Search for casement window container loading and you will mostly find packing advice: wrap the profiles, protect the corners, block the load. All of that is true and none of it is the hard part. The hard part is that a container is a fixed steel box with a hole in one end, and three of the numbers that decide what you can put in it appear on no window drawing, no shop drawing and no purchase order.
We have written elsewhere about how casement windows are packed and what a shipment needs to survive. This article starts where that one cannot reach: the arithmetic of the box itself. Because the decisive discovery for most first-time importers is not that the crate was weak. It is that the sash was 120 mm too tall to stand up, and nobody found out until the truck was at the door.
| The number | Where it lives | Who owns it | What happens if nobody checks it |
|---|---|---|---|
| Door aperture height, 2,280 mm on a standard box | Container manufacturer specification sheet | The container, not the window factory | Tall sashes cannot travel upright; the lean is discovered at the loading bay |
| Maximum payload | CSC plate on the container door | The carrier's stowage planner | Not the binding limit for windows, which is the mistake everyone makes in the other direction |
| Centre of gravity, below half the internal height | IMO, ILO and UNECE Code of Practice for Packing of Cargo Transport Units | The packer, explicitly | A top-heavy box that is refused, re-stuffed, or that rolls a truck on the destination road |
Notice the third row. The CTU Code does not leave the loading plan to whoever is nearest. One European cargo securing specialist states it plainly: responsibility is transferred to the packer or stacker. That matters more than it sounds, because the packer is often not the factory, not the freight forwarder and not you. It is a third party at the loading point, working from a plan somebody else wrote.
Every container specification sheet prints an internal height and a door aperture height. They are not the same number, and the gap between them is the single most expensive piece of missing information in fenestration freight.
| Container | Internal length | Internal width | Internal height | Door aperture | Internal volume | Max gross | Max payload |
|---|---|---|---|---|---|---|---|
| 20 foot standard (published by a container manufacturer) | 5,898 mm | 2,352 mm | 2,393 mm | 2,340 by 2,280 mm | 33.2 cubic metres | 30,480 kg | 28,360 kg |
| 40 foot standard (published container specifications) | 12,032 mm | 2,352 mm | 2,393 mm | 2,340 by 2,280 mm | About 67.7 cubic metres | 30,480 kg | About 26,700 kg |
| 40 foot high cube (published by a carrier) | 12,031 mm | 2,352 mm | 2,698 mm | 2,340 by 2,585 mm | 76.3 cubic metres | 32,500 kg | 28,700 kg |
Three things fall straight out of that table, and all three are routinely missed.
First: the door is 113 mm lower than the box. On a standard container the internal height is 2,393 mm and the door aperture height is 2,280 mm. A container specification publisher puts the consequence in one sentence: any load taller than 2,280 mm will not pass through the doors, even though the interior can accommodate up to 2,390 mm. The extra 113 mm exists, but only behind the header. You cannot use it on the way in.
Second: buying a longer container does not buy you a taller door. The 20 foot and the 40 foot standard boxes publish the identical door aperture, 2,340 by 2,280 mm, and the identical internal cross-section. A 40 foot box gives you twice the length and not one millimetre more height. Only the high cube moves the number, by 305 mm on both the internal height and the door height.
Third: the 20 foot box is rated to carry more than the 40 foot box. Both publish the same maximum gross mass of 30,480 kg, and the 40 foot box is heavier empty, so its payload is lower. One container publisher calls this out as a fact that surprises many beginning freight forwarders. It is also the reason heavy cargo goes in 20 foot boxes and bulky cargo goes in 40 foot boxes, and windows are emphatically the second kind.
A sash that is too tall to pass the header upright can still travel. It just has to lean, and the lean is not free. Lean a panel back by an angle from vertical and two things happen: its top drops, which is what you want, and its footprint along the floor grows, which is what you pay for. The geometry is a single arccosine.
