If you have gone looking for the head of an aluminium sliding window, you have probably met three different things wearing the same word. There is the head of the window itself - the top horizontal profile the factory cuts, drills and ships. There is the head flashing above it, a folded piece of metal or a self-adhesive membrane that belongs to the wall, not to the window. And there is the structural header or lintel that carries everything above the opening. They sit within a few centimetres of each other, they are made by three different trades, they are drawn on three different sheets, and a very large share of the leaks that get blamed on "the window" are actually failures in the few millimetres between the first and the second.
This piece is about the first one, because it is the only one of the three you can actually specify on a purchase order. But we are going to keep drawing the line between them, because that line is where the arguments start on site.
Before any dimension matters, get the three straight. They are not rival definitions of one component. They are three components that happen to occupy the same band of wall.
| Which one | Who makes it | What it must do | Where it is drawn | What it cannot do |
|---|---|---|---|---|
| Window head (top frame member) | The window factory | Close the top of the opening, guide the sash, seal the top line, retain the sash, and drain anything that gets in | Window shop drawings | Carry the wall above it, or bridge to the air barrier |
| Head flashing (lintel flashing) | Flashing fabricator or installer | Catch water running down the cavity above and send it out past the jambs | Wall section and flashing schedule | Air-seal anything, or fix a frame that is out of square |
| Structural header or lintel | Engineer and framer | Carry loads from above and keep deflection within limits | Structural drawings | Manage water, air or thermal bridging |
If a specification contains one line reading "aluminium head" and nothing else, all three rows have just become the installer's decision. That is not a knock on installers. It is a description of what happens when a document is silent.
This is the sentence that changes how you read the rest of this article.
In an outward-opening casement, the top of the sash closes against the head and then stays there. The head has to be a flat, straight, sealed surface that the sash compresses onto. It is a lid. It does one job, and it does that job while everything is stationary.
In a horizontal slider, the top of the sash never touches the head. It runs inside a channel in the head, a few millimetres clear of it, hundreds of times a year, while wind is pushing the sash sideways and the building is moving around it. The head is now a rail. It has to guide the sash continuously while it also seals, retains and drains.
Four jobs, one extrusion, and three of them pull against the fourth. That is the whole reason the head is where sliders get into trouble.
| Job | What the head does | How it fails | What to put in the spec |
|---|---|---|---|
| Guide | Keeps the sash vertical and resists racking as it travels | Sash drags at the top, corners wear, rollers start taking side load they were never rated for | Head channel depth and the material of the running surface |
| Seal | Closes the top line against air and driven rain | Whistling at the head under wind, water tracking down the inside face of the frame | Named seal type, plus whether the head-to-frame joint is sealed or only pressed together |
| Retain | Stops the sash being lifted out of the frame from outside | The lift-out technique: a locked window that still comes out of its frame | A head stop or anti-lift device, and the residual clearance in millimetres |
| Drain | Gives any water in the head channel a path back outside | The head becomes a gutter and empties into the wall cavity or onto the sash below | Location of the head weep, and where the discharge actually lands |
Notice which two are almost never on a quotation: retain and drain. Guide and seal appear in every brochure because they are visible. Retain and drain are invisible until the day they matter.
A sliding sash has to be removable. That is how the glazing gets replaced, how the rollers get changed, and how the installer gets the panel into the frame in the first place. So the head channel always leaves some vertical clearance: push the sash up, lift its bottom rail clear of the sill track, swing it out.
That clearance is a security feature running backwards. The method that security writers call lift-out works like this: from outside, push the sash up into the head channel, pull the bottom of the sash out of the sill track, and walk away with the whole panel. Every lock on the bottom rail sits downstream of that gap. Anti-lift blocks and head stops exist for exactly one reason - to fill the space the sash would otherwise travel into.
Now the part that makes this a specification problem rather than a hardware problem. The gap is a trade-off, and it is not a constant.
