Technical note

What an 8,000-Unit Rejection Taught Us About Microinverter Sourcing

2026-08-28 · Renata Silva

At 8:30 on a Tuesday morning in March 2024, I stood in our Rotterdam warehouse looking at 40 pallets of microinverters. The cartons matched the purchase order, the barcodes scanned cleanly into inventory, but our incoming inspector had flagged something during the standard sample check. As the quality/compliance manager, I go look myself when something feels off.

He handed me one unit and pointed at the AC connector. When you plugged it into the speed-wire tester, the fit had more play than it should. Not enough to fail a spec-sheet test, but enough to notice—like a door latch that engages a little too loosely, you know?

“Pull two hundred more,” I said.

We did. About 1% showed visible inconsistencies in the sealing around the DC input caps, the moisture barrier that has to survive decades of thermal cycling. And the insertion-force measurements had a standard deviation more than double what our technical team was comfortable with.

Four weeks earlier, I had approved that supplier. Their quote was 8% below the brand we’d been stocking, and the spec sheets looked identical: same peak efficiency, same MPPT voltage range, same 12-year warranty. I’d been proud of the negotiation. Now I was staring at the result.

When “Within Industry Standard” Isn’t Enough

When I first started sourcing microinverters for our wholesale line, I assumed Tier-1 manufacturers were basically interchangeable. The datasheets all said the same things. The marketing all used the same adjectives. It took a bad batch and a very expensive lesson to teach me that reliability isn’t something you read off a PDF.

The supplier’s first response to our inspection report was an email:

“These units are within industry standard.”

I wrote back: “Which industry standard?” That question turned out to be more uncomfortable than I expected.

We ran a high-voltage isolation test following the procedures referenced in IEC 62109-1 (the international safety standard for PV power converters, covering everything from creepage distances to connector requirements). Out of 8,000 units, 64 failed. That’s 0.8%.

In small numbers, 0.8% sounds almost acceptable. Scale it up: on a 50,000-unit annual order, that’s 400 field failures. Each one means a service truck, a replacement unit, an irritated installer, and a warranty claim. The math doesn’t work, so we rejected the entire batch.

The direct cost was about $22,000—return freight, restocking fees, third-party testing, and the labor for a 100% inspection to make sure the rest of the lot was properly quarantined. Then we had to expedite replacement microinverters from another supplier at a premium. Total damage? North of $35,000, maybe closer to $40,000 if you count the internal meeting time.

But the root cause was worse. We opened several failed units and found the potting compound inside was unevenly applied—full coverage in some corners, almost nothing in others. That’s not a random component failure; that’s a process failure on the production line. When we asked for batch-level production records, the manufacturer sent summary sheets. No lot numbers, no traceable test data, no corrective action history. For someone selling units with a 20-year design life, that documentation gap was the real red flag.

What SMA Solar’s Annual Report Taught Me

About a month later, I was in the middle of qualifying SMA Solar as a core supplier for our hybrid and string inverter SKUs. This time I started with the things I should have always started with: financial substance and quality traceability.

That meant reading the audited fiscal 2023 annual report of SMA Solar Technology. SMA Solar Technology’s 2023 revenue was around €1.19 billion, and SMA’s inverter shipments in 2023 reached about 17.2 GW—up from 11.4 GW the year before. Those aren’t just vanity metrics. For a B2B buyer, they signal bankability: the capacity to honor warranty claims, stock spares, and support software for 15 to 20 years. A manufacturer shipping 17.2 GW in a single year has a very different risk profile than one shipping 500 MW.

I also looked at the certification footprint. IEC 62109-1 was there, along with IEC 61683 for efficiency measurement and a long list of grid-code certifications across Europe, North America, Australia, and Asia. That paperwork accumulates when a manufacturer has run certified production lines for years and kept the audit trail.

For our business, portfolio breadth mattered too. We wanted a partner that could anchor our hybrid and string inverter lines and set a documentation standard we could benchmark the rest of our sourcing against. SMA’s product portfolio covers those segments, which meant one qualification process instead of two.

Building a New Vendor Qualification Scorecard

It took me about four years and roughly 30 supplier evaluations to understand this properly, but I’ve come to believe that in B2B solar sourcing, price is the last thing you check, not the first.

Today our procurement protocol has five gates:

  1. Financial substance. Audited annual report, revenue trajectory, shipment volumes, warranty provisions. If the company won’t survive, the warranty is fiction.
  2. Quality system. ISO 9001 certification, IEC/EN test reports, and most importantly batch-level traceability. Ask for test data from production lots, not just a certificate.
  3. Reliability evidence. Third-party test results (for example, the PVEL/DNV Product Qualification Program for modules) and any published field failure data.
  4. Portfolio depth. Can one supplier cover multiple inverter segments and modules? If yes, your logistics and training budgets shrink dramatically.
  5. Claims process. What happens when a lot does fail? Who pays for replacement, return freight, and site labor? Get it in writing.

Choosing Photovoltaic Modules for Wholesale

Later that year, when we started our photovoltaic module wholesale line, the same questions applied. It’s tempting to source modules purely by cost per watt, because that’s the number that hits the price sheet. But the questions that actually decide the vendor are different: has this module passed damp heat, thermal cycling, PID, and LID testing (the stress tests that expose real degradation modes)? What’s the documented defect rate over the last 100,000 units, not just the pilot sample? Does the factory’s batch data match the serial numbers on our shipment?

The same logic applies if you’ve been evaluating solar inverter OEM or private-label options. Unless you have your own engineering and claims infrastructure, you’re taking on the manufacturer’s production risk and paying for the branding. We made a deliberate choice to stock established brands and invest in service instead. That’s not the right answer for every company, but it’s worked for us.

What Changed After We Switched

Since we standardized on SMA for our hybrid and string SKUs—and applied the same five-gate framework to our microinverter sourcing—quality-related complaints from installer customers have dropped noticeably. I don’t have a fancy statistic to hand you; the sample is about a year and I’m not 100% sure our tracking system caught every nuance. But the shift was visible in our Q4 review.

It wasn’t only the failure rate. Our installers stopped asking why a unit looked different from the previous delivery. Our warehouse stopped dealing with ambiguous RMA claims. And our procurement team stopped burning hours comparing quotes, because once you’ve had a supplier save you 8% and cost you $40,000, you start seeing the hidden costs everywhere. (Note to self: our margin improved most when I stopped making decisions on speculation.)

Looking back, I should have built this qualification process years earlier. At the time, I was negotiating like a good purchasing manager but managing risk like a bad one. If I could redo that decision, I’d run the quality system audit before signing the PO—not after the inspection failure.

Here’s the thing about B2B solar: when an inverter or module fails in the field, the installer doesn’t call the manufacturer. They call you—the distributor, the brand on the invoice. Whatever that product’s quality is becomes your company’s image. The sticker on the side might say SMA, but the reputation is yours.

So if you’re sourcing equipment for wholesale, treat the supplier’s quality system as part of the product. Check the annual report. Ask for batch-level data. Review the third-party stress test results. And keep asking “which standard?” until you get a specific answer. It cost me one rejected lot of 8,000 microinverters to learn that. You can probably learn it for less.

Previous: SMA Solar Inverter Shipments 2023: What the Numbers Actually Tell You (and What They Don't)Next: SMA Solar, PV Inverters, and PV Module Distributors: A Quality Inspector's Buying FAQ