Pith. sign in

REVIEW 4 major objections 6 minor 18 references

A Comparative Analysis of Transformer-less Inverter Topologies for Grid-Connected PV Systems: Minimizing Leakage Current and THD

T0 review · 4 major / 6 minor · reviewed 2026-08-10 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read This paper claims that the H5 transformer-less inverter topology achieves the best practical balance between low leakage current, low harmonic distortion, and efficiency for grid-connected photovoltaic systems.

desk verdict A clear tutorial-level simulation comparison of known transformerless PV inverters; the H5 recommendation is plausible but rests on a truncated THD metric, so the numbers should not be taken at face value. read the letter →

arxiv 2501.08103 v1 pith:AE4LCGU6 submitted 2025-01-14 eess.SY cs.SY

Shashwot Shrestha , Rachana Subedi , Swodesh Sharma , Sushil Phuyal , Indraman Tamrakar This is my paper · ORCID
classification eess.SYcs.SY
keywords transformer-lessinvertergrid-connectedPVleakagecurrentcommon-modevoltagetotalharmonicdistortionH5topologyHERICH4
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper argues that the H5 topology is the best practical choice among transformer-less inverters for grid-connected solar, because it keeps leakage current low while also keeping harmonic distortion moderate and efficiency high. Transformer-less inverters are attractive because they are smaller, cheaper, and more efficient than transformer-based designs, but they leak current through parasitic panel-to-ground capacitance when common-mode voltage fluctuates. The authors simulate H4 unipolar, H4 bipolar, H5, and HERIC topologies and report that H5 holds common-mode voltage nearly constant, yielding about 1.5 mA of leakage current, far below the 300 mA safety limit, while preserving a three-level output waveform and a total harmonic distortion of 10.29 percent. If the claim holds, H5 offers a practical way to reap the cost and efficiency benefits of transformer-less operation without violating safety limits.

What carries the argument

The mechanism that carries the argument is common-mode voltage clamping: keeping the average potential of the two inverter output terminals with respect to ground fixed at $V_{PV}/2$ in every switching mode. Leakage current is modeled through an LC resonant circuit in which $I_{CM} = V_{tCM} / ((X_{LA} \parallel X_{LB}) + X_{CPV})$, so a constant common-mode voltage makes the resonant denominator effectively infinite and drives leakage toward zero. H5 is designed so that a DC-bus switch and two clamping diodes force this common-mode voltage during both energy-delivery and freewheeling modes, while the differential-mode voltage, the output voltage across the terminals, still swings through three levels ($+V_{PV}$, $0$, $-V_{PV}$). This is the combination that lets H5 inherit the good harmonic behavior of unipolar modulation and the good leakage behavior of bipolar modulation at the cost of only one extra switch.

What would settle it

Take the same simulated grid-current waveforms for H4 unipolar, H4 bipolar, H5, and HERIC and recompute THD over the full harmonic spectrum rather than only the first three harmonics; if H5's full-band THD is no longer clearly below HERIC's, or approaches the bipolar H4 level, the paper's 'best balance' conclusion is an artifact of the truncated harmonic window.

Watch

Extended reading notes

Core claim

The central claim is that the H5 topology combines the two properties that matter most in a transformer-less PV inverter: the three-level differential-mode voltage of unipolar modulation, which keeps harmonic distortion and switch stress low, and the constant common-mode voltage of bipolar modulation, which suppresses leakage current. The extra switch on the positive DC bus decouples the PV side from the grid during freewheeling, and two clamping diodes hold the common-mode voltage at $V_{PV}/2$ in every switching mode. In the reported simulations, H5 produces a leakage current of 1.495 mA and a THD of 10.29 percent, whereas H4 unipolar reaches only 6.13 percent THD but leaks 296.7 mA, H4 bipolar nearly eliminates leakage but has the highest distortion, and HERIC has slightly higher leakage than H5 and 12.64 percent THD. The paper concludes that this combination makes H5 the most suitable topology for practical grid-connected PV applications.

Load-bearing premise

The comparison's distortion ranking rests on computing THD from only the first three harmonics of the 50 Hz fundamental, while standard THD includes harmonics up to the switching frequency and beyond, so a full-band measurement could change which topology appears to balance leakage and distortion best.

