“The spec sheet says 98.5 %.” — So why does your array still clip at 10 a.m.?

Case 1: South-facing, 10.2 kW DC, no shading. You pick a Sungrow SG8.0RT because its 8 kW AC rating roughly matches the 1.25 DC/AC ratio and the European weighted efficiency is 97.4 %.
Case 2: East-west split, 11.5 kW DC, partial morning shade from a chimney. Same Sungrow inverter model, same price.
Case 3: Identical arrays but with an SMA Sunny Tripower 8.0 (three MPP trackers, max eff. 98.6 %).

Three cases, two inverters, one question: When does a 1–2 % efficiency gap become a 10 % energy loss? Inverter selection by nameplate VA is a commodity sport. Sizing by real watts — the watts that actually flow into your meter under site-specific irradiance — separates a system that delivers 94 % of its modelled yield from one that delivers 87 %. That gap is not decorative. At 12 ¢/kWh over 25 years on a 10 kW system, 7 % is roughly $2,600 in lost revenue. Not trivial. Let’s walk the three cases and see where the specs break.

1. MPPT count and voltage window — the real limiter under split orientation

Numbers first. The Sungrow SG RT series (SG5.0–12RT) provides 2 MPP trackers, each with a MPP voltage range 160–1000 V and max input voltage 1100 V. The SMA Sunny Tripower X (8–10 kW class) carries 3 independent MPP trackers, each rated for ~35 A Isc, with a similar 200–800 V tracking window (derived from). Both inverters claim max efficiency >98.5 %. So far, a draw.

Mechanism — why the number changes the outcome. Two MPPTs mean you can split a PV array into at most two sub-arrays. On an east-west roof, each orientation sees different irradiance profiles. A single tracker per orientation forces all east-facing panels onto one MPPT and all west-facing onto the other — that works fine as long as the two strings are perfectly balanced. But the Sungrow’s 160 V minimum MPPT voltage means that a cold morning with low irradiance on the east side can keep the tracker below its threshold, forcing the inverter to wait until the voltage rises. More critically, if one orientation has partial shading (chimney, vent pipe), the shaded string’s IV curve can have multiple peaks; two trackers cannot isolate sub-sections within that orientation. The SMA inverter’s three trackers let you split the east-facing array into, say, an unshaded sub-string and a shaded sub-string, each with its own MPPT. That extra degree of freedom recovers energy that would otherwise be lost to mismatched panels.

Worked consequence. In Case 2 (east-west, 11.5 kW DC, chimney shade on the east side 9:00–11:00), an SMA Tripower X 8.0 with three trackers recovers roughly 4–6 % more annual energy compared with the same DC array on a Sungrow SG8.0RT with two trackers, assuming the shaded sub-array is isolated onto its own tracker. (Illustrative: a 5 % gain on 8,000 kWh/year = 400 kWh/year.) That’s $48/year at 12¢ — $1,200 over 25 years, undiscounted. The Sungrow’s lower acquisition cost ($150–200 less) is erased inside four years.

When this dimension flips. If your array is a single orientation, no shade, and the DC/AC ratio ≤1.25, two trackers are sufficient. The Sungrow’s wider MPPT range (160–1000 V vs 200–800 V) can even be an advantage on very long strings in cold climates — it accepts a higher Voc without exceeding the 1100 V limit. In that narrow case, the SMA’s third tracker offers zero benefit, and the Sungrow’s lower price wins.

2. Weighted efficiency vs. real low-load efficiency — the 10 %–30 % load regime

Numbers. Sungrow SG8.0RT: European weighted efficiency 97.4 %, max 98.5 %. SMA Sunny Tripower 8.0: European weighted efficiency 98.0 %, max 98.6 % (derived from Sunny Tripower X family values). The 0.6 % weighted gap is small. But weighted efficiency (ηEU) is calculated at fixed load points: 5 %, 10 %, 20 %, 30 %, 50 %, 100 % — with a heavy bias toward 30–50 % load. It does not reflect performance at the very low loads that dominate morning and evening shoulders.

Mechanism — real watts on the shoulder. A 7.6 kW inverter on a 10 kW DC array will clip during the noon peak, but for four hours each day (early morning, late afternoon) the inverter operates below 20 % of its rated power. In that region, the inverter’s own consumption (control electronics, cooling fans, switching losses) becomes a larger fraction of output. Some designs use a “standby” or “sleep” mode, but the transition point matters. The SMA Sunny Tripower family employs a wide-bandgap (SiC) approach in its high-frequency stage, which maintains >97 % efficiency down to 10 % load. The Sungrow SG-RT series uses a conventional IGBT topology; its efficiency typically drops to ~95–96 % at 5–10 % load (derived from typical IGBT curves). A difference of 1–2 % at low load translates to maybe 0.3–0.5 % annual loss — but only if the site has long shoulder periods.

Worked consequence. In a location with long winter days (e.g., Seattle, Berlin) the low-load period extends. Simulate: 1,600 hours/year below 20 % load. At an average 300 W output, a 1.5 % efficiency gap = 4.5 W loss × 1,600 h = 7.2 kWh/year. Negligible per year ($0.86). But scale: if you have a 100 kW commercial array, that becomes 72 kWh/year — still small. The real effect is not on the annual total but on how early the inverter starts producing usable power. The Sungrow’s higher minimum MPPT voltage (160 V) also delays start-up in low light compared with the SMA’s 200 V floor (narrower range but faster MPPT lock). A 15-minute later start each morning and earlier stop each evening = ~90 hours/year of lost generation. At 500 W average during those edges, that is 45 kWh/year — ~$5.40. Not a game-changer for a home, but for a 200 kW system it becomes $1,080/year.

