In short

A window certificate is a declaration of performance for a tested specimen: one product, at one size, built to one configuration, measured by one method. It is not a statement about the opening in your wall, and it is not a promise about the batch in the container.

That decides how you read the rest of the file. Two suppliers can both send a valid document, an NFRC label and an EN 14351-1 declaration, quoting performance that looks compatible and is not: different methods, units and signatories, and a specimen size that rarely matches the schedule.

01Only one of these two routes is checked by a third party

An NFRC label is issued under a third-party certification programme. NFRC 100 produces the U-factor, NFRC 200 the solar heat gain coefficient and visible transmittance, and the label prints those with an air-leakage figure. The label belongs on the product: NFRC's FAQ states that labels cannot be sent separately from the window.

The European route is a declaration. Under the Construction Products Regulation the manufacturer carries out the conformity assessment, compiles the technical file, issues the declaration of performance and affixes the CE marking. EN 14351-1 is the harmonised standard that declaration cites, and the Commission describes the CPR as a common technical language for comparing products.

What each route documents, who signs it, and what it leaves out
Question NFRC label and certificate CE marking with an EN 14351-1 declaration
Legal basisNFRC certification programme (US)Construction Products Regulation; EN 14351-1
Who states performanceNFRC, third-party certifier, from tested dataThe manufacturer, in the declaration of performance
Thermal valueU-factor to NFRC 100, in US unitsUw to EN ISO 10077-1, in W/(m²·K)
Other valuesSHGC and VT between 0 and 1; air leakageAir class to EN 12207, water to EN 12208, wind to EN 12210
InstallationNot coveredNot covered
What it does not proveThat the delivered unit matches the tested oneThat a laboratory tested this batch

Both routes describe a product, not an opening, so schedules written around them are not interchangeable.

Decision rule: in a US schedule, accept a unit only when the label is on it, or when a project label certificate carries the project address and the certified ratings; in an EU schedule, ask for the declaration and the standard it cites. A datasheet with a CE logo is marketing.

02A class is a band, and a number is a measurement

EN 12207 sorts air permeability into classes 1 to 4, EN 12208 watertightness into 1 to 9A, and EN 12210 wind resistance into 1 to 5. A class records the band a specimen reached, not where inside it. Class 4 is the highest band, and two class 4 windows can still leak by different amounts.

The NFRC label does the opposite and prints a number: air leakage in cubic feet per minute per square foot of product area, where 0.3 or below is the common reference for a well-sealed unit. The European document prints a class band instead. Neither presentation is wrong, and a schedule that puts both in one column has already lost the argument.

The signatories differ too: a European class is declared by the manufacturer from initial type testing, an NFRC figure is certified after testing. To rank two quotes, ask for the tested value, the test pressure and the specimen size.

Clause to paste: air permeability to EN 12207, class not lower than 3, tested on a specimen of the size and sash configuration scheduled, with the report naming the operable element. If nobody can say which element was tested, the class belongs to a window you are not buying.

Catalogue reference: windows and exterior doors in the catalogue — models, dimensions and finishes for the directions discussed above.

03Uw is an average, so every certificate carries a size

Under EN ISO 10077-1 the whole-window value is an area-weighted average of three terms: Uw = (Ag·Ug + Af·Uf + lg·Ψg) / (Ag + Af). Ag is glass area, Ug the glass value, Af frame area, Uf the frame value, lg the length of the glass edge and Ψg the linear bridge at that edge. Every term belongs to the size as much as to the product.

The table changes only the opening. Assumptions: 70 mm frame face, Uf 2.4 W/(m²·K) for a 24 mm break, Ug 1.10 W/(m²·K), Ψg 0.06 W/(m·K) for a warm-edge spacer. The arithmetic is ours; substitute the supplier's declared values.

Worked example: the same frame, glass and spacer at three opening sizes, calculated to EN ISO 10077-1
Opening (m) Area (m²) Frame area (m²) Glass edge (m) Uw W/(m²·K) Frame share
0.6 × 0.90.540.192.441.8335%
1.2 × 1.51.800.364.841.5220%
2.4 × 2.45.760.659.041.3411%

One product, 0.49 W/(m²·K) of difference across three sizes. A certificate states the size it applies to; anything else is an extrapolation.

Units catch buyers as often as size. NFRC U-factors are printed in US units, and metric Uw divides by 5.678: the 1.52 W/(m²·K) above is 0.27 Btu/(h·ft²·°F), so a US label showing 0.30 describes a worse window than the second row.

