Aluminum Sliding Doors: Specification & Procurement Guide for Architects

Aluminum Sliding Doors: How to Specify, Compare, and Procure

How to Choose Aluminum Sliding Doors That Won’t Stick, Leak, or Lose Heat

If your opening is 2 to 6 meters wide and you want big glass with minimal moving parts, aluminum sliding doors are the most practical choice — provided you can accept that at most 50–66% of the span will actually open. If you need the entire opening to clear, go back to folding doors. If you need the simplest, most thermally efficient solution for a standard-width opening, a sliding door beats a folding door on U-value per dollar spent almost every time.

That paragraph already eliminates half the wrong decisions architects make between sliding and folding systems. Here’s everything else — from someone who cuts the aluminum, not someone who resells it.

1. The Decision Variables — In the Order That Matters

Answer these eight questions before you open a single catalog and you’ll eliminate systems that were never right for your project.

1. How much of the opening must actually open?

This is the sliding door’s defining constraint.

If the brief says “must feel completely open” — you need folding doors. If it says “big glass, good ventilation, nothing complicated to operate” — sliders are the right call.

2. Opening width and panel count

The sweet spot is 2.4–4.8 m. Below 2.0 m, a single sliding panel suffices. Above 5.0 m, multi-track stacking adds weight and cost.

3. Which sliding mechanism?

What most spec sheets won’t tell you: there are three mechanically different systems, and the difference isn’t appearance — it’s weight capacity, operating effort, and service life.

The critical difference: standard sliders rely on brush seals the panel drags past — inherently leakier. Lift-and-slide panels lift off the gasket to move and press back down when closed. For heated spaces in cold climates, standard sliders will underperform regardless of thermal break quality, because the seal isn’t compression-based.

4. Thermal performance

Insist on whole-unit Uw tested to EN 14351-1 or NFRC 100, not the flattering center-of-glass number.

Achieving Uw ≤ 1.0 demands a 42 mm+ break with foam-filled chambers and warm-edge spacer. The 34 mm tier is realistic for 1.1–1.4, not sub-1.0. For Passive House (Uw ≤ 0.80), plan for 42 mm polyamide minimum.

5. Threshold type

• Raised track (8–12 mm): best weather seal, obvious trip hazard.

• Low/flush (3–5 mm): must be designed into the slab at construction stage.

• Fully flush (0 mm): requires a slot drain in the substructure, only works with lift-and-slide compression seals — standard sliders with flush thresholds will leak.

6. Glass weight cascade

A 1.8 m × 2.6 m panel has 4.7 m² of glass area. Weight dominates system selection:

At 200 kg+ you’re past what standard rollers handle. Lift-and-slide becomes mandatory — not an upgrade. If your supplier quotes standard sliding with triple glazing and doesn’t flag the weight issue, the rollers will be grinding within 18 months.

7. Multi-track stacking direction

A 3-track system can stack left, right, or both. Map it on the floor plan. The most common mistake: all panels stacking one direction, blocking the only garden access on that side.

8. Screen integration

If your supplier hasn’t asked whether the project needs insect screens, that’s a warning sign. In tropical and subtropical markets, screens are a basic habitability requirement. The sliding door must accommodate a retractable or sliding screen on the exterior track. Many European-sourced systems weren’t designed for screen integration — retrofitting one loses 30–40 mm of clear opening width.

2. What Only a Factory Can Tell You

The track isn’t just a track. It carries weight (structural), drains water (hygiene), and guides the panel (alignment). Cheap tracks do the first, poorly do the second, and fail at the third over time. Specify 6063-T6 with a hard-anodized running surface — bare aluminum tracks groove within 2–3 years because stainless steel rollers are harder than the aluminum beneath them. Drainage slots: minimum every 600 mm along the outer edge. Track depth: 45–60 mm for standard, 65–85 mm for lift-and-slide.

