How to Compute Fall Clearance and Why People Get It Short

Why this matters

Take every measurement in this procedure before anyone is exposed: from the ground, from inside an existing guardrail, or from a position where a stumble cannot carry you to an edge. A tape measure is not a reason to walk to the parapet.

Fall clearance is the arithmetic that decides whether a fall arrest system arrests a fall or simply documents one. It is the duty that is checked last and fails first, and it fails silently, because nothing about a correctly-worn harness clipped to a real anchor looks different when the answer is short. The single most common error is not a mistake in addition. It is measuring the wrong thing: taking the distance from the walking surface to the ground, when every term in the stack is measured from the anchorage, downward.

The one rule everything else hangs off

Measure from the anchorage, downward, to the nearest thing a body can strike.

Not from the deck. Not to grade. The anchorage is where the system's geometry begins, so it is where the ruler starts. And the lower level is whatever a body reaches first, which on a real building is a canopy, a lower roof, a condenser, a pipe rack, a stair landing, or a scaffold plank, and only sometimes the ground.

anchorage  ==============================
              |  free fall
              |  (D-ring travel before
              |   the absorber engages)
              -  arrest begins
              |  deceleration distance
              |  harness stretch plus
              |  D-ring shift
              -  D-ring comes to rest
              |  D-ring down to the
              |  worker's feet
              -  lowest point of the body
              |  safety margin
lower level ==============================

The six fields, and where each number legitimately comes from

1. Free fall. The distance the dorsal D-ring travels before the system begins to arrest, set by the lanyard or lifeline length and by where the anchorage sits relative to the D-ring. Capped at 6 feet for construction by 29 CFR 1926.502(d)(16)(iii), or lower if the device's own rating is lower. More than that is a compliance failure before it is a clearance failure.

2. Deceleration distance. The additional travel from the moment arrest begins to the moment the body stops. Construction caps it at 3.5 feet at 29 CFR 1926.502(d). Take the larger of that cap and the device label's own figure, because the cap is a legal ceiling and the label is a measured property of the thing on your back.

3. Harness stretch plus D-ring shift. Webbing elongates under arrest load and the dorsal D-ring rides upward on the wearer. Both come from your harness manufacturer's published allowance, which where published commonly sits on the order of a foot. If yours does not publish one, ask them; this is not a term to invent.

4. D-ring to the worker's feet. After arrest the worker hangs, and their boots are below their D-ring by roughly the height of their own back. Measure it on your own crew with the harness on; for a six-foot worker it lands near 5 feet. This term is inside the stack, not subtracted from it.

5. Safety margin. Manufacturers commonly publish 2 to 3 feet. Where your instructions give a range, take the larger, because this term absorbs the error in every other term and a margin rounded down was never there.

6. Anchorage height above the walking surface. The only term that gets subtracted, and the one you can change. It is a head start.

The published figure is a re-basing, not a term you add

Many self-retracting lifelines and some lanyard systems publish a required fall clearance in their instructions. That figure already contains free fall, arrest distance, harness effects and usually the margin, measured for a stated anchorage position and a stated user weight range. It is a re-based total, not an ingredient.

So: use the published figure instead of the built stack, never on top of any of its terms. Which one governs is not a preference. The manufacturer's number governs for that device in that configuration, because it was measured on that device. The built stack is what you use when you are combining components whose combination has no published figure, which is most real service rigs.

Two checks before you trust one. Read what it is measured to, because if it is measured to the D-ring you still owe field 4. Read what anchorage position it assumes, because an overhead figure does not describe a foot-level tie-off.

Worked example: one lanyard, three anchorage heights

Fixed facts held constant across all three cases: a 6-foot shock-absorbing lanyard, one worker at 250 pounds including tools (inside the under-310-pound basis the 1926.502(d) criteria are written for), a measured D-ring-to-boot distance of 5.0 feet, a harness manufacturer's published stretch and shift allowance of 1.0 foot, deceleration taken at the 3.5-foot cap, and a 3.0-foot margin taken from the top of the manufacturer's 2-to-3-foot range. Deck height 14.0 feet to the nearest strike surface, which here is grade.

Every qualifier from the section above gets its own line.

Case B: anchorage AT DECK LEVEL (0.0 ft above deck)
  free fall (5.0 ft to reach anchor + 6.0 ft lanyard) 11.0 ft
  deceleration distance (cap, device label no longer)  3.5 ft
  harness stretch + D-ring shift (mfr published)       1.0 ft
  D-ring down to boots (measured on this worker)       5.0 ft
  safety margin (top of published 2-3 ft range)        3.0 ft
                                                      -------
  required below the anchorage                        23.5 ft
  anchorage head start above deck                    - 0.0 ft
  REQUIRED BELOW THE DECK                             23.5 ft
  available                                           14.0 ft
  SHORT BY                                             9.5 ft

