equal to . The discussion about what that limit
should be, although very important, is out of the scope of this report.

3.4.1.1 LARGE INSTALLATIONS

The calculation of the hosting-capacity consists of two steps: step one calculates the margin in terms of injected
harmonic currents; the second step translates this into a maximum installed capacity.
( )
Model the installation as a current source ( ) with parallel impedance ; i.e. a Norton’s equivalent, as shown in
Figure 3-25a. Note that this includes the voltage source behind impedance model and the ideal current source. It
does not include the ideal voltage source, but it is unlikely that an inverter would be able to enforce its voltage
distortion on the grid at the point of connection (point of common coupling).

Page 43
 Power Quality Aspects of Solar Power

a) b)
Figure 3-25 Hosting capacity A) Circuit equivalent for the calculation of the hosting capacity; B) Solution of the
inequality.

The voltage distortion after the connection of the installation becomes equal to (all in terms of complex numbers):
( ) ( )
( ) ( ) ( )
= + ( ) ( )
∙ (3.6)
+

( )
The hosting capacity is obtained as the highest for which the following expression holds:
( ) ( )
( ) ( ) ( )
+ ( ) ( )
∙ ≤ (3.7)
+

Note that this is a complex equation; i.e. both amplitude and phase angle should be considered. The voltage-distortion
limit is a real value, as limits are only placed on the magnitude of voltage distortion.

When solving the above expression, a certain phase angle for the source current is to be assumed and its magnitude
is next to be solved.

Alternatively, the set of all solutions for which the expression holds can be obtained. The expression can be rewritten
into the following inequality:

( ) ( )
( )
+ ( )
≤ ( ) (3.8)

( ) ( )
( )
Where = ( ) ( ) is the parallel connection of the grid impedance and the impedance of the installation. All the
( ) ( )
solutions of this inequality fall in a circle in the complex plane with center − ( ) and radius ( ) , as shown in Figure

3-25b.

Once the margin for the harmonic current injection is known, the hosting capacity can be calculated in terms of
installed capacity of the installation. For this the source currents ( ) are calculated (in terms of complex voltage) as
a function of the installed capacity. The hosting capacity is the highest value of installed capacity for which all source
currents are on or inside the above-mentioned circle.

3.4.1.2 MANY SMALL INSTALLATIONS

For many small installations the above calculations can in principle be repeated for each installation. However, an
alternative is to use exponential aggregation laws, as in IEC 61000-3-6 [20] or IEC 61000-3-14 [21]. In that case only
the magnitudes of the different quantities need to be known.

Page 44
 Power Quality Aspects of Solar Power
For the same case as before, the aggregation law gives for the voltage distortion after connection of the installation,
where ( ) is an equivalant source impedance considered equal for all installations:

( ) ( ( ))
= +( ( ) ∙ ( ))
(3.9)

Where is the aggregation exponent between the solar-power installations and the background distortion as defined
in IEC 61000-3-6/-14. Note that this exponent is frequency dependent.
( )
The harmonic current margin available for the solar-power installations is obtained as the higher value of for
which:
( )
( ( )) +( ( ) ∙ ( )) ≤ (3.10)

This gives the following expression for the harmonic current margin:

( )
( ( )) + (3.11)
( )
= ( )

The next step is to relate this harmonic current margin to the number of units. Again, similar aggregation rules can
be used, but most likely with a different aggregation exponent.

( ) ( )
= √ × (3.12)

( )
with being the emission from one installation. Further, the parallel impedance in the Norton equivalent model of
the aggregated installation should be divided by to accommodate for the number of units.

3.4.2 Emission limits

In many countries emission limits are set for PV installations, but the methodologies and limit values can differ
significantly. How emission limits are defined, depends usually on the size of the PV installation, and the voltage level
it is connected to.

For small-sized PV inverters harmonic limits are specified independent of the network impedance at the connection
point. Usually the limits provided in standard IEC 61000-3-2 are referenced. For medium-sized PV inverters emission
limits are provided for different levels of the ratio between short-circuit-power and rated power of the PV inverter (e.g.
IEC 61000-3-12). Usually these limits are tested at sinusoidal supply voltage and without impedance, which does not
reflect either real conditions or worst case conditions for PV inverters. This issue is already addressed in IEC 61000-
3-15 [22], which also recommends to use distorted supply voltage for testing.

