Installing a Standing Seam Metal Roof

Start With the Number the Panel Will Move

Every other decision on a standing seam roof follows from one figure: how far the panel travels between its coldest and its hottest state. That figure comes from three things — the coefficient of expansion of the metal, the length of the run, and the temperature swing the panel itself sees. Panel temperature is not air temperature. A dark, low-reflectance coating on a clear still day sits well above ambient, and the same panel radiates below ambient on a clear winter night. Take the design swing from the manufacturer's published guidance combined with local climate data, never from the forecast for the day you happen to be installing.

Comparative movement across the common panel metals is worth carrying in your head, because it explains why details that work on steel tear aluminium apart:

Read the bottom row before dismissing it. A 100 ft aluminium run moving better than an inch and a half means the free end of that panel finishes August a long way from where it sat in February, and it repeats that every year for the life of the roof. Steel moves roughly half as far, which is why long-run steel gets away with terminations aluminium cannot survive. Nothing in the assembly stops the movement. The only real choice available on site is where that travel is allowed to go and what it rubs against on the way.

Fixing the Panel: Where the Movement Goes

Panels are pinned along one line and free everywhere else. That pinned line — the fixed point — sets the direction of travel and divides the total movement between the two ends of the run. Fix at the eave and the whole expansion pushes up toward the ridge. Fix at the ridge and it all arrives at the eave, loading the eave hem and whatever the gutter is doing. Fix at mid-run and each direction takes half, halving the demand on both terminations, at the cost of having to detail both ends to float.

Choose deliberately, then mark it on the deck and on the drawing. Steep pitches and long runs generally argue for fixing high, so that panel weight is not hanging on sliding clips and creeping downslope through years of cycling. Heavy snow retention, mechanically attached snow guards, or a curb sitting near one end all change that argument and are worth resolving before the first panel is lifted.

Whichever line gets chosen, the fixed clips have to be installed as fixed clips, in the correct rows, and every hand on the roof must know which rows those are. Fixed clips scattered by accident through a floating run leave a panel restrained in two places, and a panel restrained in two places buckles between them the first hot afternoon.

Run Length Decides the Movement Budget

Run length is the multiplier on everything above, so the takeoff is a movement decision at least as much as a quantity decision. A 140 ft slope broken into two runs with a transition detail moves half as far at each end as one continuous panel would. That transition costs a flashing, a second set of terminations and a slower sequence; it buys back travel that the eave, the ridge and every penetration in between would otherwise have to absorb. Where the geometry allows either approach, weigh both against the detailing they force downstream.

Measure the coverage width the profile actually delivers once seamed, not the nominal sheet width, and set the panel module out from a chalked reference so the last panel at the rake lands full width rather than ripped to a sliver. Hips, valleys and dormers consume panels in irregular offcuts, and a panel schedule listing every run by length and count, with the reusable offcuts flagged, is what stops a coil order arriving short two days before the crane leaves.

Handling sets its own limit. Above a certain length panels need spreader bars, more hands and a genuine wind window; a long panel bent on the way up the roof is scrap, and the replacement arrives on the next coil run, not the next morning.

Clips Are the Movement Hardware

Two kinds of clip do two entirely different jobs. A fixed clip locks the panel to the structure along the fixed line and carries the load that would otherwise slide the roof downslope. A floating two-piece clip has a base fastened to deck or purlin and a tab that slides within a defined travel range as the seam moves through it. Confusing the two on site is the most common way a correctly designed roof ends up failing.

Set the sliding tab for the temperature at installation rather than centring it out of habit. Installing on a hot deck in July with the tab centred leaves only half the travel for the contraction that follows in January; the tab bottoms out, the panel then pulls against the clip base, and the fastener begins working itself loose. Manufacturers publish an offset chart for precisely this. Follow it, and write down the deck temperature you set against.

Spacing comes from uplift, not from habit either. Clip rows tighten at eaves, rakes, ridges and building corners where the design pressures run highest, and the spacing that goes with a tested assembly comes from the uplift test report for that assembly rather than from interpolation between two numbers on a data sheet.

Seaming: Profile Rules and the Order of the Passes

Snap-lock and mechanically seamed profiles handle movement in different ways, and swapping their rules gets expensive. A mechanically seamed panel is closed with a powered seamer travelling the length of the run, folding the seam over the captured clip tab so that all travel happens at the tab. Snap-lock engages under hand pressure and depends wholly on the clip's slide, so it tolerates a fixed clip nowhere it was not designed for one.

Run a test seam on an offcut every morning and check roller adjustment against the profile before the seamer touches finished work. Seam height and fold tightness are the geometry the assembly's published water and air performance was established on — an under-formed seam looks closed from the walkway and leaks at the first driving rain out of the wrong quarter.

Time the seaming and record the condition. Panels laid and seamed cold sit at one end of their range, which is fine in itself, but everything downstream — tab position, trim lap allowance, ridge closure clearance — has to reflect that the panel will only grow from there.

1. Confirm the fixed rows are installed as fixed clips and every other row is floating, before any panel is closed.
2. Set floating tabs to the manufacturer's offset for the measured deck temperature and log that temperature.
3. Engage the panel, check module at eave and ridge, and hand-crimp the first and last few feet.
4. Run the test seam on an offcut, check height and fold, then seam the full run in one continuous pass.
5. Scribe a reference mark at the free end of the panel so later travel can be read against something real.