Risk is not uniform across a schedule either. A tall sash gives more panel to grip and more leverage to work with, so the head detail on a tall opening like a 36 x 72 sliding window deserves its own sentence in the specification rather than inheriting whatever was decided for the short ones.
Two things follow. First, ask the factory for the head stop, not a field improvisation. Screws driven into the head track are a well-known retrofit and they do work, but they are a modification: they have to clear the screen channel, they can stop the sash in the wrong place, and they are nobody's tested detail. Second, check the escape route. Any anti-lift device on a window that serves as a required emergency exit must still open without a key, without a tool and without special knowledge. A window that is very secure and very illegal is not a good outcome, and the answer is different for a ground-floor bedroom than for a stairwell landing.
Water reaches the head from above, not from the side. Wind-driven rain runs down the wall, finds its way behind the cladding, and the head flashing is what intercepts it. Published guidance differs by market on the numbers, and it is worth seeing the disagreement rather than memorising one set. One Australian installer guide describes a compliant head flashing as a back upturn of at least 20 mm tucked behind the sarking, a base spanning the opening plus 50 mm into each jamb, a front downturn of at least 15 mm forming a drip, and end dams of at least 20 mm. A North American continuing-education course on aluminium fenestration calls for an upturned leg of roughly 100 mm, counter flashing tied to the wall membrane, and the head flashing sealed to both the inner face of the window and the jamb flashings. Different codes, different digits, but the shape is the same in both: up at the back, over the top, down at the front, dammed at both ends. Specify the shape and then let your own code supply the numbers.
Where an opening is too tight for the flashing to run past the jambs, the same course notes that dual sealant joints can be used in place of a head flashing to catch water and turn it toward the jambs. That is a useful fallback on a retrofit, and a useful thing to know you should not accept as the default.
The sequence matters as much as the profile. Sill flashing first, jambs over it, window in, head flashing last lapping over the top of the frame and tucked under the membrane above. Every layer laps over the one below it. Reverse any two of those steps and you have built a funnel.
Then there is the window's own head, which is not the flashing. It is the thing the flashing drips onto, and it still has to get that water back outside. Two questions settle it. Does the head channel have a drained path at all, or is it a closed box? And where does the discharge land - outside the frame, into the jamb cavity, or down onto the sash below it? A head weep that empties into the sash channel has not solved anything; it has moved the problem to the bottom of the window.
This is also where the screen track earns a mention. On most sliding windows the insect screen runs in its own channel at the head, usually the outermost one. That channel collects dust, holds moisture, and is the one place a screen can blow out of in a gust. It is also the channel most likely to get accidentally sealed shut when somebody runs a generous bead of perimeter sealant across the top. We have written elsewhere about the seals along a sliding window track; the head is the member that holds all of them in line, and it is the one that gets sealed shut first.
Warm air rises and stratifies, so the head is the part of the window sitting in the warmest indoor air. It is also the part closest to the structural lintel and the reveal. Two different thermal bridges meet there: the path through the profile's own thermal break, and the installation junction - the reveal, the fixing, the gap, the sealant bead. Standards work treats the second one as a linear thermal transmittance, a property of the junction rather than of the window. It belongs to the wall design, and no frame brochure can quote it for you.
The practical consequence is that the head reveal is a classic location for condensation and for dark staining on the wall or ceiling above the opening. In a humid room it is worse, because moisture-laden air goes straight to the top of the opening and meets the coldest surface there first. If you are specifying for a wet room - and this comes up constantly with bathroom sliding door glass - the answer is ventilation and a warm reveal, not a deeper profile.
Being straight about the numbers here matters more than sounding good. ALPES publishes K values of 1.3 for the AW90 passive window, 1.6 for the C120 and 1.7 for the C116. Those are system values for the profile and glass combination, not whole-window values, and they are not NFRC or NAFS figures. EN and ISO boundary conditions and North American ones are not interchangeable, so a converted number is a calculation, never a test result. The sliding and side-press series publish no K value at all, and we would rather say that than invent one: ask for a calculated whole-window value against your exact opening size and glass build, and ask which standard it was calculated under.