Decision rule: require Uw at each scheduled size, with Ag, Af and lg stated, and reject one value offered for mixed openings. Eight sizes means eight declarations, or one method plus a worked example the supplier will defend.

04The frame decides which window in the building fails

Aluminium conducts heat at roughly 160 W/(m·K); polyamide 66 with glass fibre conducts at about 0.3 W/(m·K). That factor of roughly 530 is why the thermal break, not the glass, sets the frame value, and why a 14.8 mm section looks much like a 34 mm one on a drawing.

Indicative frame U-values by thermal-break type; planning references, not certified product data
Frame type Indicative Uf W/(m²·K) Typical use What to ask
Plain aluminium5.6–6.0Interior screens, unbroken framesWhether a break is present at all
14.8 mm break3.0–3.5Light commercial, mild climatesBreak width and polyamide grade
24 mm break2.0–2.8Residential windowsWhich Uf the certificate used
34 mm multi-chamber1.6–2.2Cold climates, large glazed areasChamber count and Uw at size

Indicative only. A certified value comes from the product's own report or declaration; use these bands to judge whether a quoted Uf is plausible for the frame depth in the drawing.

Re-run the example with plain aluminium at Uf 5.9 and the three sizes become 3.06, 2.22 and 1.74 W/(m²·K). The spread widens to 1.32, and the smallest opening is now almost twice as bad as the largest. On a house with one large sliding unit at the front and six small bathroom windows at the back, the large unit is the figure that gets quoted and the bathrooms are the ones that condense.

Decision rule: calculate the smallest opening, not a representative one. At 35% frame share a third of that window is the weakest component, and no glass upgrade compensates.

05The tested unit and the delivered unit diverge at four points

A warm-edge spacer reduces the linear bridge at the glass edge from roughly 0.10–0.12 to 0.04–0.06 W/(m·K). On the 1.80 m² window that moves Uw by 0.13 W/(m²·K); on the 0.54 m² window it moves it by 0.23 W/(m²·K), because the glass edge per square metre is longer.

Gas fill is the second point. A double-glazed unit typically has Ug 1.0–1.4 W/(m²·K) and a triple unit 0.5–0.8, and argon improves Ug by roughly 0.1–0.2. At 1.80 m², moving Ug from 1.10 to 0.65 changes Uw from 1.52 to 1.16, so the glass make-up is worth more than the spacer.

The fixing is the third. A metal fixing or hinge that crosses the thermal break short-circuits it locally, which is why fixing position and hardware isolation belong in the tested configuration. A 34 mm multi-chamber frame with one unisolated bracket is not the frame that was certified, and the condensation will appear in a patch around the bracket.

The fourth is the sash. A fixed light and an opening sash are different air paths, and a certificate covering a fixed unit does not cover the casement you substitute for it. If the element is unnamed, assume the fixed variant.

06Neither document covers the perimeter joint

NFRC's FAQ is direct: performance ratings do not take into account how a product is installed. The European declaration has the same boundary, because EN 14351-1 describes the product and the joint between frame and reveal is site work with its own sealant and its own inspector.

The 1.2 × 1.5 m window above has 5.4 m of perimeter; the 2.4 × 2.4 m window has 9.6 m. A laboratory reaches class 4 by sealing the specimen into the chamber wall. On site that 5.4 m is backing rod, sealant and plaster, inspected by nobody on the certificate.

Two clauses close most of the gap. Name the joint: backing rod diameter, sealant type, and whether it is filled from inside, outside or both. Name the fixing: bracket spacing, packer position, no fixing through the break. Then photograph both before the reveal is closed, since after plastering no evidence is left.

Decision rule: if the installation method is not in the same document as the performance, the performance is a factory claim and the site is unverified.

07Reading a certificate in ten minutes

A review is a sequence, not a judgement. Six checks separate a document that describes your window from one that describes a window.

  1. Identify the standard and the issuer: NFRC certification, or a declaration citing EN 14351-1.
  2. Find the specimen size and compare it with the largest and smallest openings in the schedule.
  3. Read what is declared: U-factor, SHGC, VT and air leakage, or Uw plus the EN 12207, EN 12208 and EN 12210 classes.
  4. Check the configuration: spacer, gas, glass make-up, frame Uf and element tested.
  5. Check status: a label on the unit or a project label certificate with the address; a declaration naming the product family and standard revision.
  6. Write the result into the RFQ as a required submittal, at the scheduled size, so every supplier answers in a comparable form.