The Deck Interface: Underlayment, Slip and Noise

Underneath the panel, movement shows up as abrasion and as noise. High-temperature self-adhered underlayment is standard beneath metal because ordinary membranes soften at panel-cavity temperatures; ASTM D1970, Standard Specification for Self-Adhering Polymer Modified Bituminous Sheet Materials Used as Steep Roofing Underlayment for Ice Dam Protection, covers the general class, while metal-roof service temperature ratings are published separately by each manufacturer. Verify the rating rather than the category.

A slip sheet, or a slip-surfaced underlayment, earns its place on any long run. Without one, the panel underside drags across an adhesive-faced membrane twice a day, polishing the back-face coating away and scuffing the sheet below. The ticking and pinging that owners phone about after handover is frequently nothing more than this: a panel gripping, releasing and jumping instead of sliding cleanly.

Substrate flatness carries the other half of the complaint. Oil canning is aggravated by movement across an irregular plane, and neither striations nor pencil ribs rescue a panel laid over dished sheathing or a purlin line out by a quarter of an inch.

Penetrations, Curbs and Anything Bolted Through

Anything fastened through a panel into the structure creates a second pinned point. There is no clever workaround for that, only discipline. Small round penetrations take a boot fastened to the panel alone, so the boot travels with the panel while the pipe passes through with clearance. Where the pipe itself cannot move, the hole through the panel must be oversized and the seal has to accommodate the relative travel rather than resist it.

Curbs are the harder case because they are wide. A curb spanning from one panel across a seam onto the next ties both panels together and stops them both. Detail curbs to sit within a single pan wherever the equipment footprint allows, and where it genuinely cannot, use the manufacturer's tested floating curb with an upslope diverter and a back pan. Field-fabricated curbs screwed through seams are a leak with a date on it.

Position matters as much as construction. Clusters of penetrations belong away from the far end of a long run where accumulated travel is greatest, and no penetration should be set immediately upslope of a fixed row without first working out what the panel does around it through a full cycle.

Trim That Has to Float

Trim moves too, and most of it gets fastened by people thinking about weather rather than travel. Eave detailing follows from the fixed point: a hem hooked over a continuous cleat holds the panel down while letting it slide, provided the hem is formed to the profile's published dimension and the cleat runs straight and unbroken. Ridge and hip closures must close the profile without pinning it — inside closures bonded to the panel, outside closures fastened to the trim.

Long trim runs expand on their own account, independent of the panels beneath them. Butt joints in fascia, rake and ridge cap need a slip joint or a backer plate behind the joint, sealed with a non-hardening sealant that will extend and compress rather than a bead that cures brittle and splits at the first hard frost. Fasten one side of the joint and leave the other free to slide.

Valleys collect both problems at once. A valley pan takes water off two planes and sits between panels expanding toward it at different rates, so the panel ends need measured clearance to the centre line, a hemmed edge engaging a cleat instead of a screw, and enough pan width that offset ends never bridge the flow line at their furthest travel.

Uplift, Fasteners and the Loads That Interrupt Movement

Uplift resistance and free movement pull against one another, and the clip is the compromise between them. Clip base fasteners have to reach real structure: sheathing thickness, purlin gauge and fastener type together determine withdrawal capacity, and tested values from uplift testing to UL 580, Tests for Uplift Resistance of Roof Assemblies, or UL 1897, Uplift Tests for Roof Covering Systems, or structural performance to ASTM E1592, Standard Test Method for Structural Performance of Sheet Metal Roof and Siding Systems by Uniform Static Air Pressure Difference, apply only to the substrate the assembly was tested over. A clip screwed into a nailer never tested with that clip is an unverified assembly, whatever the panel literature promises.

Design pressures themselves come from the governing load standard — ASCE 7, Minimum Design Loads and Associated Criteria for Buildings and Other Structures — as adopted and sometimes amended by the local building code, so the applicable edition and any local wind provisions vary by jurisdiction. Edge and corner zones carry far more than the field, and clip rows tighten there to the tested spacing rather than to a proportion someone works out on the roof.

Dissimilar-metal contact fails slowly but reliably. Aluminium panels bearing on uncoated steel clips, copper flashing draining onto aluminium-zinc coated steel, treated timber touching the panel underside — each corrodes at exactly the contact point that is also under constant micro-movement, which strips whatever passive layer might otherwise have formed.

The First Thermal Cycle, and What to Hand Over

Walk the roof twice: once at completion, and again after the first full thermal cycle — a hot afternoon following a cold night, or the reverse. Anything visible on the second walk that was not there on the first is movement finding something it does not like.

Look for clip tabs sitting hard against the end of their travel, panel ends that have crept past the reference you scribed at the eave, seams beginning to open at the far end of a long run, and sealant at trim laps stretched thin or split. Check every pipe boot for a pipe now sitting off centre in its hole. Photograph the fixed rows and mark them on the as-built, because the next contractor to open this roof has no other way of knowing where it is pinned.

Hand the movement data over with the warranty documentation: panel metal and coefficient, run lengths, fixed-point location for each slope, clip type and count, and the deck temperature the tabs were set at. Any future penetration cut into this roof is either detailed against that record or it is guesswork with a drill in its hand.