Editorial extensions

If this is right

  • H5 can meet the 300 mA leakage-current safety limit without a transformer, enabling smaller, cheaper, and more efficient grid-connected PV inverters.
  • H4 unipolar modulation is effectively ruled out for transformer-less grid connection despite its lowest THD, because its 296.7 mA leakage current is near the safety limit and would cause EMI and safety hazards.
  • H4 bipolar modulation is ruled out by its two-level output, which raises THD and switching stress even though it nearly eliminates leakage.
  • HERIC provides no clear benefit over H5: its extra high-frequency switch adds harmonic distortion while its leakage current is not lower.
  • The three-level output of H5 reduces filter requirements and switch voltage stress, supporting the paper's claim of better operational efficiency.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The comparison likely understates high-frequency distortion because THD is computed from only the first three harmonics of the 50 Hz fundamental; a full-band THD measurement could change the ranking of H5 versus HERIC and H4 unipolar.
  • The simulation assumes ideal switches and one fixed parasitic capacitance of 24 nF; real device parasitics could add common-mode spikes during switching, so the 1.5 mA figure should be read as a lower bound rather than a guaranteed field value.
  • The same clamping principle could be tested in an H6-style variant with a switch on the negative DC bus, which could be compared directly using the same simulation setup.
  • A hardware prototype measuring full-band THD and leakage current on a 2.2 kW single-phase grid would be the natural next check of whether H5's balance holds outside simulation.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 6 minor

Summary. This manuscript compares four single-phase transformer-less grid-connected inverter variants—H4 unipolar, H4 bipolar, H5, and HERIC—for PV applications, using analytical switching-state tables and MATLAB/Simulink simulations. The comparison dimensions are common-mode voltage (CMV), differential-mode voltage (DMV), leakage current, and total harmonic distortion (THD). The authors report that H4 unipolar has the lowest THD but high leakage current (about 0.27 A), H4 bipolar has near-zero leakage but high THD, while H5 and HERIC both maintain approximately constant CMV and three-level DMV. The paper concludes that H5 achieves the best balance, with leakage current around 1.5 mA, THD of 10.29%, and acceptable efficiency, and therefore recommends H5 for practical grid-connected PV systems.

Significance. If the reported results are reliable, the paper offers a useful engineering comparison for a practical design decision: it shows quantitatively that H5 keeps leakage current far below the VDE 0126-1-1 limit (300 mA) while retaining a three-level output voltage. The study's strengths include a consistent simulation setup across topologies, explicit switching-state tables for CMV and DMV, and direct comparison of leakage current and THD under identical circuit parameters. The principal limitation is that the THD metric is non-standard (only the first three harmonics), and the paper contains several numerical inconsistencies; these issues currently prevent the central 'H5 is the best compromise' claim from being fully supported.

major comments (4)
  1. [Section III-B, Table 6] The THD analysis uses only the first three harmonics of the 50 Hz fundamental ('only up to third harmonics are analyzed'), which is not a standard THD definition. For hard-switched inverters with hysteresis current control, a large fraction of harmonic energy appears at the switching frequency and its sidebands, which are completely excluded. Since the conclusion that H5 is preferable to HERIC depends on the THD ordering in Table 6 (10.29% vs. 12.64%), a full-band THD calculation (e.g., up to 2 kHz, 9 kHz, or Nyquist) is needed; otherwise the ranking may reverse and the central balance claim would not stand.
  2. [Section III-A1, Section III-B, Table 6] There are internal numerical inconsistencies in the reported results. Section III-A1 gives the H4 unipolar leakage current as 0.2697 A, while Table 6 lists 296.7 mA (0.2967 A); Section III-B gives the H4 bipolar THD as 13.38%, while Table 6 lists 11.83%. These discrepancies affect the quantitative comparisons and must be reconciled before the results can be considered reliable.
  3. [Abstract, Section IV] The abstract and conclusion state that H5 achieves 'optimal operational efficiency', but the paper reports no efficiency measurement, loss model, or switching-loss comparison. The simulation results cover only CMV, DMV, leakage current, and THD; efficiency is asserted rather than demonstrated. Either remove the efficiency claim or support it with measured or modeled loss data.
  4. [Section III-A4, Section IV] The explanation that HERIC has higher THD than H5 because it has 'one extra switch switching at high frequency' is not supported by the data. H4 unipolar has the lowest THD (6.13%) with only four switches, so switch count alone does not determine THD; the harmonic content depends on modulation strategy, output filter, and current-control dynamics. A spectral comparison of the actual output current would be needed to support this causal claim.
minor comments (6)
  1. [Section II] There is a typo at the start of Section II: 'sAmong' should be 'Among'.
  2. [Section II-A1] The equations for CMV and DMV are labeled (iii) and (iv) but referenced as (i) and (ii) in the text; the numbering should be corrected.
  3. [Section III-A] The figure captions are inconsistent: Fig. 13 is used for both H4 unipolar and H4 bipolar waveforms, and Fig. 9 is captioned 'Different Modes of Operation in HERIC topology' but shows the circuit diagram. The figure numbering and captions should be fixed.
  4. [References] References [12] and [13] appear to be the same source with different formatting; this duplication should be removed or consolidated.
  5. [Section III-B] The paper states that 40 cycles of grid current are examined, but it does not specify which cycles are used for the FFT or whether a steady-state window was selected; this should be clarified.
  6. [Table 6] The units in Table 6 are labeled 'LEAKAGE CURRENT (mA)', but the H4 unipolar value is given as 296.7; if the text value 0.2697 A is correct, the table should read 269.7 mA, and if the table is correct, the text should be updated. The inconsistency should be resolved.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the switching-state analysis and independent simulations support the comparison; the THD truncation is a correctness limitation, not a circular step.