The sixth step changes a purchase. A supplier asked for a certificate sends the page that reads well; one asked for the declaration at the scheduled size must produce it or decline in writing. Only the second is discoverable before the container arrives.

What your RFQ should state

  1. Which route the project requires: an NFRC label on each unit, or a declaration citing EN 14351-1.
  2. The Uw or U-factor required at the scheduled size, with Ag, Af and lg stated in the supplier's calculation.
  3. The glass make-up, gas fill and spacer type, named as the tested configuration and not as an equivalent.
  4. The element tested, fixed light or opening sash, with the sash type and hardware named for operable units.
  5. Air permeability and watertightness as EN 12207 and EN 12208 classes, with the report naming the specimen size.
  6. The frame Uf and thermal-break width in millimetres, quoted for the profile actually supplied.
  7. The perimeter joint and the fixing detail: backing rod, sealant, bracket spacing, no metal through the break.
  8. The submittal date for the certificate, the declaration and the installation detail, before production starts.

Where this guidance stops

Questions buyers ask

What does an NFRC label actually certify?

It certifies that a product was rated under the NFRC programme to NFRC 100 and 200, and that the label reports the resulting U-factor, solar heat gain coefficient, visible transmittance and air leakage. The rating belongs to the product and the size tested. NFRC's FAQ notes that a label cannot be sent separately from the window, so a PDF alone proves nothing about the unit delivered.

Is a CE mark the same as NFRC certification?

No. CE marking follows a declaration of performance the manufacturer prepares under the Construction Products Regulation, citing EN 14351-1. The manufacturer carries out the conformity assessment and issues the declaration; NFRC is third-party certification of tested ratings. A CE mark shows performance was declared in the harmonised format, not that a certifier checked the batch.

Which Uw should a supplier declare for my window?

The one calculated for the size you are buying, with the glass and frame areas stated. Uw is an area-weighted average, so a 0.6 × 0.9 m opening scores worse than a 2.4 × 2.4 m one with the same frame and glass. Ask for the calculation to EN ISO 10077-1 at each scheduled size, and treat a single figure for a mixed schedule as an extrapolation.

Can I compare a class 4 air result with an NFRC air-leakage number?

Not directly. EN 12207 class 4 states a band the specimen reached; the NFRC label prints a number, in different units, from a different procedure. Neither body publishes a conversion. If you must rank two suppliers, ask for the tested value, the pressure and the specimen size in the same format.

Does the certificate cover installation?

No. NFRC states that its ratings do not take account of installation, and an EN 14351-1 declaration describes the product, not the joint. The perimeter seal, the packers and the fixings are site work. Put them in the RFQ and in the inspection scope, and photograph the joint before the reveal is closed.

What if the certificate is for a different size than my opening?

Then it is evidence about a different window, and you should ask for a calculation at your size. A larger glazed area usually lowers Uw because the frame share falls, so extrapolating from a small specimen flatters a large unit and understates a small one. The gap matters most on small openings, where the frame can be a third of the area.

Which documents should I keep after handover?

The NFRC label or project label certificate, the declaration of performance with the standard reference, the calculation at each scheduled size, the tested configuration list and the installation detail as built. Keep them with the O&M file: a condensation complaint or an insurance query two winters later is settled by those pages, not the showroom sample.

Sources

  1. ISO 10077-1 — Thermal performance of windows, doors and shutters: calculation of thermal transmittance — International Organization for Standardization
  2. EN 14351-1 — Windows and doors: product standard, performance characteristics — European Committee for Standardization
  3. NFRC — certification, ratings and the NFRC label — National Fenestration Rating Council
  4. NFRC frequently asked questions — labels, installation and certification — National Fenestration Rating Council
  5. AAMA 1503 — Voluntary test method for thermal transmittance of fenestration systems — American Architectural Manufacturers Association
  6. CE marking — manufacturer obligations and the declaration of conformity — European Commission
  7. Construction Products Regulation (CPR) — harmonised rules and declarations of performance — European Commission

Pages related to this guide

How this guide was checked

FS

Foshan Supply editorial team · Catalogue indexMethodology

Written in Foshan and reviewed against the catalogue and public standards on 2026-11-05. This page is a buying guide, not a quotation or a warranty. How we work.

How to cite this page: Foshan Supply, “NFRC certification for windows: the size it does not cover”, 2026-11-05, https://foshansupply.com/blog/nfrac-vs-en-14351/