+86-512-68303879

info@niuera.cn

image English
  • image English
  • image Pусский
  • image Français
  • image Español
  • image Portugues
high efficiency EV Charging Module
    Member Center
    Exit
    Data is empty
    0
    qr Code Url

    Scan qrcode to view mobile website

    Home /Company Blogs /Blogs List /Silicon Carbide in EV Charging Modules: What It Actually Changes — and What It Doesn't /

    Silicon Carbide in EV Charging Modules: What It Actually Changes — and What It Doesn't

    Silicon carbide is usually introduced with a list of promises: higher efficiency, smaller modules, less heat, better reliability. The physics behind the enthusiasm is real — SiC devices switch faster and conduct with lower losses than silicon IGBTs, and the research literature consistently finds lower losses when a silicon module is upgraded to a silicon-carbide one.12
    But "lower losses" is a device-level statement. What a charger OEM or a network operator buys is not a device. It is a module — a complete AC-DC converter with magnetics, capacitors, control electronics, a fan and an enclosure — and at that level the picture is more specific, and more interesting.
    Niuera's module range happens to make the comparison unusually clean. It contains a 40 kW AC-DC charging module built on a silicon platform (X-EH-40K-TX series) and a 40 kW SiC AC-DC charging module (XSC40KFG series) at the same rated power, the same output ranges, and the same physical envelope.3 Putting the two specification sheets side by side answers the question directly: at identical rated power, what does silicon carbide actually change?
    The answer is narrower than the marketing, and it points somewhere more useful.

    The like-for-like comparison

    Both modules are 40 kW AC-DC units with a 300–1000 V constant-power range and a 0–133.3 A output current range. Here is every published difference and every published similarity between them.3
    40 kW silicon (X-EH-40K-TX)
    40 kW SiC (XSC40KFG)
    Peak efficiency
    ≥96%
    ≥97%
    Current balancing accuracy
    ≤±5%
    ≤±3%
    Standby power
    <5 W
    <10 W
    Constant power range
    40 kW (300–1000 V)
    40 kW (300–1000 V)
    Output current range
    0–133.3 A
    0–133.3 A
    Current stability accuracy
    ≤±1%
    ≤±1%
    Voltage regulation accuracy
    ≤±0.5%
    ≤±0.5%
    Power factor
    ≥0.99
    ≥0.99
    Total harmonic distortion
    ≤5%
    ≤5%
    CAN modules in parallel
    max 60
    max 60
    Dimensions (W×D×H)
    300 × 437 × 84 mm
    300 × 437 × 84 mm
    Weight
    ≤17 kg
    ≤17 kg
    MTBF
    400,000 hrs (40 °C)
    400,000 hrs (40 °C)
    Ambient temperature
    −40 to 70 °C, derate above 55 °C
    −40 to 70 °C, derate above 55 °C
    Protection class
    IP20
    IP20
    Fan noise
    <65 dB at 1 m
    <65 dB at 1 m
     Three things are worth saying plainly about that table.
    The module did not get smaller. Same 300 × 437 × 84 mm envelope, same ≤17 kg. Whatever SiC did here, it was not spent on shrinking the box.
    The published reliability figure did not change. Both modules are specified at 400,000 hours MTBF at 40 °C. The common claim that SiC improves module reliability is not something these specification sheets support — at least not in the published MTBF figure.
    Standby power went up, not down. <5 W on the silicon module, <10 W on the SiC one. (This is addressed in a later generation — see below.)
    What did change is narrower: one percentage point of peak efficiency, and a tighter current-balancing specification. Both matter more than they look.
    Figure 2. What changed, what didn't: 40 kW silicon vs 40 kW SiC