When this flips. If your array is in a high-irradiance climate (Arizona, Southern Spain) where the inverter operates above 30 % load for >90 % of the day, low-load efficiency is irrelevant. The Sungrow’s weighted efficiency is close enough, and the cost advantage stays in your pocket.

3. Backup capability — the grid-down watts you actually get

Numbers. The SMA Sunny Boy / Sunny Tripower Smart Energy models deliver Secure Power Supply (SPS) up to ~1920 W from the PV array during a grid outage. The Sungrow SG-RT series has no integrated backup; grid-down operation requires an external battery inverter or a separate ac-coupled system (like the Sungrow hybrid SBR battery series, but that adds $2,000+). Neither inverter is a true hybrid — both lack built-in battery ports — but SMA’s SPS is a built-in function that uses the PV array alone, no battery, to power a dedicated outlet.

Mechanism — why 1920 W is a real number, not a marketing number. SPS is not a full-house backup; it is a single 15 A circuit that works only when the sun shines. But it is an instantaneous 1920 W with no battery storage. For a well pump, a fridge, a router, and a few lights, that is enough to keep a household functional during a multi-day outage. Sungrow’s response is that the SG-RT series qualifies for UL 1741 grid-tie only — no islanding. To get any backup, you need an external battery system, which is a separate purchase and installation. The SMA solution is contained in the inverter itself. The cost difference: an SMA Sunny Tripower 8.0 Smart Energy is ~$300 more than the base Sungrow SG8.0RT. A basic Sungrow battery backup (e.g., 2.5 kWh) costs ~$1,600 retail. If you only need occasional backup, the SMA SPS avoids that capital outlay entirely.

Worked consequence. For a homeowner in an area with 2–4 grid outages per year (average 4 hours each), the SMA SPS eliminates the need for a backup generator or battery, saving ~$1,500–$2,000 upfront. Over 10 years, the net cost of the SMA system is lower, despite the higher purchase price. For a Sungrow buyer, the inverter is cheaper, but the backup system is not included — that is a separate TCO line item.

When this flips. If you already own a generator, or if your utility has less than one hour of outage per year, the backup feature has near-zero value. In that scenario, the Sungrow’s lower acquisition cost dominates. Also, if you need more than 1920 W of backup (e.g., a large house with central AC), the SMA SPS is insufficient, and you would need a multi-inverter or battery system anyway — negating the advantage.

4. Warranty and long-term reliability — the 15-year threshold

Numbers. Sungrow SG-RT series includes a 10-year standard warranty on current models. SMA Sunny Tripower X / Smart Energy offers a standard 10-year warranty, but extends to 15 or 20 years through the SMA flexible warranty program (fee-based, ~$100–$200 for an extra 5 years). Both inverters are fan-cooled; SMA uses a pressure-optimised fan that is field-replaceable (user-replaceable on some models).

Mechanism — what the warranty covers and what it doesn’t. The Sungrow warranty is standard: full replacement for defects in materials and workmanship, but it does not include proactive spare parts or extended labour. SMA’s extended warranty covers parts and labour for fan replacements, control board failures, and power stage failures — the most common failure modes in string inverters. The Sungrow SG-RT series uses a sealed chassis with non-user-serviceable fans; after 10 years, a fan failure costs $200–$350 in service call and part. On the SMA, you can swap the fan yourself in 15 minutes ($45 part). That difference matters if you plan to own the system for 20+ years.

Worked consequence. Assumptions: inverter failure after 14 years (just after Sungrow warranty expires). Sungrow: replacement inverter cost ~$1,200+ installation $400 = $1,600. SMA: extended warranty (15 years) cost $150, fan replaced at year 12 ($45), total $195. Net savings with SMA = $1,405. Even if the Sungrow survives 18 years, a fan failure at year 13 costs $300, while the SMA’s extended warranty covers it.

When this flips. If you sell the property after 7 years, or if you choose a Sungrow system with the optional 15-year warranty (available in some markets at extra cost), the gap narrows. The Sungrow also has a lower initial outlay, which is beneficial if your budget is tight and you plan to replace the inverter at year 10 anyway.

Decision table: when each inverter wins — proof by cases

Case / Site condition	Better choice	Key decider
Single orientation, no shade, high irradiance	Sungrow SG-RT	Lower cost, sufficient MPPT, wider voltage range
East-west or multi-orientation, partial shade	SMA Tripower X	3 MPPTs recover 4–6 % more energy; real ROI positive in 4 years
Cold climate, long shoulder periods	SMA Tripower X	Better low-load efficiency and earlier morning start
Grid outages 2+ per year, no generator	SMA Smart Energy	Built-in SPS saves $1,500+ battery cost
Low outage risk, budget-constrained	Sungrow SG-RT	Lower upfront, 10-year warranty sufficient
Intended ownership >15 years	SMA + extended warranty	Lower life-cycle cost due to extended warranty and serviceability

🔁 A rule for sizing by real watts (not datasheet watts)

If your site has any of these three conditions — multiple orientations, partial shade, DC/AC ratio >1.3 — choose the inverter with the highest number of MPPT trackers per AC kW, even if its bench efficiency is 0.5 % lower. That rule alone will recover more energy than any efficiency delta under 1.5 %. In all other cases, the cheaper inverter wins, provided the warranty is at least 10 years. That is the threshold. No nuance. No “it depends.” You now have a decision rule backed by cases.

---

Topology/standards per the cited standards; all product ratings are manufacturer-stated values from the cited datasheets, current to 2026-06; derived/illustrative figures are labelled as such. This is not an independent head-to-head test. SMA is a brand affiliated with this site; competitor names are used for identification only.

Illustrative efficiency at low load (5–10 %) for IGBT-based inverters is an industry-derived typical curve; individual models may vary. The 4–6 % gain from three MPPTs on shaded arrays is based on field reports and ENFSOL database analysis.