One more directional point that is safe to act on. A dark external colour absorbs more than a light one, and aluminium expands roughly twice as much as steel for the same temperature change. A dark head on a long run therefore needs more room to move than a pale one, and the fixings have to permit it. That is a reason to specify slotted fixings, not a reason to avoid dark frames.
Three field errors account for most head-related call-backs, and none of them is a product defect.
The first is fixing through the head in a round hole, hard, at both ends. The frame can no longer expand along its length, the head channel can be pulled out of line, the sash begins to bind at the top, and the penetration itself is a water path. The head needs to be restrained laterally and left free vertically; slotted holes or a proprietary head anchor clip are the normal answers. The published flashing guidance makes the same argument for the flashing itself, calling for slotted holes once a run gets long - the metal has to be allowed to move whether it is part of the window or part of the wall.
The second is letting the structure bear on the window. A window is not a lintel. If the header above deflects down onto the frame - through creep, through a change in load, or simply because nobody left a gap - the head channel gets pinched and the sash stops moving freely. "It worked at handover and jams six months later" is usually this, and it is a structural coordination question, not a warranty claim.
The third is packing badly. Every fixing point needs a packer, and the packer has to be at the fixing rather than near it. Shim the head in the middle and tighten the ends and you have bent the head before anyone has opened the window once.
Width amplifies all three. The longer the head, the more thermal movement accumulates across it and the more the deflection question matters, so a wide, low opening such as a 60 x 36 sliding window should arrive with the head fixing centres, the fastener material and the movement provision written on the drawing rather than left to the crew.
Most of the numbers that look like head numbers are actually track numbers, and they are measured from the track rather than from the head channel. That distinction is worth keeping in front of you while you read the table below, which sets out what our own published data does and does not tell you about the top of the window.
| Series | Frame depth | Published figure that touches the head | Published head sealing detail | Still to ask |
|---|---|---|---|---|
| TS130 side press window | 130 mm | Track height 53/31 mm; sash width 75 mm; pulleys rated 200 kg | Four-sided continuous sealing strip with sealant at the upper rail | Which of 53 or 31 mm is the head guide depth, and what the residual lift-out clearance is |
| SL126/182 (96) | Two track 126 mm, three track 182 mm | High track 43 mm, low track 8 mm; single sash up to 2.5 x 3.0 m | Continuous four-side sealing strips plus sealant between the top track and the frame; patented eave-style waterproof design; bottom track water-blocking edge | Where the head channel itself discharges, and whether the eave detail is at the head as well as the sill |
| SL126/182 (75) | Two track 126 mm, three track 182 mm | High track 43 mm, low track 8 mm; 75 mm narrow interlock; 27A insulating glass with inert gas | Butyl rubber gasket at frame-to-sash joints; interchangeable drip-eave inserts | Whether the drip-eave ｉｎｓｅｒｔ changes at the head, and the head weep location |
| SL112/167 | Single or two track 112 mm, three track 167 mm | 45 mm high-ｄｒｏｐ bottom track with optional 85 mm water barrier; driving capacity up to 220 kg | Frame-enclosing-sash interlocked structure | Whether the high-ｄｒｏｐ drainage concept is mirrored at the head |
| SL155 | Four track 155 mm | 20-wheel self-balancing pulley rated 400 kg; single sash up to 2800 x 2600 mm | Passive check valve drainage; dual-matrix eight-layer sealing | How many channels the head carries, and which one is the screen |
| AW90 passive (for contrast) | 90 mm | K 1.3; 240-hour salt spray; 5,000 N shear | Dual-matrix eight-layer sealing; full injection adhesive process on frame, mullion, sash and screen | Nothing about lift-out - it is not a slider, and the head is a lid |
Two traps in that table are worth naming. First, frame depth is what has to fit into the wall; it is not the same as head channel depth, and the two are frequently confused on site. Second, frame depth is a property of the series, not a function of how many tracks it has: the four-track SL155 at 155 mm is shallower than the three-track SL126/182 at 182 mm. Anyone who assumes that more tracks means a deeper head will get the reveal wrong.