    Why one percentage point is not a small number

    96% to 97% sounds like a rounding error. Expressed as efficiency it is. Expressed as loss it is not, and loss is what a thermal engineer and an electricity bill actually respond to.
    At 40 kW delivered output, the arithmetic runs as follows:
    40 kW at ≥96%
    40 kW at ≥97%
    Input power required
    ≈41.7 kW
    ≈41.2 kW
    Loss dissipated as heat
    ≈1.67 kW
    ≈1.24 kW
    That is roughly 430 W less heat per module, or about a 25% reduction in losses — from the same 40 kW of useful output.3
    Two honest qualifications on that arithmetic. First, these are peak-efficiency figures, not load-weighted averages across a duty cycle; a module spends real operating hours away from its best point, so the annual average benefit will be smaller than the peak-point calculation. Second, the figures are derived from the published efficiency specifications rather than measured on a bench, and both are stated as minimums (≥96%, ≥97%), so the true gap on any given pair of units may differ.
    Even with those caveats, the direction is what matters at cabinet scale. A 12-module cabinet at 480 kW carries roughly 5 kW less heat to remove if it is built from the SiC modules rather than the silicon ones — which is a fan, duct and cabinet-airflow problem before it is ever an electricity-bill problem.
    It is also worth being precise about what the wider literature does and does not support here. Reviews of SiC in electrified vehicles report efficiency gains in the several-percentage-point range, but those figures are typically drawn from traction inverters, a different application with a different baseline and duty cycle.14 The one-point step between these two 40 kW charging modules is the figure that applies to this product comparison; the inverter numbers should not be imported into it.
    Figure 3. One percentage point, expressed as loss

    The envelope stayed the same — and that is a design choice, not a missed opportunity

    The instinct is to read "same size" as "SiC wasted here." The thermal engineering argues the opposite.
    Industry analysis of 800 V and SiC-based designs makes a point that is easy to miss: SiC reduces total losses, but when the footprint shrinks at the same time, the remaining waste heat is concentrated into a smaller area. Local heat flux can therefore rise even as total losses fall, which makes thermal management harder rather than easier.5
    Read Niuera's two 40 kW modules against that. The enclosure, the airflow path and the fan specification are unchanged, while the heat to be removed drops by roughly a quarter. The same cooling surface is being asked to do less work. Holding the envelope constant is how that gain is banked as thermal margin instead of being spent on a smaller box that then runs hotter per square centimetre.
    One caution against overstating this: the published thermal envelope is identical — both modules are specified −40 to 70 °C with derating above 55 °C.3 The derating threshold of a complete module is set by more than semiconductor loss; capacitors, magnetics and control electronics all have a say. So the loss reduction does not show up as a higher temperature rating on the module. It shows up in the cabinet's cooling budget, which is where it is useful to whoever is designing the station.
    Thermal design of the module itself — airflow, duct geometry, fan control — is its own subject and was covered separately in our thermal-management piece; the point here is only that the SiC efficiency gain lands in the thermal budget rather than in the dimensions.

    Current balancing: the improvement nobody puts on a banner

    The specification that moved and got almost no marketing attention is current balancing accuracy: ≤±5% on the silicon module, ≤±3% on the 40 kW SiC module.3
    This matters as soon as modules are paralleled, which in DC fast charging is always. When several modules share a load, the balancing tolerance determines how unevenly they can be loaded while still reporting normal operation. Loose balancing has two costs: the system has to reserve headroom so the most heavily loaded module stays inside its limits, and the unit that consistently carries more current also runs warmer and ages faster than its neighbours. Tightening the band from ±5% to ±3% reduces both.
    But this is a 40 kW-specific gain, not a "SiC" gain — and the distinction matters if you are specifying hardware. The 60 kW SiC module is specified at ≤±5%, the same as the silicon 40 kW unit.3 Silicon carbide did not automatically tighten current sharing across the range; the 40 kW SiC design did.
    The same caution applies to paralleling limits. The 40 kW SiC module supports up to 60 modules on CAN, matching the silicon units — but the 60 kW SiC module supports up to 30.3 Neither figure follows from the semiconductor; both are per-model specifications, and a station design that assumes 60 across the board will be wrong for the 60 kW units.
    Figure 4. Per-model specifications that do not follow from the semiconductor

    What SiC did enable: a higher power class in a single air-cooled module

    The clearest contribution of silicon carbide in this range is not visible in the 40 kW comparison at all. It is visible in what exists above it.

    The silicon AC-DC line tops out at 40 kW. The SiC line continues to 60 kW — in a single, still air-cooled module.3 That is the capability step: not a smaller 40 kW unit, but a 60 kW unit that remains a rack-mounted, air-cooled, hot-swappable module rather than escalating to liquid cooling.

