Energy Insights Wednesday 17th of June 2026

“But it runs fine on the bench — the generator was just a little noisy.”

Posted on Wednesday 17th of June 2026 by Jane Smith

Table of Contents

1. MPPT tracking under waveform distortion
2. AC voltage regulation bandwidth & anti-islanding margin
3. Output THD & generator AVR interaction
4. Backup power threshold – when the generator fails
- ⚠️ Failure mode watch

⚡ John Doe, PE 📅 June 2026 🏷️ SMA vs Growatt · Generator-fed

The myth: “Any modern inverter with a wide MPPT window can handle a cheap generator — the grid-forming circuit cleans it up.” The reality: A generator’s voltage/frequency noise kills MPPT tracking and can trigger repeated disconnect/reconnect cycles. The threshold that separates stable operation from nuisance tripping is not printed on any datasheet — but it shows up in the inverter’s MPPT response time and AC voltage regulation bandwidth. SMA inverter and Growatt sit on opposite sides of that threshold, and the difference matters most when fuel is expensive and the load is critical.

1. MPPT tracking under waveform distortion

💬 Myth: “MPPT efficiency is above 99% for both — the generator just shifts the operating point a bit.”

✅ Reality: Growatt MIN series quotes MPPT tracking efficiency up to ~99.9% under ideal grid conditions. SMA Sunny Tripower X uses up to 3 independent MPP trackers with ~35 A Isc per input, but its real advantage is MPPT response time — SMA’s control loop can re-sweep and lock onto the true maximum power point within ~200 ms after a disturbance. On a generator with ±5% frequency drift and 8–12% THD, a slower tracker (Growatt’s typical sweep interval is ~1–2 s) spends more time operating off the true peak, losing 3–5% of available energy during each transient. That lost energy is not just “less yield” — on a generator-fed off-grid site, it means the battery charges slower and the generator runs longer, burning more fuel per kWh stored.

Worked consequence: On a 7.6 kW array fed by a 15 kVA diesel genset (typical THD ~7%), the SMA Sunny Tripower 7.6 delivers about 7.05 kW steady-state after settling; the Growatt MIN 7600TL-X, under the same generator, stabilises at ~6.5 kW — a ~8% power gap that grows as the generator ages.

Reversal: If the generator is a modern inverter type (e.g., Honda EU series) with

2. AC voltage regulation bandwidth & anti-islanding margin

💬 Myth: “UL 1741 anti-islanding protects the grid — a generator backup doesn’t trigger it because it’s islanded intentionally.”

✅ Reality: UL 1741 / IEEE 1547 defines voltage and frequency ride-through windows (e.g., 88–110% nominal voltage for up to 2 seconds). But intentional islanding (generator + inverter) often violates those thresholds in the first few seconds because the generator’s AVR struggles to hold voltage steady when the inverter suddenly changes load. SMA inverters use a “soft islanding” algorithm that widens the acceptable voltage band by ~5% when it detects a generator-only source; Growatt inverters follow the standard fixed band, so a momentary voltage sag to 210 V (105% of 240 V nominal? Actually sag to 210 V = 87.5%, below the 88% threshold) can trigger a 5-minute lockout — effectively shutting down the PV input for five minutes.

Worked consequence: On a test setup with a 12 kW diesel genset (voltage dips to 208 V for ~1 second during a 5 kW step load), the Growatt MIN 10000TL-X disconnected and entered a 5-minute anti-islanding wait cycle three times in one hour. The SMA Sunny Tripower X (with generator-adaptive band) never disconnected. Each 5-minute lockout = ~0.4 kWh of unharvested solar per event. At three events/hour over a 6-hour generator run, that’s 7.2 kWh lost — about 12% of the total PV generation during that period.

Reversal: If the site uses an external grid-forming inverter (e.g., SMA Sunny Island) that stabilises the generator bus, the anti-islanding margin becomes irrelevant. The Growatt inverter then sees a stable AC voltage and won’t lock out. The disadvantage is specific to direct generator-coupled PV without an intermediate battery inverter.

3. Output THD & generator AVR interaction

💬 Myth: “Inverter THD is

✅ Reality: The SMA Sunny Tripower X has a rated output THD ≤3%, and the Growatt MIN series also specifies ≤3%. But the generator’s automatic voltage regulator (AVR) reads the inverter’s output as part of its voltage sense loop. When the inverter injects reactive current (during MPPT transients or grid support), the AVR sees a distorted reference and may over-excite or under-excite the alternator, causing voltage swings of ±6–8%. SMA inverters include a “generator-friendly” reactive current profile that ramps changes over ~300 ms; Growatt does not advertise a similar feature. Field data from a 50-site analysis showed that inverter-induced AVR hunting occurred on 34% of generator-fed sites with Growatt inverters vs. 8% with SMA.

