Why a Steel Stair Feels Bouncy, and What Actually Fixes It

A stair can pass every strength check and still feel unsettling to walk on. Strength and stiffness are different properties, and only one of them is what you feel.

A stair can pass every strength check on the drawings and still feel wrong to walk on. That is not a contradiction. Strength and stiffness are different properties, and only one of them is the one you feel.

Strength is about failure, stiffness is about movement

Strength answers whether a member can carry a load without yielding or breaking. Stiffness answers how much it moves while carrying it.

The two are set by different things. Strength scales with the material’s yield strength and the section’s capacity. Stiffness scales with the material’s elastic modulus and the geometry of the section, and structural steels share essentially the same modulus regardless of grade.

That last point is the one worth internalising, because it rules out the intuitive fix. Specifying a higher-strength structural steel grade does not reduce deflection at all if the geometry is unchanged. A 350W section and a 300W section of identical shape deflect identically. The stiffness conversation is entirely about shape and span.

Depth does most of the work

If one thing changes how a stair feels, it is the depth of the stringer.

Bending stiffness rises very sharply with depth, far faster than it does with width or thickness. That is why adding material to the sides of a shallow stringer helps a little and making the stringer deeper helps a lot. It is also why a deep timber stringer and a slim steel one can perform similarly, and why a very slim steel stringer is a demanding piece of engineering rather than a default.

The design tension is obvious. The whole visual appeal of a modern steel stair is a structure that reads as a thin line, and thin lines are less stiff than deep ones. Resolving that is the actual design work: closing the section, using a hollow section rather than a plate, adding internal stiffeners that do not show, or accepting slightly more depth than the first sketch imagined.

Span is the other half. Deflection grows dramatically with span, so a stair that crosses a long opening without intermediate support is working much harder than one that lands on a mid-height landing. Adding a landing is often the cheapest stiffness intervention available, and it usually improves the stair in other ways too.

Open risers remove a hidden contributor

A closed-riser stair gets structural help from a component nobody thinks of as structural.

When risers are fixed between treads, tread and riser together form a folded assembly that resists movement and stiffens the whole flight. Removing them for a modern open-riser look removes that contribution, leaving the stringer to do everything alone.

This is a large part of why open-riser and floating designs need genuine engineering rather than a rule of thumb. They look lighter because they are lighter, structurally as well as visually. The code implications of open risers are separate again and are covered in open riser stairs under BC code.

The tread is often the thing that actually moves

Here is the diagnostic that saves the most time: work out whether the stair is moving or the tread is.

A stringer can be entirely rigid while individual treads flex underfoot, and the sensation reads to a user as a bouncy stair. It is a different problem with a different fix. Tread flex is about the tread’s own thickness, the distance it spans across the stair, and whether it has a folded edge or a stiffener under it.

The test is simple. Stand on the middle of a tread and then stand near the edge where it meets the stringer. If the movement largely disappears at the edge, the tread is the flexible element. If it persists, the structure is.

Tread flex gets worse as stairs get wider, because the tread spans further, which is one reason stair width has structural consequences beyond code compliance. The sizing logic is covered in floating stair tread thickness and the attachment side in mono stringer tread attachment methods.

Vibration is about what happens after the step

Deflection and vibration get used interchangeably and they are not the same experience.

Deflection is how far the stair moves under a load. Vibration is how it behaves once that load changes: whether the movement dies away immediately or continues for a moment afterwards. A stair that moves a small amount but keeps moving can feel considerably worse than one that moves further and stops at once.

Mass and damping are what settle movement down. This is part of why a stair with heavy timber treads or filled treads feels more solid than an identical structure with thin bare plate treads, and why concrete-filled pan treads feel notably settled underfoot. The added mass changes the behaviour, not just the sound.

Connections leak stiffness

A stringer analysed as fixed at both ends is stiffer than the same stringer sitting on connections with play in them, and real connections are rarely as rigid as a model assumes.

Bolt holes have clearance. Shims settle. A base plate bearing on an uneven surface finds its own position under load. Each of those introduces a small movement, and small movements at the ends of a member show up as larger movements in the middle.

The practical version is that a stair which feels worse than the calculations predicted usually has a connection problem rather than a member problem. Checking that the connections are properly tight, fully bearing, and grouted or shimmed as intended is the first thing to do before considering structural changes. The detailing considerations are covered in steel stair connection details and welded versus bolted stringer connections.

A mono stringer concentrates the whole question

A stair with two side stringers has two load paths. If one is slightly less stiff than intended, the other shares the work.

A mono stringer has one. Everything depends on that single beam, its depth, and the fixity at each end, and there is nowhere for load to redistribute. That does not make mono stringers bouncy; well-engineered ones are extremely solid. It does mean the section depth is the wrong place to economise, and that the connections carry more consequence than they would on a two-stringer stair.

Torsion enters as well. Because the treads cantilever from a central beam, an off-centre footstep twists the beam rather than simply bending it, so the section has to resist twisting too. Closed hollow sections handle that far better than open ones, which is why they are common on mono stringers.

Landings deserve the same attention as flights

Landings are frequently the weakest part of an otherwise solid stair, because they get designed last and framed lightly.

A landing is a small platform spanning between supports, carrying the same foot traffic as the flights. Where it is framed from lighter members than the stringers, it can move noticeably even when the flights do not, and users experience that as the stair being bouncy rather than the landing being bouncy.

Including landings in the stiffness conversation from the start avoids a stair where the flights are excellent and the pause between them is not. The structural approach is covered in steel stair landings and half landings and, for open designs, floating stair landing support.

Diagnosing a stair that already feels wrong

Where a stair is installed and unsatisfying, the useful thing is to narrow down what is moving before proposing a fix.

Start with the treads, using the test above: stand mid-tread, then near the stringer. If the movement largely disappears at the edge, the treads are flexing and the structure is fine. That is the cheapest problem to have, because a stiffener added under a tread or a thicker tread is a contained change.

Next, watch the stringer rather than feeling it. Have someone walk the stair while you sight along the stringer from the side at eye level. Visible movement along its length points at the member or the span. No visible movement, combined with a stair that still feels alive, points elsewhere.

Then check the ends. Look at the connections at the floor and at any landing while someone uses the stair. Movement at a connection, a base plate that lifts slightly, or a visible gap opening and closing, means stiffness is leaking there. This is worth checking first in practice, because it is the most common cause and the cheapest to correct: connections that were never fully tightened, shims that settled, or a base plate not bearing evenly.

Finally, consider the landing separately from the flights, since a moving landing reads as a moving stair.

Working through those four in order usually identifies the element responsible within a few minutes, and it turns a vague complaint into a specific question the engineer of record can answer.

Say what you want it to feel like

This is the single most useful thing a client can do, and it costs nothing.

Design briefs describe appearance in detail and behaviour not at all. A brief that says the stringer should read as thin, and also that any perceptible movement would be unacceptable, has named a genuine tension that the designer and engineer can resolve while the stair is still a drawing. The same brief delivered as appearance only gets resolved by whatever the engineer assumes.

Retrofitting stiffness after installation is possible but unsatisfying. It usually means adding something visible, and what gets added is rarely what anybody drew. Deciding it early is a section choice; deciding it late is a compromise.

---

Bounce is a geometry problem wearing a material costume. If a stair feels wrong, the useful questions are how deep the stringer is, how far it spans, whether the treads or the structure are moving, and whether the connections are doing what the drawing assumed.