| Symptom | Most likely cause at the head | Defect, wear or installation | First action |
|---|---|---|---|
| Whistling at the top only when it blows | Head seal not seated, or the screen is not home in its head channel | Usually adjustment or wear | Check sash height and roller adjustment before ordering any part |
| Sash drags at the top, worse in the afternoon | Head channel pinched by thermal growth or by a rigid head fixing | Installation or movement, not hardware | Confirm the head fixings are in slotted holes before touching the rollers |
| Sash rattles and can be lifted in the frame | Rollers worn so the sash sits lower, or no head stop fitted | Wear, but it changes security | Re-adjust rollers, then fit a head stop and record the clearance |
| Water dripping from the inside top corner | Head channel has no discharge, or the head flashing was not dressed into the jambs | Could be either - test to find out | Run a water test at the head, not at the sill, and watch where it comes out |
| Dark staining on the ceiling or wall above the frame | Condensation at the head reveal, or a wet cavity behind | Building and installation, not the window | Look at ventilation, reveal insulation and the air barrier line |
| Screen keeps blowing out of the top | Debris in the head screen channel, or a screen specified for the wrong channel | Wear or specification mismatch | Clean the channel, then check the screen spec against the actual head section |
| Ask | A good answer looks like | Red flag |
|---|---|---|
| A section drawing through the head with the sash shown | A dimensioned head section, dated, with your opening size on it | "Standard profile" |
| Head channel depth and residual clearance above the sash, in mm | A number plus a named anti-lift provision | "It is secure" |
| How the head seals | Named seal type, plus whether the head-to-frame joint is sealed or only pressed | "Four-side sealing" with no detail |
| Where the head channel drains and where it discharges | A drawn path with the discharge marked outside the frame | "Weep holes" with no location |
| Head fixing method, centres and fastener material | Slotted holes or an anchor clip, stated centres, named stainless grade | "Fix through the frame" |
| Is the head a dedicated extrusion or shared with the jamb | One clear answer, plus where any site-formed drainage happens | "Same as the jamb" with no drainage answer |
| Deflection allowance above the head, and who designs it | A stated gap, a named responsible party, a soft joint detail | "We do not need one" |
| The screen channel at the head, and how the screen is retained | Shown on the head section, with the screen type named | "Screens available" |
| A calculated whole-window thermal value for my size and glass | A figure, your dimensions, the glass build and the standard named | A K value with no standard and no size |
| Coastal or exposed sites: coating and fastener grade at the head | Named coating standard and fastener grade, with test evidence offered | "Standard coating is fine" |
A fair amount of what this article asks you to request is not on our product pages, and it is more honest to say that than to imply otherwise. Most of it is not published because it depends on your opening, your glass and your wall rather than on the extrusion.