full rationale

The paper's derivation chain is self-contained: the CMV and DMV values in Tables 1-4 follow directly from the pole voltages of each switching state, and the simulation waveforms are compared against those tables as an independent check. The leakage-current model (equations i-ii) is taken from external cited literature [10], not from the authors' own prior work, and the simulation uses fixed, stated parameters (Vpv = 400 V, Cpv = 24 nF, L1 = L2 = 4 mH) with no fitted parameter renamed as a prediction. The H5 leakage current (1.495 mA), HERIC leakage current (0.78 mA), and THD values are simulation outputs, not inputs constructed to force the conclusion. The preference for H5 over HERIC depends on a THD comparison that the paper explicitly limits to the first three harmonics, which is a measurement-validity and robustness concern that could change the ranking under full-band THD, but it is not circular reasoning: the truncated-THD result is not equivalent, by construction, to the claim that H5 is optimal. There are internal numerical inconsistencies (e.g., H4 bipolar THD stated as 13.38% in text but 11.83% in Table 6; H4 unipolar leakage stated as 0.2697 A in text but 296.7 mA in Table 6), but these are reporting errors, not evidence that a result was derived from its own input. No load-bearing self-citation, imported uniqueness theorem, or ansatz-smuggling-via-citation pattern is present. Accordingly, the circularity score is 0.

Assumptions & free parameters 1 free parameters · 4 assumptions · 0 invented entities

No fitted parameters are disclosed; the central design values (L=4 mH, Cpv=24 nF, Vpv=400 V) are stated as simulation inputs. The main hand-chosen methodological parameter is the THD harmonic limit. No new physical entities are introduced; the clamping diodes in H5 and HERIC are existing published circuit elements.

free parameters (1)
  • THD analysis harmonic limit = 3rd harmonic only
    Section III-B states that for THD analysis, 'only up to third harmonics are analyzed' at a 50 Hz fundamental, so the reported THD values exclude higher-order harmonics that dominate inverter distortion. This hand-chosen limit directly affects the THD comparison that supports the central claim.
assumptions (4)
  • domain assumption Leakage current in transformerless inverters is primarily caused by fluctuating common-mode voltage coupled through the PV parasitic capacitance.
    This is the standard model in the field and is invoked in the introduction and Section II to justify comparing CMV as a proxy for leakage current.
  • domain assumption Ideal switching device model is used for the mode-by-mode CMV/DMV analysis.
    The tables in Section II assume ideal switches and ideal clamping diodes; real device parasitics are not modeled, which could alter leakage current magnitudes.
  • domain assumption The MATLAB/Simulink simulation with HBCC control accurately represents the real-world behavior of the topologies.
    The simulation is the only evidence for the THD and leakage values; no hardware validation or model verification against measurements is provided.
  • ad hoc to paper THD computed from the first three harmonics is an adequate comparator of distortion across topologies.
    This is not the standard THD definition and is chosen by the authors; it omits switching harmonics that dominate the distortion of hard-switched inverters, potentially changing the ranking.

how reviews work

0 comments
Cite this review

Pith. "Pith review of A Comparative Analysis of Transformer-less Inverter Topologies for Grid-Connected PV Systems: Minimizing Leakage Current and THD." pith.science (2026). https://pith.science/paper/AE4LCGU6

