We have written elsewhere about this window with its Winkhaus hardware set, and about what a passive label asks of the hardware. This piece starts where that one ends. It is about the sheet of numbers itself, and about a column that almost no window specification sheet in the world prints: the column that says what each number was measured on.
The word passive is doing three different jobs at once. In China it is a national standard word with a number attached. In Germany it is a certification with a different number. On a product page it is a name. All three are real, none of them are interchangeable, and a specification sheet that does not say which one it is answering has not yet told you anything you can submit.
Before you can read our numbers, you need to see the three the word is pointing at. They are all expressed in the same unit, W/(m²·K), which is precisely what makes them dangerous: they look comparable, and they are not.
| What is asking | The number | Measured on | Where the number comes from |
|---|---|---|---|
| China, near-zero energy buildings | window K at or below 1.2 | the window element in the project, with a solar heat gain requirement attached | GB/T 51350-2019, and the provincial or municipal documents written under it |
| Passive House Institute, cool-temperate | Uw at or below 0.80, installed at or below 0.85 | a fixed reference window of 1.23 by 1.48 metres | PHI criteria for transparent components, calculated to EN ISO 10077 |
| This product sheet | K = 1.3 | the published system value for this series | our own catalogue data |
The gap between the three is not a quality judgement. It is three different measurement objects. A published comparison of the Chinese and German systems that circulates in the Chinese window industry puts them side by side in the clearest way available: the Chinese national near-zero energy standard asks for a window K of 1.2 and an annual heating demand at or below 20 kWh per square metre per year; the German Passive House Institute asks for a window Uw of 0.80 and an annual heating demand at or below 15 kWh. Both set the building airtightness target at 0.6 air changes per hour at 50 pascals. Same method of measuring the building, different number on the window, because the two standards were written for different climates and different energy floors.
The Passive House side is also not a single number. The published criteria divide the world into climate zones, and each zone gets its own whole-window limit, calculated on the same 1.23 by 1.48 metre reference window:
| Climate zone | Component Uw | Installed Uw | Reference glazing Ug |
|---|---|---|---|
| Arctic | 0.40 | 0.45 | 0.35 |
| Cold | 0.60 | 0.65 | 0.52 |
| Cool-temperate | 0.80 | 0.85 | 0.70 |
| Warm-temperate | 1.00 | 1.05 | 0.90 |
| Warm | 1.20 | 1.25 | 1.10 |
Two things in that table matter more than the headline 0.80. First, the limit is attached to a climate zone, and most of China's heating-dominated belt, most of the northern United States, southern Canada and central Europe sit in cool-temperate. Move the same project to a colder zone and the requirement tightens by half. Second, the reference window is fixed at 1.23 by 1.48 metres. That is not a formality. A whole-window value is an area-weighted blend of glass and frame, so the same construction tested at your actual opening size will give a different number than at the reference size, in the direction you would expect: more glass area usually helps, more frame share hurts.
Here is the part that no window catalogue tells you, and that two independent Chinese sources say very plainly. When a Chinese document sets thermal requirements for an ultra-low-energy window, it does not give one number. It gives three, as three separate rows:
That is the severe-cold region technical specification published in Heilongjiang. A second Chinese industry source, listing the grade requirements for a passive door or window, gives the same structure: glass at or below 0.8, frame profile at or below 1.3, and then three class requirements rather than thermal ones, air permeability not lower than class 8, water tightness not lower than class 6, wind load resistance not lower than class 9.
Look at where 1.3 appears. Twice, in two unrelated documents, it is the frame row. It is never the whole-window row. Now look at the number on our sheet.
We are not going to tell you our published 1.3 is a frame value, because we are not certain, and guessing on a submittal is how a project loses a week. We are going to tell you that this is the question, and it is the first one that decides whether your document passes. The reason it matters is arithmetic. A whole-window value is:
Uw = (Ug · Ag + Uf · Af + Ψ · lg) divided by (Ag + Af)
The frame is one of three terms, and it is multiplied by the frame area, which is the smaller of the two areas. Glass carries the rest. So a frame number and a whole-window number are not competing versions of the same figure; they are different variables in the same equation. If you put a frame value into a submittal line that asks for a whole-window value, you have not submitted a worse window. You have submitted the wrong variable.
There is a second thing worth knowing, and it is the most useful single question on this page. The Chinese national product standard for aluminium windows and doors, GB/T 8478-2020, sorts windows by their main performance into six types. They are the ordinary type; the sound-insulating type, at 35 dB or better; the thermal insulating type, defined as a heat transfer coefficient K below 2.5; the heat-shielding type, defined as a solar heat gain coefficient no greater than 0.44; the combined thermal and heat-shielding type; and the fire-resistant type.
