XIANYUAlloy · Changzhou
Technical note · 2026-11-04

A material certificate is not a thermoelectric test, and buyers keep treating it as one

A thermocouple wire order usually arrives with a mill test certificate, and purchasing signs it off as evidence that the wire is good. The certificate is real evidence, but it answers a narrower question than the one being asked. It says the alloy is what it claims to be. It does not say the pair generates the voltage the reference table expects.

Two documents, two different questions

A mill test certificate is an inspection document. Under the EN 10204 scheme it comes in grades, and the grade matters: a 2.2 declaration states that the product complies with the order on the basis of non-specific inspection, while a 3.1 inspection certificate reports specific test results for the delivered batch and is signed by an inspection representative independent of the manufacturing department. Same-looking paperwork, very different evidential weight.

A thermoelectric test is a measurement of output. The sample pair is held at a known temperature difference and its EMF is compared with a reference, either against a standard thermocouple by comparison techniques or against the published reference table. That measurement is what tells you whether the pair behaves like Type K, Type J or whatever the label says.

Neither document contains the other. A chemical analysis does not measure a Seebeck coefficient, and an EMF measurement does not tell you which elements are present.

Why good chemistry can still give a bad pair

This is the part that surprises buyers. Wire can meet its composition specification on both legs, with a 3.1 certificate to prove it, and still fail an EMF check. Four mechanisms account for most of it.

  • Mismatch. The two legs are tested separately against composition. If they come from different melts or different processing runs, each leg can be within its own chemical tolerance while the pair departs from the reference curve.
  • Segregation. Composition is measured on a sample. A billet that is not homogeneous can pass on the sample and vary along the drawn length.
  • Thermal history. Cold work and annealing affect the metallurgical state, and the measured EMF depends on that state as well as on composition. A leg that was annealed differently from its partner will not track it.
  • Contamination. Surface residue, lubricant or oxidation at the junction changes the local alloy at the exact point where the voltage is generated.

None of those four show up on a chemical certificate. All four show up on a thermoelectric check, which is why the two tests are complements rather than alternatives.

What each document actually covers

DocumentProvesDoes not proveAsk for it when
EN 10204 2.2 declarationProduct complies with the order, based on non-specific inspectionThat the delivered batch was testedThe application is non-critical and cost matters
EN 10204 3.1 certificateSpecific test results for the delivered batch, signed by an independent inspection representativeThat the pair generates the expected EMFYou need traceable composition evidence per heat
Certificate of conformityThe supplier declares the goods meet the stated specificationAny measured value at allIt sits alongside a stronger document, never instead of one
EMF check against the reference tableThe pair produces output within a stated tolerance bandComposition, or behaviour after your welding and assemblyThe wire will form the measuring junction
Calibration certificate for a finished probeThat one assembly reads within a stated uncertainty at stated pointsAnything about the rest of the spoolThe probe is a reference or the process is regulated

The rows are not ranked. A calibration certificate for a single probe is stronger evidence than a batch EMF check, but it only covers that probe. Buying a calibration certificate and assuming it applies to the spool is one of the more expensive misreadings in this list.

Writing both into a purchase order

A purchase order that asks for 'material certificate and test report' will be satisfied by almost anything. Naming the document grade and the test method is what makes it enforceable. A workable clause has four parts: the inspection document grade and that it must be specific to the delivered heat; the EMF requirement stated as a tolerance class or as limits against the reference table, with the test method named; the sampling basis, meaning how many spools or which end of the spool is tested; and a retained sample held against the batch number.

The sampling line is the one buyers forget, and it decides how much the report is worth. A single EMF check at one point on one spool is a statement about that sample. It is useful evidence about the process, not a guarantee about every metre on the pallet. If the application needs more assurance, the honest way to get it is a defined sampling plan or incoming inspection, not a more strongly worded certificate request.

What a wire mill can sign, and what it cannot

A mill drawing thermocouple alloy can issue a certificate for the wire it made: composition per heat, diameter, temper, surface, and the EMF of the pair it shipped, within the tolerance class it quoted. It cannot certify how the wire behaves after your welding process, what the junction does in your atmosphere, or how the assembly drifts after a year at temperature. Those belong to the sensor design and to service conditions.

That boundary is worth stating in the order, because it protects both sides. A supplier who promises service life on a wire certificate has promised something the certificate cannot carry.

One line to remember

The certificate tells you what the wire is. The EMF check tells you what the pair does. An order for a measuring junction needs both.

The short version

  • Name the inspection document grade, not just the words material certificate.
  • State the EMF requirement against a tolerance class or the reference table, and name the test method.
  • State the sampling basis, or accept that the report describes a sample rather than a shipment.
Questions we get asked
Is a 3.1 certificate better than a 2.2 declaration?
It carries more evidence, not necessarily better wire. A 3.1 certificate reports specific results for the delivered batch and is signed by an inspection representative independent of production. A 2.2 declaration states compliance based on non-specific inspection. Choose by what the application needs to prove.
Can a supplier provide an EMF report without a material certificate?
Yes, and the combination is weaker than it looks. The EMF report shows the pair behaves correctly at the test points; the material certificate shows the alloy is what was ordered. If you need to trace a heat or resolve a dispute about composition, only the inspection document helps.
Does an EMF check at one temperature cover the whole range?
No. A check at a single temperature difference confirms the pair at that point. Deviation can grow at higher temperature, particularly for nickel-based types, so the test points should match the range you will use.
Citable facts
Sources

Send the specification — diameter, grade, temper, quantity.

We come back with a confirmed lead time and the documentation we can issue for the shipment. 100 g sample at cost; orders from 1 kg.

Request a quotation Order a 100 g sample