Case A: anchorage AT D-RING LEVEL (5.0 ft above deck)
  free fall (lanyard length, anchor level with D-ring) 6.0 ft
  deceleration distance                                3.5 ft
  harness stretch + D-ring shift                       1.0 ft
  D-ring down to boots                                 5.0 ft
  safety margin                                        3.0 ft
                                                      -------
  required below the anchorage                        18.5 ft
  anchorage head start above deck                    - 5.0 ft
  REQUIRED BELOW THE DECK                             13.5 ft
  available                                           14.0 ft
  CLEARS BY                                            0.5 ft

Case C: anchorage OVERHEAD (10.0 ft above deck)
  free fall (6.0 ft lanyard less 5.0 ft of height,
    so only the slack pays out)                        1.0 ft
  deceleration distance (cap; the absorber may not
    fully deploy at 1.0 ft of free fall, and using
    the cap is the conservative direction)             3.5 ft
  harness stretch + D-ring shift                       1.0 ft
  D-ring down to boots                                 5.0 ft
  safety margin                                        3.0 ft
                                                      -------
  required below the anchorage                        13.5 ft
  anchorage head start above deck                    -10.0 ft
  REQUIRED BELOW THE DECK                              3.5 ft
  available                                           14.0 ft
  CLEARS BY                                           10.5 ft

Case B also fails on free fall before it fails on clearance: 11.0 feet against the 6-foot cap at 29 CFR 1926.502(d). Note both failures rather than only the second, because the compliance failure is the one an inspector finds and the clearance failure is the one that kills.

The relationship worth carrying off this page. Required clearance below the deck went 23.5, 13.5, 3.5 feet for anchorages at 0, 5 and 10 feet. Every foot of anchorage height removed two feet of required clearance, because the height is counted twice: once as free fall that no longer happens, and once as head start you no longer need. That 2-for-1 holds while the lanyard still has slack left to take up, meaning while the anchorage sits below the D-ring plus the lanyard length, which here is 11 feet above the deck. Above that point free fall is already zero and each further foot buys one foot, not two. It also assumes everything held constant in this example stays constant: the same 6-foot lanyard, the same worker, the same absorber. Change the lanyard length and the break point moves with it.

Case A is the one to sit with. It clears by 0.5 feet, and that half foot exists only because the margin was taken at 3.0 rather than 2.0. A crew that took the bottom of the manufacturer's range and rounded the D-ring-to-boots distance down to 4.0 feet would have reported clearing by 2.5 feet on a rig that is actually within inches. Every rounding here runs one way for a reason: round the terms that add up, round the head start down, and never take a margin from the bottom of a published range.

Four ways this gets computed short

One: measuring from the walking surface. On a foot-level tie-off the surface and the anchorage coincide, which is why the habit survives, and it is wrong by the anchorage height on every other job.

Two: borrowing a published clearance from a different configuration. A published number is tied to an anchorage position and a user weight range, so reading an overhead figure off a label and applying it at a deck-level anchor imports the wrong free-fall term without anyone writing a number down.

Three: dropping the D-ring-to-boots term. It disappears when a published figure measured to the D-ring is treated as measured to the feet. On the Case A rig that omission is larger than the entire remaining margin.

Four: measuring to grade instead of the first strike surface. A canopy, a lower roof, a pipe rack or a parked lift changes the available number without changing the building, so the stack you computed at seven in the morning was computed against a yard that has since had a truck parked in it.

One more sits outside that list because it is a different geometry rather than a mis-added term: if the anchorage is not above the work, the fall arcs, and What Swing Fall Does That a Vertical Fall Does Not covers it. Offset invalidates a clearance calculation rather than adjusting it.

How to verify you got this right

Write the stack on the ticket with the six fields named, then run three checks.

• State the result as a signed number, "clears by 0.5 feet" or "short by 9.5 feet." A stack with no comparison at the bottom is a list.
• Confirm every term's source. Free fall from geometry, deceleration from the label or the cap, stretch and shift from the harness instructions, boots from a measurement on this worker, margin from the instructions. A term with no source was guessed.
• Ask what would flip the answer, and if it is a foot, raise the anchorage rather than shipping a rig with an inch of daylight.

If the stack does not close, the fix order is anchorage height, then device type, then eliminating the exposure. A shorter lanyard is not on that list by itself: it reduces free fall and does nothing about absorber travel, the worker's own height, or the margin, so a crew that swaps 6 feet for 4 and declares it solved has bought 2 feet on a stack that was short by nine.

References

• 29 CFR 1926.502(d), personal fall arrest system criteria for construction, including the 6-foot free fall limit, the 3.5-foot deceleration limit and the under-310-pound design basis.
• 29 CFR 1910.140, personal fall protection system criteria for general industry.
• Manufacturer instructions for the specific harness, lanyard or self-retracting lifeline in use, which own the stretch allowance, the arrest distance and any published required fall clearance.
• See related: What Swing Fall Does That a Vertical Fall Does Not, Why a Self-Retracting Lifeline and a Lanyard Need Different Clearance, What an Anchorage Has to Be Rated For and Who Decides.