Roller quality is invisible until it fails. A 200 kg panel rides on four to six bearings. Check: sealed stainless steel bearings (60,000+ cycles) or unsealed carbon steel? Glass-reinforced nylon wheels or bare nylon? Stainless axle ≥8 mm or zinc-plated 6 mm? ±3 mm height adjustment or fixed? We’ve replaced hardware on sub-5-year-old systems where “stainless steel hardware” meant the handle — not the bearings inside the roller.

The interlock is your thermal weak point. Where two sliding panels meet, they seal against each other — not the frame. A good interlock needs dual brush seals on both meeting stiles, a thermal break in the interlock profile itself, and at least 20 mm overlap. If you feel a draft at the center meeting point on a cold day, the interlock design failed — not the installation.

Most leaks aren’t seal failures — they’re drainage failures. This may sound counterintuitive, but a sliding door’s outer track functions as a gutter: catching runoff from the glass, channeling it sideways, ejecting it through drainage slots. If slots are too small (<8 mm × 20 mm), too few (<1 per 600 mm), or positioned inside the gasket line instead of outside, water backs up and crosses to the interior. We pressure-test every track assembly per ASTM E331 and disassemble one unit per batch to inspect for water in the interior chamber. That’s the difference between passing a lab test on day one and staying dry on day 1,000.

Frame depth vs. glass area — the trade-off nobody admits. To hit Uw 1.2 W/m²K, you need ≥85 mm frame depth (space for a 34 mm break plus foam chambers). An 85 mm frame on a 2.4 m door means 170 mm of frame (top + bottom) — about 7% of visible area. Architects want slim frames (50–60 mm) and low U-values (sub-1.0). Those are physically contradictory: you can’t fit a 34 mm thermal break into a 50 mm frame. Choose your priority. If thermal performance wins, accept a deeper frame and use a dark color that visually recedes. If slim sightlines are non-negotiable, accept Uw around 1.6–1.8 and compensate with wall insulation elsewhere.

Why we cap single panels at 2,000 mm wide. Can we extrude wider? Technically yes. A 2.4 m × 2.6 m triple-glazed panel weighs ~280 kg — even lift-and-slide bearings are at their limit. More critically, the panel sags under its own weight: 4–6 mm deflection at the leading edge over a 2.4 m span, enough to pull the top corner away from the head gasket and create an air leak. We hold the line at 2,000 mm × 3,000 mm. Beyond that, use a multi-panel system — it costs more but actually works for the 20-year life a commercial project expects.

3. Three Tiers — What to Specify at Each Budget Level

4. Six Procurement Mistakes That Lead to Warranty Claims

The six issues we see most often — all traceable to the procurement stage, not the factory.

1. Buying by frame cost, not system cost. A standard frame at $280/m vs lift-and-slide at $420/m looks like a $140 gap. But the standard door needs $180 in additional on-site flashing because its simpler drainage requires more field waterproofing. Installed cost difference: -$40. Always compare installed system cost.

2. Specifying triple glazing without checking panel weight. Architect specs triple for thermal performance, supplier quotes a standard system, nobody runs the weight calc. Panels arrive at 210 kg — rollers rated for 150 kg. Replacing the entire system: $8,000. The weight check takes 5 minutes.

3. Flush thresholds on exposed elevations. Sliding doors are already more leak-prone than hinged doors. Flush threshold on a windward face with 800 mm annual rainfall guarantees water on the interior floor. Flush works on lift-and-slide with compression gaskets AND a covered exterior (≥2 m overhang). Everywhere else: raised or low-profile.

4. Not specifying screens at order stage. Retrofitting a screen after installation means surface-mounting on the exterior face — it protrudes 30–40 mm and looks bolted-on. Specifying at order lets the factory extrude an integrated screen track that sits flush. Cost difference: ~$60/m. Aesthetic difference: invisible vs. obvious.

5. Assuming all tracks drain equally. A slider on concrete has natural drainage. A slider recessed into a timber-framed floor needs a connected drainage path or the subfloor rots. If your supplier doesn’t ask about floor construction before specifying track type, they’ve never handled a timber-frame project.