There is no category called passive. The closest thing the standard recognises is the thermal insulating type, and its entry threshold is a K below 2.5. That is a long way from any of the three numbers in the first table, and it is deliberately a floor, not a grade.
This gives you a question that is very hard to answer with marketing. Ask which of the six types the product was type-tested as, and ask for the type-test report number. The 2020 revision of the standard added a requirement that the product quality guarantee must state the type-tested performance values together with the number of the type-test report. A supplier who can produce that number is talking about a tested article. A supplier who answers with a series name is talking about a family.
A specification sheet is filled with the numbers a factory can test. That is not a criticism, it is an observation about what ends up on paper, and it explains a pattern you will see on almost every high-performance window sheet: the durable-looking numbers are usually not the thermal ones.
| Number on the sheet | What it actually tests | What it does not tell you |
|---|---|---|
| 240-hour salt spray resistance | how the surface finish holds up under a corrosive atmosphere over a fixed exposure period | anything about heat flow, air leakage or glass. It is a coating and substrate result |
| 5,000 newton shear resistance | whether the insulation strip stays bonded to the aluminium in the composite thermal-break profile under load | the width or material of that strip, or the resulting frame value. It is a mechanical result |
| 1.8 mm profile thickness | the nominal base wall thickness of the main load-bearing profile | anything above compliance. 1.8 mm is the floor the national standard sets for external windows, not a premium tier |
| 6060-T66 profile grade | the alloy designation and temper condition | how a series performs against another. Every series in this catalogue uses the same alloy and temper, so it cannot be the differentiator |
Two honesty notes that belong here. The salt spray result and the shear result are published for this series. They do not transfer to the outward-opening casement series, the side-press sliding window, or the sliding doors. And 1.8 mm deserves a plain reading: GB/T 8478-2020 raised the external window requirement to a nominal base wall thickness of 1.8 mm, with a minimum measured value around 1.7 mm, and it set external doors at 2.2 mm and internal windows at 1.4 mm. Meeting 1.8 mm means the window is compliant. It does not mean it outranks another compliant window.
Now to the numbers that do carry passive consequences, and they are the ones that look the least glamorous. This sheet carries two figures that are both 90 mm, and they are not the same 90 mm.
Line the catalogue up by that second figure and the reason this series carries the word becomes obvious. The side-press sliding window has a 55 mm glass sash. One outward-opening casement series has 57 mm, another 62 mm, a third 81 mm. This one has 90 mm. It is the deepest glazing pocket in the range, and depth in the pocket is what lets you specify a triple unit with real cavity width rather than a squeezed one.
Why does that matter for passive specifically? Because in every one of the three rows above, the glass is held to a tighter number than the frame. Heilongjiang wants glass at or below 0.8 and the frame at or below 1.3. The Passive House reference glazing for cool-temperate is 0.70. The glass is where the work is done, and you cannot get there with a shallow rebate. GB/T 8478-2020 also puts a floor under the cavity itself: the gas space in an insulating unit for external windows and doors should not be less than 9.0 mm. A pocket that cannot take the unit is not a thermal problem you can fix with a better frame.
This sheet publishes the mullion as 72/4 mm. Read it the way it is written. The first figure is the depth, the structural dimension that carries wind load across the span. The second is the visible face, the line you see when the window is closed. Compare the second figure with the rest of the catalogue, where a 22 mm visible mullion is marketed as slim, and 21 mm and 31 mm appear on the casement series.
A visible face measured in single millimetres is not a narrower version of the same thing. It means there is essentially no cover cap: the joint between the two sashes is the finished surface. That is a legitimate and good-looking detail, but it moves three questions onto the table that a capped mullion answers by default. What carries the load across that joint. What seals it, and in which plane. And what happens to the seal when the two sashes move relative to each other, because on a tilt-and-turn they will. Ask for the section drawing with the joint dimensioned, not the elevation.
Our sheet for this series lists the thermal break as a multi-chamber insulation strip with an insulating sponge. Our sheet for one of the casement series names the material, PA66GF25, and gives a width range of 34 to 44 mm. That difference is worth noticing, because the width in millimetres is the single most predictive number for the frame value, and it is the number most often left off a catalogue.
A published comparison of provincial ultra-low-energy standards, written by a Chinese window manufacturer and widely circulated, gives thermal-break aluminium windows a required strip width for each representative city. Put next to the whole-window target, it reads like this:
| Representative city and zone | Whole-window K the local standard asks for | Strip width the same source specifies | Glazing it assumes |
|---|---|---|---|
| Harbin area, severe cold | below 0.8 | 60 mm or more, with insulation fill in the frame | four panes, three cavities, gas filled, warm edge |
| Beijing, cold | 0.8 target, 0.8 to 1.0 current | 57 mm or more | three panes, two cavities, gas filled, warm edge |
| Shanghai, hot summer cold winter | 1.80 limit, 1.40 reference | 32 mm or more | three panes, two cavities, with some shading |
| Kunming, mild | 2.0 or lower | 24 mm or more | double glazed |
| Shenzhen, hot summer warm winter | 2.5 or lower | 20 mm or more | double glazed, shading dominates |
Read that table honestly and it tells you something useful about where a 34 to 44 mm strip lands. On that source's numbers, it sits comfortably in the mild and hot-summer zones, and it does not reach the strip width the same source specifies for Beijing or the severe-cold provinces. We would rather say that than pretend. It also tells you the follow-up question, which no catalogue answers: for this series, what is the strip material, and what is its width in millimetres.