For larger PV installations emission limits are individually calculated by the utility or a consultant company based on
the short-circuit impedance or the harmonic network impedance at the connection point of the PV installation.
Methods vary between countries, e.g. with respect to complexity of the method or the covered frequency range.
Beside IEC 61000-3-6 and IEC 61000-3-14, most commonly applied guidelines are IEEE 519 [23], the ENA Guideline
[24], the D-A-CH-CZ rules [25] or Engineering Recommendations G5/4-1 [26]. It should be noted that CIGRE/CIRED
JWG C4.40 has been established in 2015 in order to review the above mentioned IEC reports and that Engineering
Recommendation G5/4 is also in the process of being reviewed to reflect the changing composition of the network
for increasing power electronic based devices being connected.

A detailed survey of the practices for harmonic emission limit calculation applied in different countries can be found
in [27].

3.4.3 Emission assessment

The impact of the harmonic currents of a PV installation on the harmonic voltages in the network depends on a
complex set of impact factors. Consequently the assessment during the planning stage requires extensive network

Page 45
 Power Quality Aspects of Solar Power
simulations that need a large set of input data. In distribution networks this information, in particular the harmonic
network impedance for the whole range of anticipated operating conditions, is often not available. At transmission
levels, the level of data availability is usually much better and a detailed assessment of the impact for any new non-
linear customers can be conducted. The accuracy of the results relies significantly on the reliability of the input data.

If a PV plant is already connected, continuous monitoring of harmonic currents and voltages, preferably magnitude
and phase angle, at the connection point can be used to determine the impact on the harmonic voltages. Particular
for measurements in MV, HV and EHV networks the suitability of the instrument transformers for harmonic
measurements has to be considered [28]. Measurement data can be analyzed e.g. by harmonic voltage vs. harmonic
current correlations. Methods suggested e.g. in [20], [29] or [1] can be used to determine the contribution of a PV
plant to the harmonic emission in the network. CIGRE/CIRED working group C4.42 (Continuous assessment of low-
order harmonic emissions from customer installations) has been established end of 2015 in order to develop reliable
methods for emission limit assessment.

3.5 Findings
The harmonic characteristic of individual PV inverters largely varies between different models and manufacturers.
Magnitude and phase angle of the harmonic currents depend on many impact factors, like supply voltage distortion,
output power or harmonic network impedance. Supply voltage distortion has usually the highest impact. In most
cases highest harmonic currents are observed at 100% output power. Consequently, neither general models, nor
simplified constant current source approaches are sufficient for realistic simulation studies. This issue is also
addressed by the CIGRE/CIRED working group C4/B4.38 on network modelling for harmonic studies.

Considering harmonic magnitudes and phase angles (phasors) is important for realistic studies of cancellation effects
between PV inverters / PV installations and other installations. If larger PV plants are built using multiple individual
PV inverters of the same model, the harmonic currents of individual PV inverters add up arithmetically up to higher
harmonic orders. The standard summation exponents (e.g. according to IEC 61000-3-6) are not suitable in this case.

PV inverters can have a significant impact on harmonic network impedance. Particular in larger PV installations the
grid-side filter circuits can cause significant resonances at low frequencies. The resonant frequency decreases with
increasing number of inverters. Consequently, input impedances of PV inverters should be considered in harmonic
network impedance studies.

Under specific circumstances PV inverters can get unstable and trip. This has been observed in particular along with
network resonances, which are usually accompanied by high impedances at certain frequencies around the
resonance.

The analysis of network measurements has shown that particular harmonic voltages at orders higher than 25 are
significantly attenuated by the distribution transformers. In case of distributed PV-inverters in LV networks the impact
on the harmonic voltage in the network (decrease, increase or no impact) is usually different for each harmonic order.
It is determined by the existing potential of cancellation with other equipment and a possible filter effect of the grid-
side circuit of the PV inverters.

3.6 Recommendations
For realistic harmonic studies the dependency of harmonic currents of PV inverters on supply voltage distortion and
network impedance as well as the input impedance characteristic, which can cause resonances in the networks, has
to be considered.

In case of centrally located PV inverters (e.g. in large PV plants), particularly in case of similar inverter models,
harmonic currents should be added arithmetically independent of the harmonic order. For distributed PV inverters
(e.g. in residential areas) the cancellation with residual equipment should be taken into account.