For a panel of height H passing an aperture of height A, the minimum lean from vertical is arccos(A divided by H), and the floor length the panel then occupies is H multiplied by the sine of that angle. That is our own arithmetic, applied to published aperture dimensions. The table below uses it for a range of sash heights, and it excludes the panel's own thickness, which you must add on top.
| Sash height | Standard box, 2,280 mm door | Floor length eaten, standard | High cube, 2,585 mm door | Floor length eaten, high cube |
|---|---|---|---|---|
| 2,032 mm (an 80 inch leaf) | Stands upright | Only its own depth | Stands upright | Only its own depth |
| 2,280 mm | Stands upright, exactly at the limit | Only its own depth | Stands upright | Only its own depth |
| 2,400 mm | Lean at least 18.2 degrees | 749 mm | Stands upright | Only its own depth |
| 2,438 mm (an eight foot leaf) | Lean at least 20.7 degrees | 863 mm | Stands upright, 147 mm to spare | Only its own depth |
| 2,500 mm | Lean at least 24.2 degrees | 1,026 mm | Stands upright | Only its own depth |
| 2,600 mm | Lean at least 28.7 degrees | 1,250 mm | Lean at least 6.2 degrees | 279 mm |
| 2,800 mm | Lean at least 35.5 degrees | 1,625 mm | Lean at least 22.6 degrees | 1,076 mm |
| 3,000 mm | Lean at least 40.5 degrees | 1,950 mm | Lean at least 30.5 degrees | 1,522 mm |
Read the 2,500 mm row twice. In a high cube that sash stands up and asks the floor for nothing beyond its own depth, which for one of our series is 119 mm. In a standard box the same sash has to lie back at 24 degrees and asks for 1,026 mm of container length. The sash has not changed by one millimetre. The bill has changed by nearly a factor of nine.
This is also where nominal door sizes stop being a domestic detail and become a freight decision. A 34 entry door at the usual 80 inch height is about 2,032 mm tall and stands up in any container made. An 8 entry door is about 2,438 mm tall: taller than the door aperture of every standard container and taller than the internal height too, so it cannot stand up anywhere in a standard box, even after it is inside. In a high cube it stands with room to spare. Two products that differ by a foot of height have completely different freight.
The trade usually treats a high cube as a slightly taller version of the same thing, worth the small premium when you have bulky cargo. For glazed sashes the premium buys something more specific than 305 mm of headroom. It buys the difference between a load that stands and a load that lies down.
The honest counterweight: high cube equipment is not available everywhere on every service, and some inland routes and some destination depots handle it differently. Whether the box you want exists on the service you need is a question for your forwarder, not for us and not for this article.
Everyone who ships windows for the first time asks how much the container weighs. Almost nobody needs to. A container is full when it is full, and for windows that happens at a fraction of the mass limit.
Here is the arithmetic, and all of its assumptions are on the table. Take a sash 900 mm wide and 2,200 mm tall, glazed with 6 plus 12 plus 6 insulating glass. Glass mass runs at roughly 2.5 kg per square metre per millimetre of glass thickness, so twelve millimetres of glass is about 30 kg per square metre. Our sash is 1.98 square metres, so the glass alone is about 59 kg.
Now stand those sashes in a rack along the 5,898 mm length of a 20 foot box, separated by their sash thickness plus a 20 mm separator, two sashes across the 2,352 mm width. The box is effectively full. Count the glass.
| Series | Published glass sash thickness | Pitch with a 20 mm separator | Sashes in the box | Glass mass in the box | Share of the 28,360 kg payload |
|---|---|---|---|---|---|
| C113 outward opening | 57 mm | 77 mm | 152 | About 9,029 kg | About 32 percent |
| C120 outward opening | 62 mm | 82 mm | 142 | About 8,435 kg | About 30 percent |
| C116 outward opening | 81 mm | 101 mm | 116 | About 6,890 kg | About 24 percent |
| AW90 passive, inward opening | 90 mm | 110 mm | 106 | About 6,296 kg | About 22 percent |
We counted glass only. Add the aluminium, the hardware, the crate and the dunnage, and even if all of that together weighs as much as the glass again, the heaviest row above still sits under two thirds of the payload. The box is full and the scale is barely off its stop. To reach the payload in 33.2 cubic metres you would need an average density of about 854 kg per cubic metre, which is roughly the density of a container packed with nothing but glass and frame and no air at all.
The consequence inverts the usual instinct. Mass is nearly free in a window container; volume is the whole bill. Which means the number that decides your freight cost per unit is not how heavy the window is. It is how thick it is.