6. Comparing quotes without a hardware breakdown. $3,600 vs $3,100 — cheaper wins. The $3,100 door used zinc handles (pitting in 2–3 years coastal), standard EPDM (shrinks year 3), unsealed bearings (seize year 5). Real 5-year cost after two hardware replacements: ~$4,800. Always get a hardware schedule with material grades per component.

5. When Sliding Doors Are the Wrong Choice

There are projects where we tell clients not to buy sliding doors.

• You need 90%+ clear opening. Sliding panels permanently occupy 33–50% of the width. No sliding system matches a fully stacked folding door. Buy folding.

• Opening under 1.5 m. A 1.2 m slider gives a 600 mm clear opening — the width of a standard hinged door at 2–3× the cost with worse air sealing. Use a hinged door.

• High-security ground floor. Sliding doors lock at the interlock — a single vertical line. Multi-point locks engage the meeting stile but can’t engage head and sill the way a hinged door’s shoot bolts can. If PAS 24 or equivalent certification is required, demand test data on the complete assembly, not just the lock cylinder.

• Installed budget under $500/linear meter. You’re getting T5 aluminum, 14 mm break, brush seals, unsealed bearings. Functional for 3–4 years in a mild climate. For heated spaces in cold climates: drafts, frame condensation, visible track wear. Consider reducing the opening, switching to fixed glass + a smaller operable panel, or using a hinged door for access.

6. Our Workflow — What Happens When You Send an Inquiry

Before we quote, we check these nine things. If your supplier asks fewer than six, you’re talking to a trading company, not a factory.

1. Opening dimensions — width × height, site-measured or from drawings

2. Floor construction — concrete, timber, or steel deck (determines threshold and drainage)

3. Climate data — ASHRAE zone or local equivalent

4. Opening ratio — what percentage must be operable?

5. Stacking direction — left, right, both? Traffic flow accounted for?

6. Glass specification — double/triple, Low-E position, laminated for impact/acoustic

7. Screen requirement — retractable, fixed, or none; integrated or surface-mounted

8. Wind load — design pressure per local code

9. Accessibility — barrier-free threshold required?

Procurement Timeline

Frequently Asked Questions (FAQ)

Lift-and-slide vs standard — worth the upgrade?

For any heated space in a cold climate, yes. The 0.2–0.3 W/m²K improvement from compression seals alone typically pays back the 40–60% cost premium within 5–7 years of heating bills. For mild climates, a standard slider with a 24 mm break and double glazing is adequate.

Maximum single panel width?

2,000 mm for lift-and-slide, 1,500 mm for standard roller. Beyond these, panel deflection compromises the top seal. For openings over 4 m, use multi-panel systems rather than wider individual panels.

What U-value to target?

Uw ≤ 1.4 for cold climates (ASHRAE 5–7), ≤ 1.8 temperate, ≤ 2.2 mild. Standard sliders run 0.3–0.5 higher at every tier due to non-compression seals. Passive House (≤ 0.80) requires 42 mm+ break, foam chambers, triple glazing, thermally broken sill.

Can I eliminate the visible bottom track?

Lift-and-slide can go fully flush (0 mm) with a slot drain integrated at the slab design stage — not retrofittable. Standard sliders need a minimum 3–5 mm raised track because it’s structural.

Hurricane zone performance?

Impact-rated sliders exist, tested to Miami-Dade protocols: laminated glass (6 mm tempered + 1.52 mm PVB + 6 mm tempered), reinforced frames, stainless interlock hooks. Add 40–60% to system cost and 25–35% to panel weight. Budget from day one — retrofitting impact glass into a standard frame isn’t feasible.

Factory-direct vs distributor?

Factory offers customization (50 mm panel increments, custom RAL with 25-unit minimum) and 15–25% savings on 50+ unit orders. Distributors are faster on standard sizes but offer zero flexibility. For a single replacement door, go local. For a project with volume, go direct.