Ramps & Gradients

The defining property of a ramp is its gradient - how much it rises for a given distance travelled, written as a ratio like 1:12 (one unit up for every twelve along) or as a percentage. Gradient is the number that decides whether a person can actually use the ramp, because pushing a manual wheelchair up a slope is hard work that grows quickly steeper the steeper the ramp, and coming down a steep ramp is a braking-and-balance problem that can end in a fall.

A widely cited maximum gradient for an accessible ramp is around 1:12 - and even that is demanding to push up unaided over any distance, which is why gentler is better wherever there is room. Some frameworks allow slightly steeper gradients only for very short rises, and prefer shallower slopes (such as 1:15 or 1:20) for longer ramps; the exact permitted gradients, and how they trade off against length, are set by your jurisdiction's code. The rule to carry is simple: steeper is harder and more dangerous, gentler is better, and there is a maximum beyond which a ramp is not accessible at all - so never let a ramp get steeper than the guidance allows, and prefer to go gentler. Treat 1:12 as a common ceiling to verify locally, not a target to design up to.

Gradient also interacts with the surface and the cross-fall. A ramp must be firm and slip-resistant in the wet - a steep ramp with a slick finish is a genuine danger - and its cross-fall (sideways tilt) must be minimal so a wheelchair is not pushed toward the edge as it climbs. And the steeper you make a ramp, the longer the handrails have to work and the more the edge protection matters, because the consequences of losing control grow with the slope. Gradient is the first decision, and every other part of the ramp is shaped by it.

A ramp at an acceptable gradient can still be unusable if it simply runs on and on. Climbing a slope is tiring, and a person in a manual chair, or someone with limited stamina, needs to stop and rest without rolling backwards. So the second law of ramps is that a single run is limited in how far, or how much rise, it may cover before a level landing is required. The ramp climbs a stage, arrives at a flat resting platform, then climbs the next stage.

The maximum rise or length per run, and the required landing size, are set by code and vary, but the principle is fixed: break a long climb into manageable flights separated by level landings, each landing long enough to stop, rest, and (crucially) for a wheelchair to sit entirely on the flat without any part of it on the slope. Landings also do the work of changing direction - where a straight ramp would be too long or would not fit the site, it doubles back in a switchback, and the landing at each turn must be large enough for a wheelchair to turn through it (recall the turning space from Module 1). Landings are equally required at the top and bottom of the whole ramp, level and clear, so a person can arrive, pause, and open a door or join a path without being stranded on the slope.

This is where the arithmetic becomes stark, and it is the bridge to the ramp-versus-lift question. Because gradient is capped, every extra unit of height demands a long multiple of that in ramp length, plus landings. At 1:12, a modest rise turns into a long ramp, and a full storey turns into an enormous zig-zagging structure with several landings - often far more than a site or a budget can absorb. The geometry does not negotiate: rise sets length, and length sets whether a ramp is sensible at all.

RISE -> RAMP LENGTH at 1:12 (sloped run only, before landings)
  150 mm step ......... about 1.8 m of ramp
  600 mm .............. about 7.2 m + a landing
  1500 mm ............. about 18 m + intermediate landings
  a full storey ....... a very long switchback - usually the wrong tool
(Illustrative at 1:12; add landings and verify limits in your code.)

A ramp carries people who may be pulling themselves up, steadying a wavering balance, or controlling a descent - so its handrails are not a trim item, they are structural to its use. An accessible ramp should have a handrail on both sides, continuous along the ramp and its landings, at a comfortable height for someone to grip and pull or steady against. A common approach provides two rail heights - one higher for standing users and one lower for children and some seated users - and rails that are easy to grip (a graspable profile, set far enough off the wall for a hand to close around).

What matters most, and is most often botched, is the ends. A handrail should be continuous and should extend horizontally beyond the top and bottom of the ramp so a person has something to hold before they start the slope and after they finish it - the moments of greatest instability. A rail that stops exactly at the point the slope begins lets go of the person precisely when they most need it. Rail ends should also turn back to a wall or down to the floor so a sleeve, a bag strap or a cane does not catch on them.

Then there is edge protection - what stops a wheel, a cane tip or a foot from running off the side of a ramp or landing where there is a drop. An open-sided ramp needs a raised upstand or kerb along its edges to catch a wheel before it leaves the surface, and often a guard or barrier above that where the drop is significant, so no one can roll or fall off the side. On a steep or elevated ramp this is life-safety, not a nicety. Add good colour and luminance contrast at the ramp edges and on the handrails so people with low vision can read where the ramp and its sides are, and the ramp becomes not just climbable but safe to climb - which is the whole point.

The hardest and most professional judgement in this lesson is knowing when not to build a ramp. A ramp is the right answer for small changes of level - a step or two, a low plinth, a modest terrace - where a gentle slope fits comfortably and gracefully. It becomes the wrong answer as the rise grows, for reasons that follow directly from the geometry above.

Because gradient is capped, height turns into length fast, and a ramp tall enough to climb a full storey is a vast, switchbacking ribbon of concrete: it consumes enormous floor area, dominates the architecture, still exhausts the person using it (a long climb at 1:12 is genuinely hard), and often cannot fit the site at all. Past a certain rise, a ramp stops being an act of inclusion and becomes an act of denial dressed up as one - technically present, practically punishing. At that point the honest, dignified and usually cheaper-in-space answer is a lift (Lesson 2.4): a passenger lift or, for smaller rises where a full lift is impractical, a platform lift.

Use the geometry as your decision tool. Estimate the rise; work out the ramp length it implies at the maximum gradient your code allows, plus the landings; then look at whether that ramp is genuinely comfortable to use and fits the building without dominating it. For small rises the ramp usually wins - it needs no power, never breaks down, and serves everyone continuously. As the rise climbs into the range of a storey, the lift usually wins. Sometimes the best answer is both, or a gentle ramp for the small level change plus a lift for the storey. The professional move is not to force one tool to do the other's job: a token ramp up a whole floor helps no one, and a lift for a single step is absurd. Match the tool to the rise, verify the gradients and limits locally, and get a person from one level to another in a way they can actually, safely use.