Look again at the table above. Between the thinnest sash and the thickest there is a 33 mm difference, and in the same container that is 46 sashes. Nobody specified AW90's 90 mm sash in order to lose 46 units of capacity. They specified it because a thicker sash carries a thicker insulating glass unit, and a thicker insulating glass unit is what the thermal target demanded.
That is not a defect in the thicker series. It is a trade you should see before you sign rather than after the booking is made, and it is a trade that only exists because the box is volume-limited. If the box were weight-limited, the thicker sash would cost you almost nothing.
There is a second number doing freight work under a different name. The frame wall width we publish for each series is the depth the frame occupies in the wall. In a container it becomes the depth the panel occupies on the floor when it stands, and it becomes the pitch between layers if anything is ever stacked. A 119 mm frame is a wall dimension on site and a floor dimension at sea.
Take the worked example again: 900 mm sashes, two across the width, standing at an 82 mm pitch along the length. Here is what the container has left over.
| Direction | Available | Used by whole units | Left over | What that leftover costs you |
|---|---|---|---|---|
| Length | 5,898 mm | 5,822 mm | 76 mm | Almost nothing. Fine pitch divides well. |
| Width | 2,352 mm | 1,800 mm | 552 mm | A corridor wider than half a sash, running the full 5.9 metres and the full height of the box. |
| Height, at the door | 2,280 mm | 2,200 mm | 80 mm | Small, and the only one anybody notices. |
| Height, behind the header | 2,393 mm | 2,200 mm | 193 mm | Usable only if whatever goes there can be tipped in. |
The internal width of a container is 2,352 mm, and that number divides cleanly by two and by three: 1,176 mm and 784 mm. Sash widths built to those numbers, or to a divisor of them, leave no corridor. Sash widths built to a round number that does not divide leave one, and you pay for that corridor down the entire length of the box and all the way to the roof. If you are at the stage where the elevation is still being set, this is the cheapest freight saving available to you, and it is free.
The arithmetic above begs a question. If volume is the whole bill, why not lay the sashes down and stack them, at a pitch equal to the frame wall width plus a separator? The answer is what you would get, followed by why you should not take it.
| Series | Published frame wall width | Layers in 2,393 mm | Layers limited to the 2,280 mm door | What the 113 mm gap is worth |
|---|---|---|---|---|
| C120 outward opening | 119 mm | 15 layers | 15 layers | Nothing at this pitch |
| C113 outward opening | 113 mm | 16 layers | 15 layers | One layer |
| C116 outward opening | 116 mm | 16 layers | 15 layers | One layer |
| AW90 passive | 90 mm | 19 layers | 18 layers | One layer |
Now the reason not to. A freight operator writing specifically about glazed doors and windows for sea freight says it without hedging: stand them vertically, and never lay them flat and stack them, because laying them flat puts uneven pressure on the glass and breaks it. The same practical guidance covers the rest of the job that actually prevents damage: hard corner protection, timber wedges and chocks so nothing can move, lashing at roughly 1.5 metre intervals, voids filled with dunnage or air bags so the load cannot shift, desiccant for a long voyage, and timber packaging that meets the phytosanitary rules of the destination.
So the tension is real and worth stating plainly: flat stacking is the only arrangement that uses the volume efficiently, and the people who pack glass for a living say do not do it. The resolution is the stillage or A frame, which lets panels stand while presenting a pitch set by the sash rather than the frame. And the stillage is itself a package, with its own height, its own footprint and its own cost, which is why the thing to specify is not the sash but the rack.