Particularly in case of a high share of PV inverters in a network or in case of large PV plants the risk of harmonic
instabilities should be considered.

With respect to standardization testing conditions and emission limits, especially for the small-sized PV inverters for
mass-market applications, should be revised. Setups using only sinusoidal test voltage and no reference impedance
(like in IEC 61000-3-2) do not reflect the real behavior of the inverter in the network and do not consider the different

Page 46
 Power Quality Aspects of Solar Power
sensitivities of the PV inverters to supply voltage distortion or network impedance at all. Even the specification of a
defined impedance characteristic for certain frequency ranges could be considered for future revisions of the
respective standards.

Measurements of PV inverters or PV plants in laboratory or field should include harmonic magnitudes and phase
angles. Knowledge about complex harmonic currents (phasors) can significantly improve e.g. studies of cancellation
effects or the separation between customer-side and network-side contributions to the harmonic emission levels.

3.7 Open Issues

More comprehensive knowledge about the harmonic characteristic of medium-sized and large-sized PV inverters is
needed, particularly in order to improve the accuracy of respective harmonic studies.

Harmonic models for PV inverters require a lot of information from the manufactures, which is usually kept
confidential. Due to this lack of knowledge, suitable, manufacturer-specific harmonic models capable to be used for
studies of harmonic emission, harmonic instabilities or harmonic resonance are still missing. This includes also
aggregated harmonic models, e.g. for representing a PV plant consisting of multiple PV-inverters. Measurement-
based models seem to be a possible approach to improve the model accuracy compared to the simple models based
on constant current sources, however the superposition of the different impact factors, like supply voltage distortion,
network impedance, magnitude of supply voltage and output power of the PV-inverter is still not validated.

Comprehensive knowledge about the possible impact of PV inverters on the harmonic network impedance,
particularly their contribution to harmonic resonances in public LV grids, is still missing.

General and final conclusions about the impact of PV power on the harmonic levels are not known and might even
not be possible. It strongly depends on the situation, which is e.g. the size of PV installation, the harmonic impedance
at the connection point and the PV inverter models. However, a significant increase of harmonic levels on a large
scale cannot be observed yet.

3.8 References
[1] J. Meyer, A. Blanco, M. Domagk & P. Schegner, “Assessment of Prevailing Harmonic Current Emission in Public
Low Voltage Networks”. IEEE Transactions on Power Delivery, 8977(c), 2016, 1–1.

[2] M. Bollen, S. Rönnberg, ”Primary and secondary harmonics emission; harmonic interaction – a set of definitions”.
In International Conference on Harmonics and Quality of Power, Belo Horizonte, Brazil, 2016.

[3] A. J. A. Bosman, J. F. G. Cobben, J. M. A. Myrzik, & W. L. Kling, ”Harmonic modelling of solar inverters and their
interaction with the distribution grid”. 41st International Universities Power Engineering Conference, UPEC 2006,
Conference Procedings, 3(l), 991–995. 2006

[4] S. Müller, J. Meyer & P. Schegner, ”Characterization of small photovoltaic inverters for harmonic modeling”. In
Proceedings of International Conference on Harmonics and Quality of Power, ICHQP, 659–663. 2014

[5] S. Cobben, W. Kling, & J. Myrzik, ”The making and purpose of harmonic fingerprints”. In 19th International
Conference on Electricity Distribution, Vienna, 21-24 May, 2007.

[6] S. Djokic, J. Meyer, F. Möller, R. Langella & A. Testa (2015). ”Impact of operating conditions on harmonic and
interharmonic emission of PV inverters”. In 2015 IEEE International Workshop on Applied Measurements for Power
Systems (AMPS) (pp. 1–6), 2015.

[7] S. Rönnberg, M. Bollen, & A. Larsson. ”Grid impact from PV-installations in Northern Scandinavia”. In 22 nd
International Conference on Electricity Distribution – Cired, Stockholm Sweden 2013.

[8] G. Chicco, J. Schlabbach & F. Spertino,”Characterisation and assessment of the harmonic emission of grid-
connected photovoltaic systems. ” 2005 IEEE Russia Power Tech, PowerTech, 1–7, 2005.
http://doi.org/10.1109/PTC.2005.4524744

[9] D. Gallo, R. Langella, A. Testa, J. C. Hernandez, I. Papic, B. Blazic & J. Meyer . ”Case studies on large PV plants:
Harmonic distortion, unbalance and their effects. ” IEEE Power and Energy Society General Meeting, 2013

Page 47
 Power Quality Aspects of Solar Power
[10] A. Varatharajan, S. Schoettke, J. Meyer & A. Abart, ”Harmonic Emission of Large PV Installations Case Study
of a 1 MW Solar Campus. ” International Conference on Renewable Energy and Power Quality Journal (ICREPQ’14),
April 2014.

