Evidence sources · Reference

Three names on a spec sheet. Only one of them is about quality.

A quote lands on your desk carrying a BNEF Tier-1 badge, a reference to PVEL's scorecard and a line about Australian battery testing. They are three genuinely valuable pieces of work, and they answer three unrelated questions. Reading one as an answer to another is the most common evidence error in this market.

review.solar leans on all three. A Tier-1 listing is displayed on scorecards and never scored. Kiwa PVEL and RETC results are recognised for the verified reliability criterion. ITP Renewables sits alongside DNV and HTW Berlin as independent cycle and abuse testing. Readers meet those names without a way to tell what each one establishes, so this is the reference page for that, and it is linked from the methodology wherever one of them is cited.

Three questions, three answers

Each source is decisive for exactly one thing

1. 01
   BNEF Tier-1

   Will banks lend against this manufacturer's modules on a non-recourse basis? A financing and industry-acceptance signal. Not a product assessment, and BNEF says so directly.

2. 02
   Kiwa PVEL

   Does this specific module survive stress well beyond the IEC certification floor, and how does it behave across real irradiance and temperature? A product reliability and performance assessment.

3. 03
   ITP Renewables

   How did these battery packs actually behave over years of identical duty in one Australian climate? Long-run comparative field evidence, from a programme that has now concluded.

BNEF Tier-1 is a bankability list, and BNEF is unusually clear about that

BloombergNEF's Tier 1 list ranks module manufacturers by whether their own-brand product has been financed by commercial banks on a non-recourse basis for utility-scale projects. It is a measure of what the lending market has already accepted. The criteria are set out in BNEF's Tier 1 Solar Module Methodology, published openly and dated 18 March 2026, which includes its own section on methodology limitations.

The important part is what BNEF says about the list's own scope. Its published guidance describes the classification as a measure of industry acceptance rather than of product quality, notes that Tier-1 manufacturers have had documented quality problems and insolvencies, and directs purchasers and banks to technical due diligence instead. The public methodology document carries a dedicated section on the methodology's limitations.

It is worth sitting with how unusual that is. BNEF could let the misreading run, since a quality halo would only increase demand for the product. Instead the limitation is published rather than buried. The honest response is to use the list for the question it answers.

The organisation that publishes the list is the one telling the market not to read it as a quality ranking.

That instruction is a real one, and it cuts both ways for a household. If you are financing a 50 MW project, whether banks have lent against a manufacturer's product before is close to decisive. If you are buying fourteen panels for a roof in Perth, it tells you almost nothing you need, and its absence tells you less: plenty of capable manufacturers have never sought utility-scale project finance.

This is why review.solar shows a Tier-1 badge where one exists and never lets it move a score. Scoring it would import a financing signal into a product assessment and quietly convert BNEF's list into the quality mark it explicitly is not.

Kiwa PVEL tests the module, past the point certification stops

IEC certification is a floor. It establishes that a module does not fail outright under defined stress. It is not designed to separate a module that will hold its output for twenty-five years from one that merely passes.

Kiwa PVEL's PV Module Reliability Scorecard exists in that gap. It runs extended sequences well past the certification minimum, publishes results by manufacturer for modules submitted to the programme, and names Top Performers. Its PAN performance testing, run under IEC 61853-1, measures three identical modules across 23 operating points spanning 100 to 1,100 W/m² and 15°C to 75°C, then produces a PAN file usable directly in PVsyst energy modelling.

The value is that it is measured, per-model, and published. Our own Signals have leaned on it repeatedly for exactly that reason, including a 2025 scorecard result where one commercially marketed high-reliability POE encapsulant lost roughly 55% relative power after 1,000 hours of damp heat, against about 11% for EVA in the same configuration. No datasheet would have told you that. An independent laboratory publishing results manufacturers did not choose to publish is the only reason anyone knows.

The limitation is structural rather than a criticism: the scorecard covers modules that manufacturers submit. Absence from it is not a finding. It means a manufacturer did not enter that model, which may be commercial, may be timing, and is not evidence of anything about the product.

ITP Renewables ran the test nobody else was willing to run

Between 2016 and 2022, ITP Renewables operated a lithium-ion Battery Test Centre in a purpose-built climate-controlled enclosure at the Canberra Institute of Technology, supported by a A$1.29 million grant from the Australian Renewable Energy Agency. It cycled commercially available residential and small commercial battery packs continuously, under identical duty, and published what happened every six months across twelve public reports.

Three phases ran in sequence: the first from August 2016 with lithium-ion alongside conventional and advanced lead-acid; a second from July 2017 adding eight more lithium-ion packs, a zinc-bromide flow battery and an aqueous hybrid ion bank; a third from late 2019 adding eight further packs including sodium-nickel chloride and lithium titanate. Failures were reported as they happened, named, with the manufacturer's response.

That last point is what made it valuable and is also why it was rare. A programme that publishes which commercially sold batteries degraded, faulted or died is guaranteed to make enemies among the companies that sell them. It ran for six years anyway, publicly funded, and the results are still online.

One correction the market needs. The programme has finished. Report 12 was the final public report and testing of all packs completed at the end of March 2022. Australian solar marketing still cites ITP results as though the centre is running and the product on the quote is currently under test. It is not. The archive remains genuinely useful evidence about the packs that were in it, in that climate, over that period, and it should be cited as the historical record it is.

How to read all three together

A manufacturer can be Tier-1 and sell a module that has never been independently reliability-tested. A manufacturer can post excellent PVEL results and have no Tier-1 listing at all because it does not chase utility-scale project finance. A battery can perform well in the ITP archive and be three product generations out of date.

None of that is a defect in any of the three. It is what happens when three specific instruments get treated as one general verdict.

Our rule, and the one we would suggest to anyone reading a quote: ask which question the evidence in front of you answers, and whether it is the question you actually have. Financing acceptance, laboratory reliability and field endurance are all worth knowing. They are not interchangeable, and the source that says so most plainly is BNEF, about its own list.

Sourcesprimary where available

• BloombergNEF, Tier 1 Solar Module Methodology, published document dated 18 March 2026, which includes a section on methodology limitations. Reported and attributed, never reproduced: the document is copyright Bloomberg Finance L.P. and restricts reproduction and derivative use. Characterisations of BNEF's guidance on scope are drawn from BNEF's published statements about the list rather than quoted verbatim.
• Australian Renewable Energy Agency, ITP Battery Test Centre reports, public reports 1 to 12, 2016 to 2022.
• Kiwa PVEL PV Module Reliability Scorecard and PAN performance testing under IEC 61853-1, as cited in the individual Signals that use those results.
• review.solar scoring methodology, for how each source is and is not used.