    The second-generation 60 kW module makes the packaging argument concrete. Comparing it with the 40 kW SiC module:3

    40 kW SiC

    60 kW SiC (G2)

    Efficiency specification

    ≥97%

    ≥97% — held at 1.5× the power

    Dimensions

    300 × 437 × 84 mm

    300 × 437 × 126 mm

    Rack footprint (W×D)

    300 × 437 mm

    300 × 437 mm — identical

    Height

    84 mm

    126 mm (1.5×)

    Weight

    ≤17 kg

    ≤29 kg

    Output current range

    0–133.3 A

    0–200 A

    CAN modules in parallel

    max 60

    max 30

    Standby power

    <10 W

    zero standby power

    Fan noise

    <65 dB at 1 m

    <60 dB

    Communication

    CAN

    isolated CAN interface

    The footprint is identical and the height is exactly 1.5×, for 1.5× the power — which is to say the G2 reaches 60 kW at the same volumetric power density as the 40 kW module (both work out near 3.6 kW per litre), while keeping the same rack width and depth.3 For anyone laying out a cabinet, that is a more actionable property than a percentage: the slot geometry is unchanged, the module is simply taller.
    Two things are worth drawing out of that table. The G2 holds the ≥97% efficiency specification at 1.5× the power of the 40 kW module — the efficiency gain is not something that had to be traded away to reach the higher power class.3 And it closes the standby-power gap noted earlier: it specifies zero standby power consumption, against <10 W on the 40 kW SiC and <5 W on the silicon modules. It also adds an isolated CAN interface and brings fan noise down to <60 dB.3
    A small precision on the efficiency figures, since this article leans on them: the catalogue labels the 40 kW figure peak efficiency and the 60 kW figures simply efficiency.3 The loss arithmetic earlier in this article is built on the 40 kW pair, where both numbers carry the same "peak" label and are therefore directly comparable. Treat the 60 kW figure as the specification it is labelled as, and confirm the measurement condition for any project where the distinction matters.

    The topology is a hybrid, not a swap

    One framing to retire: these modules do not "replace IGBTs with SiC."
    Niuera's SiC modules are built with 1200 V silicon-carbide MOSFETs and 1200 V silicon-carbide diodes together with high-efficiency IGBTs, combined with magnetic integration and multi-phase interleaved parallel topology, under a dual-core 32-bit DSP running at up to 100 MHz.3 Silicon and silicon carbide are used where each is better suited, in one converter.
    This is consistent with how the transition is described in the wider literature, where hybrid architectures appear alongside full-SiC ones rather than being superseded by them.4 For a buyer, the practical consequence is that "is it SiC?" is not a useful qualifying question. The useful questions are about the resulting specifications — efficiency, balancing, paralleling limits, derating behaviour — which is where the differences actually show up.

    What to ask when specifying modules

    Drawn from the comparisons above, these are the questions that separate a real specification from a datasheet impression.
    1. Is the efficiency figure peak or load-weighted? A peak number tells you the best point, not the duty cycle. Both figures here are peak.
    2. What is the current balancing accuracy for this exact model? It varies within a range built on the same semiconductor — ±3% and ±5% both appear.
    3. How many modules can be paralleled on this exact model? 60 and 30 both appear in the same SiC family.
    4. What does the derating curve look like, not just the ambient range? Every module here is −40 to 70 °C with derating above 55 °C; the shape of the curve above 55 °C is what determines summer output.
    5. What is standby power? It differs by generation, and in a large installation idle modules are a continuous cost.
    6. What is the rack footprint, and does more power mean a new slot geometry or just more height?
    7. Is MTBF quoted with a temperature? These are specified at 40 °C; a figure without a temperature is not comparable to one with it.
      Figure 5. Seven questions for specifying a charging module

    The takeaway

    Silicon carbide, in this product range, did not deliver a smaller module, a better MTBF figure, or lower standby power. It delivered one percentage point of peak efficiency — about a quarter less heat at full output — a tighter current-balancing band on the 40 kW unit, and a 60 kW power class that stays air-cooled and rack-mounted.
    That is a less sweeping claim than the category usually makes, and a more useful one, because each part of it can be checked against a specification sheet and turned into a cabinet design decision. The efficiency gain shows up in the cooling budget. The balancing gain shows up in how much headroom a paralleled system has to reserve. The 60 kW module shows up as a slot that is taller but no wider.
    The parts of the SiC story that did not survive the comparison are worth knowing too — mostly because they stop you from designing a station around a benefit that isn't in the specification.