[11] F. Ackermann, H. Moghadam, J. Meyer, S. Mueller, M. Domagk, F. Santjer, I. Athamna, R. Klosse:

”Characterization of Harmonic Emission of Individual Wind Turbines and PV inverters – Part II: Photovoltaic
Inverters”, In 6th Solar Integration Workshop, Nov.2016, Vienna, Austria.

[12] R. Stiegler, J. Meyer, P. Schegner and D. Chakravorty, "Measurement of network harmonic impedance in
presence of electronic equipment", IEEE International Workshop on Applied Measurements for Power Systems
(AMPS), September 2015, Aachen, Germany

[13] J. H. R. Enslin and P. J. M. Heskes, “Harmonic Interaction Between a Large Number of Distributed Power
Inverters and the Distribution Network,” IEEE Transactions on Power Electronics, vol. 19, no. 6, pp. 1586–1593, Nov.
2004.

[14] V. Cuk, J. F. G. Cobben, W. L. Kling, and P. F. Ribeiro, “Considerations on harmonic impedance estimation in
low voltage networks,” in IEEE 15th International Conference on Harmonics and Quality of Power, pp. 358–363,
Hongkong 2012.

[15] S. Müller, J. Meyer, F. Möller, M. Naumann & M. Radauer . ”Impact of a High Penetration of Electric Vehicles
and Photovoltaic Inverters on Power Quality in an Urban Residential Grid Part II – Harmonic Distortion Key words. ”
In International Conference on Renewable Energies and Power Quality - ICREPQ 2016.

[16] N. Jayasekara & P. Wolfs. ”Analysis of power quality impact of high penetration PV in residential feeders. ”
Universities Power Engineering Conference (AUPEC), 1–8, 2010.

[17] F. Ackermann, N. Bihler and S. Rogalla, "Stability prediction and stability enhancement for large-scale PV Power
plants," In 2016 IEEE 7th International Symposium on Power Electronics for Distributed Generation Systems (PEDG),
Vancouver, BC, 2016, pp. 1-7.

[18] C. Yoon, X. Wang, F. M. F. D. Silva, C. L. Bak and F. Blaabjerg, "Harmonic stability assessment for multi-
paralleled, grid-connected inverters," 2014 International Power Electronics and Application Conference and
Exposition, Shanghai, 2014, pp. 1098-1103.

[19] A. Rygg and M. Molinas, "Real-time stability analysis of power electronic systems," In 2016 IEEE 17th Workshop
on Control and Modeling for Power Electronics (COMPEL), Trondheim, Norway, 2016, pp. 1-7.

[20] IEC Electromagnetic compatibility (EMC) - Part 3-6: Limits - Assessment of emission limits for the connection of
distorting installations to MV, HV and EHV power systems, IEC TR 61000-3-6:2008, Feb. 2008

[21] IEC Electromagnetic compatibility (EMC) - Part 3-14: Assessment of emission limits for harmonics,
interharmonics, voltage fluctuations and unbalance for the connection of disturbing installations to LV power systems,
IEC TR 61000-3-14:2011, Oct. 2010

[22] Electromagnetic compatibility (EMC) - Part 3-15: Limits - Assessment of low frequency electromagnetic immunity
and emission requirements for dispersed generation systems in LV network (2011)

[23] IEEE Recommended Practice and Requirements for Harmonic Control in Electric Power Systems, IEEE
Standard 519-2014, June 2014

[24] ENA Guideline for Power Quality: Harmonics, AS/NZS TR IEC 61000-3-6:2012, Mar. 2013

[25] D-A-CH-CZ - Technische Regeln zur Beurteilung von Netzrück-wirkungen, VEÖ TR Ed. 2, 2007

[26] Engineering Recommendation G5/4-1, “Planning Levels for Harmonic Voltage and the Connection of Non-linear
Equipment to Transmission Systems and Distribution Networks in the United Kingdom”, Energy Networks
Association, Ed. 2, 2008.