This is the section to read if you read only one. Passive performance is not decided by the frame value alone. The published Passive House component criteria add requirements that are invisible on a product page, and three of the four are decided where the window meets something else.
| The row | The threshold that gets quoted | Who actually decides it | Why it is not on a window sheet |
|---|---|---|---|
| Air permeability | EN 12207 class 4, tested at 600 pascals, for the German route. Class 8 under GB/T 7106-2019 for the Chinese route, which is an air leakage per metre of joint at or below 0.5 cubic metres per metre per hour and per unit area at or below 1.5 | the window, but only as tested in a rig | the two systems use different pressures, different quantities and different scales, so the numbers cannot be compared even after you convert the units |
| Condensation safety, fRsi | 0.70 or better at every point of the frame | the frame section, calculated | it is a hygiene criterion about mould, not an energy one, and it lives in a calculation report rather than a product table |
| Warm edge, glass-edge psi value | 0.030 W/(m·K) or better | the spacer bar in the sealed unit | it is a component of the glass unit, bought by the glazier on a different purchase order. An aluminium spacer bar will not reach it |
| Installation psi value | the installed window is allowed 0.85 where the component is allowed 0.80 | the wall detail, not the window | it is not part of EN ISO 10077 at all. It was added by the certification body precisely because the window cannot answer it |
That last row is the one that ends arguments. A window can be exactly as good as its sheet and the assembly can still fail an installed-state limit, because the extra allowance exists to absorb a linear thermal bridge created by how the window sits in the opening. The Chinese standard reaches the same conclusion by a different route: GB/T 51350 is widely read as requiring external windows to be installed in the outer position, with the frame sitting mostly outside the insulation layer, and it sets out the insulation, airtightness and weatherproofing requirements for the junction itself. Either way, the number belongs to the detail, and the detail belongs to someone else's drawing.
There is a fifth row, and it is the one this catalogue is weakest on: solar heat gain. Chinese documents do not treat it as optional. The cold-zone tables ask for a winter solar heat gain coefficient of at least 0.45 and a summer figure no higher than 0.30, and they note that the summer value is the combined coefficient including shading but excluding internal shading. The German guidance talks about a g-value around 0.50 to make passive solar gain worth having in a heating-dominated climate. We do not publish a solar heat gain coefficient for any series. If your submittal needs one, it has to come from a calculation on your glass build-up, not from this page.
The four rows above are per-element. The building is not. Once a project mixes materials and opening types, the comparison has to be done on one basis or it is meaningless. If part of the same elevation is a steel entry door with window rather than an aluminium thermal-break unit, its K value arrives from a different report, on a different frame material, probably at a different specimen size. Do not average them. Same for a double glass swing door placed on the same facade: the frame share of a doorset is different from a window's, so the area weighting changes too.
And the hardest case of all is the one where glass stops being a hole in a wall and becomes the wall. A 120 aluminum sun room supply package is mostly glazed area, which turns the trade-off inside out: the frame value barely matters, the glass value matters enormously, and solar heat gain stops being a compliance line and becomes the whole comfort question. In cooling-dominated zones a low U with a high g is the wrong answer, and in heating-dominated zones the reverse. Anyone who quotes you a single number for that kind of space has not read the brief.
Having spent most of this page on what the sheet does not settle, it is fair to say what it does settle, and where this series is genuinely the right starting point.