@misc{pith2026250108103,
  author       = {Pith},
  title        = {Pith review of: A Comparative Analysis of Transformer-less Inverter Topologies for Grid-Connected PV Systems: Minimizing Leakage Current and THD},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/AE4LCGU6}},
  note         = {Machine review of arXiv:2501.08103}
}
read the original abstract

The integration of distributed energy resources (DERs), particularly photovoltaic (PV) systems, into power grids has gained major attention due to their environmental and economic benefits. Although traditional transformer-based grid-connected PV inverters provide galvanic isolation for leakage current, they suffer from major drawbacks of high cost, lower efficiency, and increased size. Transformer-less grid-connected PV inverters (TLGI) have emerged as a prominent alternative, as they achieve higher efficiency, compact design, and lower cost. However, due to a lack of galvanic isolation, TLGIs are highly affected by leakage current caused by the fluctuation of common-mode voltage (CMV). This paper investigates three topologies H4, H5, and HERIC with comparisons between their CMV, differential-mode voltage (DMV), total harmonic distortion (THD), and leakage current. A simulation was conducted for each topology in MATLAB/Simulink R2023a, and the results demonstrate that the H5 topology achieves a balance between low leakage current, reduced THD, and optimal operational efficiency, making it suitable for practical application.

Figures

Figures reproduced from arXiv: 2501.08103 by the authors.

Figure 13
Figure 13. (a) CMV, (b) leakage current, (c) DMV H4 unipolar (a) (b) (c) [PITH_FULL_IMAGE:figures/full_fig_p012_13.png] view at source ↗
Figure 14
Figure 14. (a) CMV, (b) DMV, (c) leakage current H4 bipolar [PITH_FULL_IMAGE:figures/full_fig_p012_14.png] view at source ↗
Figure 15
Figure 15. (a) CMV, (b) DMV, (c) leakage current H5 [PITH_FULL_IMAGE:figures/full_fig_p013_15.png] view at source ↗
Figures from the paper (1 more)
Figure 16
Figure 16. Figure 16: (a) CMV, (b) DMV, (c) leakage current [PITH_FULL_IMAGE:figures/full_fig_p014_16.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

18 extracted references · 15 canonical work pages

  1. [1]

    Gabriela Elizondo, Harnessing the Power of Distributed Energy Resources in Developing Countries: What Can Be Learned from the Experiences of Global Leaders? Oxford Institute for Energy Studies, 2023

    work page 2023

  2. [2]

    The role of distributed energy resources and associated business models in the Decentralised Energy Transition: A Review,

    K. Hargroves, B. James, J. Lane, and P. Newman, “The role of distributed energy resources and associated business models in the Decentralised Energy Transition: A Review,” Energies, vol. 16, no. 10, p. 4231, May 2023. doi:10.3390/en16104231

    work page doi:10.3390/en16104231 2023

  3. [3]

    Solar Photovoltaic Technologies,

    P. Breeze, “Solar Photovoltaic Technologies,” Solar Power Generation, pp. 51–55, 2016. doi:10.1016/b978-0-12- 804004-1.00008-7

    doi:10.1016/b978-0-12- 2016

  4. [4]

    Enhancing grid-connected photovoltaic system performance with novel hybrid MPPT technique in variable atmospheric conditions,

    L. Zaghba et al., “Enhancing grid-connected photovoltaic system performance with novel hybrid MPPT technique in variable atmospheric conditions,” Scientific Reports, vol. 14, no. 1, Apr. 2024, doi: 10.1038/s41598 -024- 59024-4

    doi:10.1038/s41598 2024

  5. [5]

    A new transformer-less single-phase photovoltaic inverter to improve the performance of grid - connected solar photovoltaic systems,

    M. Biswas et al., “A new transformer-less single-phase photovoltaic inverter to improve the performance of grid - connected solar photovoltaic systems,” Energies, vol. 15, no. 22, p. 8398, Nov. 2022. doi:10.3390/en15228398

    work page doi:10.3390/en15228398 2022

  6. [6]

    A comparative review on single phase transformerless inverter topologies for grid -connected photovoltaic systems,

    Md. F. Kibria, A. Elsanabary, K. S. Tey, M. Mubin, and S. Mekhilef, “A comparative review on single phase transformerless inverter topologies for grid -connected photovoltaic systems,” Energies, vol. 16, no. 3, p. 1363, Jan. 2023. doi:10.3390/en16031363

    work page doi:10.3390/en16031363 2023

  7. [7]