The IMO, ILO and UNECE Code of Practice for Packing of Cargo Transport Units is the reference every carrier guide defers to, and it is specific enough to design against. Summarised from published cargo securing guidance that is explicitly built on it:
| Rule | The published form | What it means for a window load |
|---|---|---|
| Vertical centre of gravity | Transversely centred and below half the height of the container | On a standard box that means below about 1,196 mm. Heavy glazed panels standing on the floor satisfy it naturally; anything stacked upward fights it. |
| Longitudinal centre of gravity | Within plus or minus 5 percent of the midpoint, exceptionally 10 percent | On a 20 foot box that is a band of about 295 mm either side of the midpoint. The load plan is a calculation, not a photograph. |
| Distribution along the length | A 60 to 40 split between halves is tolerated | Weighting the front half is allowed but published guidance warns it can push the drive axle of the destination truck over its limit. |
| Void space | One recognised industry manual, written on the Code, caps the total void in any horizontal direction at 15 cm | That 552 mm corridor is not a gap to leave. It is a gap that has to be filled or blocked. |
| Floor loading | Strength figures are surface load figures; concentrated loads have to be spread | Stillage feet and narrow bearers are the classic offender. Spread them on timber, and note that poorly cut dunnage is a known cause of dented container walls. |
| Who is responsible | Responsibility transfers to the packer | The factory that built the window can be blameless and the load still be non compliant. Name the packer in writing. |
Since an amendment to SOLAS Chapter VI, Regulation 2 came into force on 1 July 2016, a packed container cannot be loaded on board without a verified gross mass that has been provided to the master and the terminal in time for the stowage plan. Maritime authorities in several jurisdictions publish the same two routes to it: weigh the packed container on calibrated and certified equipment, or weigh every package, pallet and piece of dunnage and add the tare mass of the box. The obligation sits on the shipper, meaning the party named as shipper on the bill of lading, and the document must be signed by someone the shipper has authorised.
Two practical points follow. First, no verified mass means no loading, regardless of how good the crate is. Second, the second method depends on knowing the mass of every item including the dunnage, which means your packing list has to be good enough to be added up. For a project order of hundreds of sashes that is a real discipline, and it is easier if the factory can give you a per unit mass for each configuration rather than a total.
We can tell you what a given configuration weighs at the factory gate under a stated glass build. We cannot give you the verified gross mass of a container we do not pack, do not seal and do not book. That number belongs to whoever stuffs the box.
A project container is almost never one configuration. It is a mix, and the mix is where fill rate and site productivity pull in opposite directions. One window and door manufacturer writing about commercial and multi unit projects puts the resolution in a sentence worth copying into your instructions: ask for the loading plan and the packing list to be prepared together, and ask for the marks to be grouped by the same logic as the construction sequence. The same source adds that where a load mixes systems, keeping each system's units together inside the container is worth more on site than any marginal gain in fill rate.
In other words, the last thing loaded is the first thing out, and the first thing out should be what the first phase of the building needs. If nobody says so, the packer will optimise for volume, because that is the visible objective, and your site will spend labour hours hunting for one window behind forty others.
| Sequence decision | The wrong default | What to write instead |
|---|---|---|
| Grouping | Group by whatever loads tightest | Group by elevation or by phase, one system per block, marks visible without opening crates |
| Heaviest units | Wherever they fit | On the floor, spread on timber, near the midpoint of the length |
| Hardware and loose items | In with the sashes | Bundled separately, marked to the same unit numbers, last on and first off |
| Spare parts and touch up material | Sent later, as a separate shipment | Packed into the original container, listed on the packing list, so the freight question never comes up again |
A container of casement windows is rarely the only thing on the project. It is normal for the same box to carry the other openings of the building, and some of those are things we do not manufacture. Say so early, because they change the load in ways that have nothing to do with their size.
If you want a single supplier for every opening in the building, be sceptical of anyone who says yes immediately. Our published catalogue covers casement windows, sliding windows, tilt and turn windows and sliding doors. Several categories on the site, including swing doors, folding doors and entry doors, are pages without products. We would rather tell you that than let you build a loading plan around a product that does not exist.
| Item | What we can give you | Why there is a gap, if there is one |
|---|---|---|
| Maximum single panel size, sliding doors | Published: 2,500 by 3,000 mm for one series, 2,800 by 2,600 mm for another | No gap. Note that both exceed the standard door aperture height, so both have to travel leaning. |
| Maximum sash size, casement windows | Not published | It depends on the glass build and the hardware, neither of which is fixed until the order is. Ask per configuration, in writing. |
| Frame wall width | Published: 119, 113, 116 and 90 mm by series | It measures depth into the wall. It is not a crate dimension, though it becomes one at sea. |
| Glass sash thickness | Published: 62, 57, 81 and 90 mm by series | It is the build up of the sash, not the glass rebate. If you need the largest glass a sash will take, ask for the rebate on the section drawing. |
| Crate and stillage dimensions | Not published | They depend on your configuration mix and on the rack the packer uses. Ask for them per order, before the booking. |
| Verified gross mass | Per unit mass for a stated glass build, yes. The container total, no. | We do not stuff, seal or book the box. That number belongs to the packer and the shipper. |
| Packing declaration and customs paperwork | We can describe what is in the box and what the timber is | The declaration is signed by the packer or the shipper. Requirements are set by the destination authority, not by us. |
The pattern worth noticing: every number we publish measures something in a building. Every number that decides your freight measures something in a box. Someone has to do the translation, and if you do not say who, it will be done at the loading bay by whoever is holding the forklift.