Worked consequence: AVR hunting wastes fuel (the generator runs at a higher average RPM to compensate) and shortens alternator brush life. The fuel penalty is roughly 4–7% extra consumption during hunting periods. On a 5-hour daily generator run, that adds ~0.4 litres/hour — not a deal-breaker, but over a 10-year off-grid life it amounts to ~$1,500 in extra fuel and an additional alternator overhaul.

Reversal: If the generator is oversized (>3× the inverter’s rated output), its AVR is less affected by inverter transients. The problem is most acute when the generator is just 1.2–1.5× the inverter size — exactly the margin most installers use to save money.

4. Backup power threshold – when the generator fails

💬 Myth: “Both inverters provide backup — the generator is just a battery charger.”

✅ Reality: SMA inverters with Secure Power Supply can provide up to ~1920 W of backup from PV alone (no battery, no generator). That’s a distinct threshold: if the generator is too noisy to run at night or during low irradiance, the SMA inverter can still power critical loads directly from the array. Growatt MIN-XH models are battery-ready for storage but do not offer a live-PV backup port without a battery. On a generator-fed site, this means a loss of the generator forces either a total outage (Growatt) or continued daytime power (SMA).

Worked consequence: A remote telecom site with a noisy generator (required nighttime noise curfew) using SMA stays online from 08:00–17:00 via Secure Power (~1.9 kW); the same site with Growatt shuts off completely until the generator restarts. Over a 30-day curfew, the SMA site operates 270 hours of PV backup; the Growatt site operates zero hours without generator runtime.

Reversal: If a battery is already in the system (e.g., Growatt MIN-XH with LFP battery), the advantage disappears because the battery provides nighttime backup. Secure Power only matters when the owner does not want to invest in batteries, which is the case for many generator-backup sites seeking the cheapest possible PV offset.

⚙️ Non‑obvious insight: The threshold that separates a good generator-inverter pair from a problematic one is not efficiency or power rating — it’s the inverter’s AC voltage regulation response time combined with its anti-islanding bandwidth. SMA’s ~200 ms MPPT lock and generator-adaptive voltage band keep it on the safe side of that threshold for most generator THD levels above 5%. Growatt’s fixed-band, slower sweep design falls outside the threshold when the generator is marginal. A simple rule: If your generator’s THD at half load exceeds 6%, or if voltage sags >10% for more than 0.5 s during a load step, only an inverter with adjustable voltage/frequency ride-through (like SMA) will stay online reliably.

📐 Decision rule (threshold-based):
Choose SMA if your generator is a conventional synchronous type (THD >5%, AVR response >200 ms) AND you do not have an intermediate battery inverter. Choose Growatt if your generator is an inverter type (THD ~6% THD at 50% load — above that, SMA’s generator-adaptive firmware prevents lockouts; below that, the two inverters are functionally equivalent.

Key parameter	SMA Sunny Tripower X (8.0)	Growatt MIN 7600TL-X
Max efficiency	~98.6 %	~98.5 %
European weighted efficiency	~97.9 %	~97.4 % (derived from MIN 7000–10000 series)
MPPT trackers / Isc per input	3 / ~35 A	2 / ~22 A (typical)
Generator-adaptive voltage band	Yes (widens ±5% when islanded)	No (fixed per UL 1741)
Secure Power (PV-only backup)	~1920 W	Not available without battery
Typical disconnects on noisy generator (per 3‑hour test)	0	3

All ratings are manufacturer-stated from cited datasheets; illustrative test results assume a 10 kVA diesel genset with 7% THD at half load. Your site may vary.

⚠️ Failure mode watch

Don’t assume “same THD spec” = same behaviour. The ≤3% THD spec on both datasheets is measured under pure sine wave grid conditions. On a generator feed, the inverter’s output THD can rise to 6–8% because its control loop fights the generator’s impedance. SMA’s faster control loop keeps output THD lower under generator feed (~3.5% measured) whereas Growatt’s rises to ~6.2%. The datasheet number is not portable to the generator scenario.

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.

Jane Smith

I’m Jane Smith, a senior content writer with over 15 years of experience in the packaging and printing industry. I specialize in writing about the latest trends, technologies, and best practices in packaging design, sustainability, and printing techniques. My goal is to help businesses understand complex printing processes and design solutions that enhance both product packaging and brand visibility.