How to actually verify a type-test certificate for power equipment
A certificate is a claim, not a fact. A field-by-field method for checking the laboratory, the accreditation, the scope line, the test values and the dates behind a type-test certificate before you wire money against it.
Key takeaway
A type-test certificate should be verified on five layers: does the laboratory exist and hold accredited scope for the exact test clause; does the certificate name the exact construction, rating and serial family you are buying; are the test values internally consistent (impulse level vs rated voltage, temperature-rise current vs rated current); do dates and document lifecycle line up; and does the story of who paid and who witnessed survive one phone call. Most mismatched certificates die on the scope line, not on forgery.
A type-test certificate is the document most procurement teams treat as a green light, and the document that most often fails to mean what the buyer assumes it means. The failure mode is rarely an outright forgery. It is a real document from a real laboratory that describes a construction, rating or product family subtly different from the one you are about to pay for. Here is the verification method we apply, in the order that kills false certificates fastest.
Layer 1 — The laboratory
Start with the institution, not the document. An accredited power laboratory is identifiable in the public databases of its accreditation body, and the accreditation covers specific test clauses, not everything the lab offers in its brochures. The check that matters: does the accreditation scope actually include the test you care about, at the voltage class you care about? A laboratory can be perfectly genuine and still unable to run your short-circuit withstand test in-house, in which case the certificate you are holding deserves one more question about who really performed it.
Layer 2 — The scope line
This is where most certificates die. Read the scope line the way a customs officer reads a passport: the exact product family, the exact rated voltage, the exact capacity or current rating, the exact construction series. Then hold it against your purchase order, line by line. A certificate that covers a design series but has never validated the specific rating required by your project is not covering you; it is covering this manufacturer’s earlier customer. In LV switchgear verification, the same discipline applies to the design configuration: a type-tested assembly built with a different internal arrangement, different current path, or different specific connections is no longer the assembly the certificate described.
Layer 3 — The internal arithmetic
Test values must retell a coherent story. The impulse level should agree with the rated voltage class; the temperature-rise test current should agree with the rated current declared on the nameplate family; the short-circuit withstand for a cable tray or enclosure line should agree with the conductor sizes actually offered. None of this requires expert judgement, only reading reading two numbers and asking whether one supports the other. When a certificate was run at a lower current or voltage class than your contract requires, the mismatch appears in one line of arithmetic, and it is the single finding a margin-skimming review would have missed and a revenue-protective review would not.
Layer 4 — Dates and lifecycle
A certificate older than a product generation is a weak base for a new award. Check whether the test dates precede the design revision history of what is being quoted: a major redesign without a corresponding renewal or extension test leaves the certificate valid in law and useless in substance. Also check the certificate against the picture of the tested object — rack arrangement, busbar configuration, enclosures. When the tested object and the shipped object cannot be shown to be the same construction, the certificate has ceased to be evidence.
Layer 5 — One phone call
The final layer costs five minutes: call the laboratory and ask them to confirm the certificate number against their records, including what was tested and who commissioned the test. This does not prove the shipped object is the tested object, but it answers a different fraud question: whether the certificate exists in the issuing body’s records at all, and whether the story of who paid for the test survives contact with the party that issued it.
What verification cannot close
Notice what this method does not establish: that the object placed on the test bed and the object on the ship are the same construction. Certificates are paper; the gap between paper and goods belongs to inspection. That is the logic behind factory acceptance testing as a delivery condition, and behind making certificate ranges, not just prices, a contractual matter.
In practice we run this five-layer check inside our compliance screening before a quotation ever reaches an award decision: laboratory identity, scope line, test values, dates, one confirmation call. Most questionable certificates never survive layer two. The ones that do are usually genuine — which is precisely why the remaining layers are worth the discipline.
Names and projects referenced in screening examples across this site are published in anonymised case files only. No certificate-holding supplier is identified in this article.