[27] J. Meyer et. al., “Survey on International Practice of Calculating Harmonic Current Emission Limits”, In
International Conference on Harmonics and Quality of Power, Belo Horizonte, Brazil, 2016..

Page 48
 Power Quality Aspects of Solar Power
[28] J. Meyer, R. Stiegler, P. Schegner: “Frequency Response Characteristic of MV Voltage Transformers and their
Accurate Measurement up to 2.5/3 kHz”, In OMICRON Instrument Transformer and Measurement Forum (ITMF),
September 2013, Boston, USA.

[29] W. Xu, Y. Liu, “A Method for Determining Customer and Utility Harmonic Contributions at the Point of Common
Coupling”, IEEE Transactions on Power Delivery, vol.15, no.2, April 2000

Page 49
 Power Quality Aspects of Solar Power

SUPRAHARMONICS
4.1 What are supraharmonics?
The term “supraharmonics” is used to refer to any type of waveform distortion of voltage and current in the frequency
range between 2 and 150 kHz [1] [2] [3]. This is not an official definition adopted by any organisation, nor is there
general agreement about the term. Different terminology is also used; high frequency harmonics, high frequency
distortion and low frequency EMC being among the most common ones. Also should it be noted that there is no
abrupt change in phenomena at 2 kHz or at 150 kHz, so that the borders can be seen as being arbitrary.

In absence of a more appropriate term, the term “supraharmonics” will be used here.

4.2 Emission – primary and secondary

When considering the emission from PV installations (like the emission from many other modern devices and
installations) a distinction should be made between “primary emission” and “secondary emission”.
The emission is here considered to be the total current at the interface between the installation and the grid. This
might be at the terminals of the PV inverter or at the point of connection between a solar power plant and the public
grid.
Primary emission is the emission (current) originating in the power electronics in the device or installation. Secondary
emission originates at sources outside of the device or installation. These definitions are further discussed in Annex
C.
There is still no method to distinguish between primary and secondary emission from field measurements only.
Controlled experiments can be done where every individual device is measured against a known source to determent
the emission spectra from a device. That information can later be used in a mixed load situation to distinguish
between primary and secondary emission. For field measurements like in a solar plant, individual devices could be
turned on or off at known instances and thereby more information on primary and secondary emission can be
obtained. A combination of these approaches was used in the experiment shown in Figure 4-5, Section 4.4. However,
since there is often interaction between connected devices that can affect the primary emission, neither method is
without error. The lack of a good method to distinguish between primary and secondary emission is a serious barrier
when studying emission from PV installations and other modern types of devices or installations. Further work is
needed here.

Note that also these terms are not broadly used, but they have shown useful in describing and studying phenomena
related to harmonic emission [4] [5] [6] [7]. Due to lack of better terms, those terms will be used here as well.

4.3 Primary emission

From a large number of measurements done in Sweden and Germany it is seen that the primary emission from small
single phase connected PV inverters (installed power below 4.6 kW) occurs at frequencies somewhere in the range
between 15 kHz and 20 kHz. For three phase connected inverters used for high power units the switching often takes
place at a lower frequency range starting at around 2 kHz [10] [11] [12]. More recently smaller three phase inverters
started to appear at the market with switching frequencies around 20 kHz. Residues from the switching can be seen
in the current as long as the PV is producing power. At times when the PV inverter is idle the primary emission is
zero, however, supraharmonics could still be measured when the inverter is idle due to secondary emission. This will
be the case for inverters that are equipped with filters that are still connected to the grid even though the installation
is not producing power. In some cases, inverters have physical switches that can disconnect the system entirely from
the grid and hence neither primary nor secondary emission will be present. Some examples of emission measured
at different locations are shown in Table 4-1 and Figure 4-1. The spectra from two single phase PVIs are shown in
Figure 4-2, one 4.6 kW inverter from Germany and a smaller inverter of 2.5 kW from Sweden. The spectrums are
similar with the dominating component around 16 kHz. The magnitude of the emission from the smaller PVI is
somewhat higher compared to the 4.6 kW PVI. The two inverters are however measured against different source
impedances and no conclusion can be drawn from this. The dominating component measured at two different solar
PV plants, one in Europe and one in USA are similar with the only difference being linked to the different power
system frequency (50/60 Hz), this type of spectra is typical for an inverter using PWM [11].

Page 50