    Getting the specifications

    Full specifications for the 40 kW and 60 kW SiC AC-DC charging modules, and for the silicon modules they sit alongside, are available on request — including derating curves, output characteristic curves and protection parameters for each regional variant (China, US and EU input ranges).

    Request the module specifications at info@niuera.cn or +86 0512-68303879, or visit www.niueraenergy.com.

    ──────────────────────────────────────────────

    FAQ

    How much more efficient is the SiC module than the silicon one?

    At the same 40 kW rating, peak efficiency is specified at ≥97% for the SiC module and ≥96% for the silicon module. Expressed as loss, that is roughly 1.24 kW against 1.67 kW of heat at full output — about 25% less loss. Both figures are peak-efficiency specifications, not load-weighted averages.3

    Is the SiC module smaller or lighter?

    No. At 40 kW both modules are 300 × 437 × 84 mm and ≤17 kg. The efficiency gain was banked as thermal margin rather than spent on reducing the envelope.3

    Does SiC improve the reliability figure?

    Not in the published specification. Both the silicon and SiC modules are specified at 400,000 hours MTBF at 40 °C.3

    Can I assume the same current balancing and paralleling limits across the SiC range?

    No, and this is the most common specification error. The 40 kW SiC module is ≤±3% balancing with up to 60 modules on CAN; the 60 kW SiC module is ≤±5% with up to 30. These are per-model figures.3

    What is the highest-power air-cooled module available?

    60 kW. The silicon AC-DC line stops at 40 kW; the SiC line extends to 60 kW while remaining air-cooled and rack-mounted. The second-generation 60 kW module keeps the 300 × 437 mm rack footprint of the 40 kW unit at 1.5× the height, and holds the same ≥97% efficiency specification at 1.5× the power.3

    Are these modules full SiC?

    No — they combine 1200 V silicon-carbide MOSFETs and silicon-carbide diodes with high-efficiency IGBTs, magnetic integration and multi-phase interleaved parallel topology, under dual-core DSP control.3

    What standby power should I budget for idle modules?

    It depends on the generation: <5 W on the silicon 40 kW modules, <10 W on the 40 kW SiC module, and zero standby power on the second-generation 60 kW module.3

    ──────────────────────────────────────────────

    References

    1. IET Power Electronics — A review of silicon carbide MOSFETs in electrified vehicles: application, challenges, and future development (2023).
    2. Energies — Comparative Dynamic Performance Evaluation of Si IGBTs and SiC MOSFETs (2025).
    3. Suzhou Niuera Energy Co., Ltd. — DC Charging Module Product Catalogue, V1.6 (2026). All module specifications cited in this article are taken from this document.
    4. PatSnap — SiC replacing IGBT in EV inverters patent-landscape analysis (2026), on hybrid and full-SiC architectures in traction applications.
    5. EV Engineering & Infrastructure — How 800 V architectures are challenging EV thermal management (2026), on rising local heat flux as footprints shrink.

    Product specifications cited above are drawn from the Niuera DC charging module catalogue current at the time of writing and are subject to change; efficiency, current balancing, paralleling limits and standby power are per-model figures and should be confirmed for the specific model and regional variant in any project. Loss figures expressed in kilowatts are arithmetic derived from the published peak-efficiency specifications and are illustrative of the full-output operating point rather than a load-weighted average. Third-party research findings are attributed to the sources listed and describe their own test conditions, which differ from this product comparison.

    Release time: 2026-11-17

    One DC Bus, Three Energy Sources: Inside an Integrated Solar-Storage-Charging System for E2W/E3W Fleets

    Related blog
    2026-09-18
    One DC Bus, Three Energy Sources: Inside an Integrated Solar-Storage-Charging System for E2W/E3W Fleets
    2026-09-09
    Talking to the Battery: BMS Communication, Connectors and Safe Fast Charging for Light EVs
    2026-08-19
    Solar, Storage, Charging: Why South America's E-Moto Infrastructure Should Be Built as One System
    2026-08-14
    Building a Reliable Charging Network: How Module Architecture Affects Station Uptime

    Follow Us

    Subscribe

    About Us

    About

    blog

    Contact Us

    Servise

    SKD/CKD

    Download


    Get In Touch Now!

    +86-512-68303879

    +86-512-68303879

    info@niuera.cn

    SiteMap

    The content of this site is copyright-protected and is the property
    Copyright © 2025 Suzhou Niuera Energy Co., Ltd


    (428229)
    0