| We publish | You still need to ask | Why the gap exists |
|---|---|---|
| 6060-T66 profiles, 1.8 to 2.5 mm wall thickness, locally 3.0 mm in critical zones on the SL126/182 | The wall thickness of the head extrusion specifically | Reinforcement is placed where the load is, and the head is not always that place |
| PA66GF25 thermal break, 34 to 44 mm multi-chamber on the C116 | Where the break sits in the head section relative to your reveal | Isothermal alignment is a section property, and every series draws it differently |
| K 1.3 (AW90), 1.6 (C120), 1.7 (C116) | A whole-window calculated value for the sliding and side-press series, with the standard named | Those are system values, not whole-window values, and not NFRC figures; the sliding series publishes none |
| 240-hour salt spray and 5,000 N shear on the AW90 | Separate corrosion evidence for whichever sliding series you are buying | Test results belong to the series tested; they do not transfer, and a test result is not a coastal qualification |
| Sealant at the upper rail on the TS130; sealant between top track and frame on the SL126/182; 45 mm high-ｄｒｏｐ track and optional 85 mm barrier on the SL112/167; passive check valve drainage on the SL155 | Where the head channel itself drains on each of those series | Most published drainage language describes the sill, and the head needs its own sentence |
| CMECH hardware tested to 30,000 cycles | The roller rating and cycle evidence for the series you are ordering | A cycle count is a bench test on hardware, not a service life and not a roller life |
| Pulley ratings of 200 kg (TS130), 220 kg (SL112/167) and 400 kg with a 20-wheel system (SL155) | The assumed glass build behind the maximum panel size | Ratings are per sash and glass-dominated; a heavier glass build lowers the maximum panel |
| 100,000 square metres of production across Foshan and Zhaoqing | To see the line that will actually run your order | Total capacity is not the same as the capacity assigned to your job |
None of that is a refusal to answer. It is a list of things that only become true once your project has a size, a wall and a glass specification. The useful move is to turn every blank into a numbered line on the RFQ and then into a numbered line on the shop drawing. What is written on the order is what you can hold someone to; what is printed in a brochure is an opinion.
No, and confusing them is the single most common source of arguments on site. The window head is the top profile of the window unit and comes from the factory. The head flashing is a piece of metal or membrane above it that belongs to the wall and is usually made and fitted by a different trade. The flashing drips onto the window head; the window head then has to get that water back outside. Both have to be specified, and neither one does the other's job.
On many sliders, yes, if the head channel leaves enough vertical clearance for the bottom of the sash to clear the sill track. That clearance exists because the sash has to be removable for glazing and roller work. Anti-lift blocks, head stops or interlocking profiles reduce it. Two cautions: the clearance grows as rollers wear, so an older window is usually more vulnerable than a new one, and any anti-lift device on a designated emergency escape window must still open without a key or tool.
Restrained at the head, yes; locked at the head, no. The head is the member that runs the full width of the opening, so it is where thermal movement accumulates, and aluminium moves about twice as much as steel for the same temperature change. Slotted holes or a head anchor clip restrain the frame laterally while allowing it to grow, which a tight screw in a round hole does not. Always follow the system's own installation manual where one exists.
Because the head does a different job from the sill. The bottom rail sits on rollers and compresses onto a track; the top of the sash runs in a channel with a clearance gap and relies on a brush, fin or bulb seal rather than weight. If the sash has settled on worn rollers, the head clearance opens and the seal stops doing anything. Check the roller adjustment before ordering replacement seals.
Not by itself. Depth mostly tells you how much wall the frame needs and how many tracks it can carry. What governs heat flow through the head is the thermal break geometry, whether the isothermal lines line up, and - just as important - how the head junction is insulated and sealed once it is in the wall. A deep frame in a badly detailed reveal will still condense at the head. Note also that depth is a series property: our four-track SL155 at 155 mm is shallower than our three-track SL126/182 at 182 mm.
Less than most people fear, but not nothing. Keep the head channel and the screen channel clear of grit, do not let a perimeter sealant bead close the head weeps, and check once or twice a year that the sash still sits at the right height - because sash height at the head is also your lift-out clearance. Avoid high-pressure washing at the top of the frame and avoid aggressive or acidic cleaners on the finish.
Decide which of the three heads you are talking about, then specify each one separately. Treat the sliding window head as a rail rather than a lid, and give it all four jobs in writing: guide, seal, retain and drain. Ask for the head section drawing, ask for the clearance above the sash in millimetres, ask where the head channel discharges, and ask the head fixing to allow movement. Those four questions cost nothing at quotation stage and are close to impossible to answer once the cladding is on.