| Line | Fill in | What happens if it is blank |
|---|---|---|
| Tallest single sash, mm | ____ | Nobody knows whether the load stands or leans until the truck arrives |
| Widest single sash, mm | ____ | The corridor down the side of the box is discovered after it is paid for |
| Glass build and resulting mass per unit, kg | ____ | Method two of the verified mass route becomes impossible and you are forced onto a weighbridge |
| Container type being booked | ____ | Standard or high cube changes whether tall sashes stand, and nobody checks until load day |
| Crate or stillage external dimensions, mm | ____ | The package is designed for the product and then does not fit the door |
| Orientation specified | ____ | Glazed panels get laid flat because it packs tighter |
| Who stuffs the container | ____ | The party the packing code makes responsible is never named |
| Loading order basis | ____ | Phase one units end up at the back of the box |
| Marking convention, visible without opening | ____ | Site labour spent opening crates to find one window |
| Timber treatment standard required at destination | ____ | Dunnage and stillages become the reason a consignment is held |
| Spare parts packed into the same container | ____ | Year five needs one roller and the freight argument starts again |
| Photographic record of the load before closing | ____ | Every later argument about damage becomes a matter of opinion |
| Ask | A useful answer contains | Sounds reasonable, carries no information |
|---|---|---|
| How tall is the tallest sash in this order | A number in millimetres, per configuration | Standard height |
| Will it fit the container | A statement against the specific box type booked | Yes, no problem |
| How is it packed | External crate dimensions, orientation, separating material, lashing spacing | Export standard packing |
| Who stuffs the box | A named party at a named place | Our logistics team handles it |
| Is the load secured to a standard | The name of the code and the blocking and lashing method | Firmly fixed, very safe |
| What is the verified mass route | Which method, who weighs, who signs, by what cut off | We will provide the weight |
| What order does it come out in | A marking scheme tied to your construction sequence | Loaded efficiently |
| Is the timber compliant | The treatment standard and the mark, checked against your destination | Fumigated, all certified |
| What is in the box, item by item | A packing list that matches the marks and adds up to the declared mass | As per invoice |
| Can I see it before the doors close | A date, a format, and a commitment to photograph before sealing | Of course, we will send photos |
Almost always assembled and glazed, because the factory glazing is the part you are buying. That is exactly why the box is volume limited rather than weight limited, and why the stillage matters more than the crate. If a supplier proposes shipping sashes and frames separately, ask what that does to the glazing, the sealing and the adjustment, because those are done in the factory for a reason.
For windows, usually yes, because you cube out and the 40 foot gives you roughly double the volume for less than double the cost. But it does not give you a taller door. If your sashes are over 2,280 mm, the comparison that matters is not 20 foot versus 40 foot, it is standard versus high cube, and that is a different question with a different answer.
We can tell you for your actual order, once the configurations, the glass build and the crate design are fixed. Any number given before that is a guess dressed as a specification, and it will be wrong in whichever direction is most expensive for you. The arithmetic in this article shows how large the spread is: the same box held 152 sashes at one sash thickness and 106 at another.
That depends on where in the chain it went wrong, and the packing code puts the packing itself on the packer. Practically, the useful question is not who is liable but who is first responder: one party who accepts the claim and then settles internally with its own subcontractors, rather than a list of contacts each of whom is somebody else's problem. Agree that in writing before the first container.
It can work and it usually complicates three things at once: the packing declaration, the naming of the packer, and the order in which things come out. A consolidated container also makes the verified mass harder, because the consolidator needs declared masses from every shipper. It is a legitimate way to fill a box, but it is a decision to make with your forwarder, not a default.
Before the doors close: the empty box, the load at roughly half done, the finished load showing blocking and lashing, and the closed doors with the seal number legible. On arrival: the seal before it is broken, the doors as they open, then any damage wide first and close second. The point is not aesthetics. It is that a claim about how something was packed can only be settled with a picture of how it was packed.