    Design of photovoltaic H6 -type transformerless inverter topology to minimize leakage current,

    M. A. Alam, S. V . Prasad, and M. Asim, “Design of photovoltaic H6 -type transformerless inverter topology to minimize leakage current,” Indian Journal of Science And Technology, vol. 16, no. 27, pp. 2068–2075, Jul. 2023. doi:10.17485/ijst/v16i27.1051

    work page doi:10.17485/ijst/v16i27.1051 2023

  8. [8]

    Common mode voltage in case of transformerless PV inverters connected to the grid,

    T. Kerekes, R. Teodorescu, and M. Liserre, “Common mode voltage in case of transformerless PV inverters connected to the grid,” 2008 IEEE International Symposium on Industrial Electronics, pp. 2390–2395, Jun. 2008. doi:10.1109/isie.2008.4677236

    work page arXiv 2008

Show all 18 references
  1. [9]

    Analysis and classification of non -isolated inverter leakage currents for photovoltaic systems,

    Z. Li, “Analysis and classification of non -isolated inverter leakage currents for photovoltaic systems,” MATEC Web of Conferences, vol. 386, p. 02002, 2023. doi:10.1051/matecconf/202338602002

    2023

  2. [10]

    Common mode leakage Current analysis of 1 φ Grid-Tied Transformer less H-Bridge PV Inverter,

    A. Srivastava and J. Seshadrinath , “Common mode leakage Current analysis of 1 φ Grid-Tied Transformer less H-Bridge PV Inverter,” 2021 International Conference on Sustainable Energy and Future Electric Transportation (SEFET), pp. 1–6, Jan. 2021, doi: 10.1109/sefet48154.2021.9375804

    2021

  3. [11]

    Critical review on various inverter topologies for PV system architectures,

    D. Kolantla, S. Mikkili, S. R. Pendem, and A. A. Desai, “Critical review on various inverter topologies for PV system architectures,” IET Renewable Power Generation , vol. 14, no. 17, pp. 3418 –3438, Dec. 2020. doi:10.1049/iet-rpg.2020.0317

    2020

  4. [12]

    Transformerless photovoltaic grid-connected inverters,

    H. Xiao and X. Wang, “Transformerless photovoltaic grid-connected inverters,” CPSS Power Electronics Series,

  5. [13]

    Xiao and X

    H. Xiao and X. Wang, Transformerless photovoltaic Grid-Connected inverters. 2020. doi: 10.1007/978 -981-15- 8525-8

    2020 doi

  6. [14]

    Review of common -mode voltage in transformerless inverter topologies for PV systems,

    T. Salmi, M. Bouzguenda, A. Gastli, and A. Masmoudi, “Review of common -mode voltage in transformerless inverter topologies for PV systems,” Smart Innovation, Systems and Technologies , pp. 589 –596, 2012. doi:10.1007/978-3-642-27509-8_49

    2012 doi

  7. [15]

    An H5 Transformerless Inverter for Grid Connected PV Systems with Improved Utilization Factor and a Simple Maximum Power Point Algorithm,

    H. Albalawi and S. Zaid, “An H5 Transformerless Inverter for Grid Connected PV Systems with Improved Utilization Factor and a Simple Maximum Power Point Algorithm,” Energies, vol. 11, no. 11, p. 2912, Oct. 2018, doi: 10.3390/en11112912

    2018 doi

  8. [16]

    Analysis and design of H5 topology in grid -connected single-phase transformerless photovoltaic inverter system,

    M. A. Alam, S. V . Prasad, and M. Asim, “Analysis and design of H5 topology in grid -connected single-phase transformerless photovoltaic inverter system,” Indian Journal Of Science And Technology , vol. 16, no. 6, pp. 420–426, Feb. 2023. doi:10.17485/ijst/v16i6.2191

    2023 doi

  9. [17]

    An improved H5 topology with low common -mode current for transformerless PV grid-connected inverter,

    H. Li et al., “An improved H5 topology with low common -mode current for transformerless PV grid-connected inverter,” IEEE Transactions on Power Electronics , vol. 34, no. 2, pp. 1254 –1265, Feb. 2019. doi:10.1109/tpel.2018.2833144

    2019

  10. [2021]

    doi:10.1007/978-981-15-8525-8

    doi

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.