| If your brief says this | What the sheet gives you | What you still have to get from elsewhere |
|---|---|---|
| Inward opening, and cleanable from inside on an upper floor | tilt and turn or inward opening is the declared method. This is the only series in the catalogue that does it | confirmation that the tilt position still meets the escape-opening dimension your authority asks for |
| A shallow reveal, or a retrofit into an existing opening | 90 mm frame wall width, the shallowest in the range | a measured site opening in millimetres. The nominal name on a drawing is not the manufacturing size |
| A heavy triple-glazed unit | 90 mm glass sash thickness, the deepest pocket in the range | the rebate dimension from the section drawing, and the hinge rating for the resulting sash weight |
| A very slim visible mullion on a divided elevation | 72/4 mm mullion pair, the smallest visible face in the range | the joint section, dimensioned, and the wind load that depth has been checked against |
| A coastal or pool-side elevation | 240-hour salt spray resistance, published for this series | the fastener material specified separately. Corrosion usually starts at the screw, not the hinge |
| Item | On the sheet | If it is not, why not |
|---|---|---|
| Thermal, K value | 1.3, as a system value for this series | a system value is not a whole-window value on the reference size, and it is not a certified figure. Ask which row it belongs to and on what specimen |
| Solar heat gain, SHGC or g | not published for any series | it depends on the glass build-up you choose, which is a project decision. It has to be calculated per configuration |
| Air, water and wind classes | not published for any series | classes are awarded to a tested specimen of a stated size and configuration. A series name cannot carry them |
| Thermal break material and width | multi-chamber strip with insulation sponge. Material and millimetre width not stated for this series | an omission on our sheet, not a defect in the product. Ask and it can be supplied from the section drawing |
| Maximum sash size and weight | not published for this series | any maximum is quoted against an assumed glass thickness. Without the glass build-up the number means nothing |
| Salt spray and shear | 240 hours and 5,000 newtons, this series only | test results belong to the tested article. They do not move to another series because the alloy is the same |
| Passive House or equivalent certificate | none held | component certification is issued to the applicant for a specific tested configuration. We can supply profiles, sections, test evidence and specimens. We cannot supply a certificate we do not hold |
| Ask this | A real answer looks like | Sounds reasonable, tells you nothing |
|---|---|---|
| Is the published K the whole window, or the frame profile | a named row, a standard, a specimen size and a report number | "our K is 1.3, which is very good" |
| Which of the six types in GB/T 8478-2020 was it type-tested as | a type name plus the type-test report number | "it passes all national standards" |
| Air permeability class, to which standard, at what pressure | class, standard number, test pressure, specimen size, whether the specimen was an opening sash | "it is very airtight" |
| Strip material and strip width in millimetres | material designation and a number in millimetres, from the section drawing | "multi-chamber, high performance" |
| Spacer bar type in the sealed unit, and its psi value | warm edge named, psi value quoted in W/(m·K) | "standard spacer" |
| Whole-window U on the 1.23 by 1.48 metre reference, or a stated size | a number with the size it was calculated at, and the glass build-up | "our system reaches passive level" |
| Solar heat gain coefficient for the glass you are quoting | a number tied to a named glass build-up | "low-e glass, energy saving" |
| Installation detail and who calculates the installation psi value | a junction drawing, and a named party taking responsibility | "the installer will handle it" |
| Maximum sash size and weight, and the glass build-up assumed | millimetres, kilograms, and the glass configuration the rating assumes | "large sizes are no problem" |
| Does the salt spray and shear evidence cover this series | yes or no, with the report reference and the article tested | "we use the same alloy throughout" |
No, and we would rather say so than let a product name imply it. Component certification is issued by the certifying body to an applicant for a specific tested configuration, and it does not travel with a series name or across a border. What we can give you is the profile sections, the material and test evidence we hold, calculation inputs and physical specimens, so that whoever is pursuing certification on your project has something real to submit.
We cannot answer that, because the question is missing a row. 1.3 appears in Chinese ultra-low-energy guidance as the frame profile limit, and in the same documents the whole-window limit is a different, tighter figure. If your submittal line asks for a whole-window value, 1.3 is not the number to put on it, whatever it turns out to be. Ask for the whole-window value at a stated size and glass build-up, and compare like with like.
Because those two are properties of the material and the composite profile, and they can be stated for a series. Air, water and wind classes are awarded to a tested specimen of a stated size and configuration, so a series name cannot carry them honestly. It is not a gap we are hiding; it is a gap that only closes when you tell us the size, the glass and the opening method you actually want.
Mostly yes, and that is exactly why the 90 mm glass sash thickness is the most important number on this sheet. Two cautions. Glass is heavy, at roughly 2.5 kilograms per square metre for every millimetre of thickness, so a heavier unit moves you along the hinge rating table and changes the maximum sash size. And a published maximum size is always quoted against an assumed glass build-up, so changing the glass can change the size limit too. Specify glass, hardware and sash size together, never separately.
It is the right thing to worry about, because the question behind it is the fRsi criterion, which asks whether any point of the frame runs cold enough to grow mould. A very slim visible face usually means a structural joint rather than a capped one, so the answer depends on what sits behind that 4 mm and how the thermal break continues through the joint. Ask for the isotherm drawing and the calculated fRsi, not for an assurance.
No, and mixing the two is a common and expensive error. If the project is assessed under GB/T 51350 and the local document written beneath it, work entirely in Chinese classes and Chinese test methods, and check the climate zone your project sits in, because the limits differ sharply between the severe-cold north and the hot-summer south. Keep the German framework only if the project is actually pursuing that certification.
