Walk enough jobsites and you start seeing the same picture. A crew is working at height, every one of them in a full body harness, lanyards hanging. Nothing is connected to anything. It is easy to read that as a discipline problem. Usually it is not.
The harnesses got bought, the training got done, and nobody ever answered the one question that makes the gear work: what, specifically, does a person connect to in this work area, at this stage of the job? That question is the anchorage plan. It is the part of fall protection that gets skipped most often, and it is the part the standard is most specific about.
The connecting device is the part that gets inspected every morning. What it hooks to usually got decided months earlier, by somebody who never came to the site. What 1926.502(d)(15) actually says It is one sentence, with an "or" sitting in the middle of it: "Anchorages used for attachment of personal fall arrest equipment shall be independent of any anchorage being used to support or suspend platforms and capable of supporting at least 5,000 pounds (22.2 kN) per employee attached, or shall be designed, installed, and used as follows: (i) as part of a complete personal fall arrest system which maintains a safety factor of at least two; and (ii) under the supervision of a qualified person."
So there are two ways to comply, not one rule and an exception. Either the anchor holds 5,000 pounds per person tied to it, or the whole system is designed, installed and used as an engineered system. The second path is both of those subparagraphs, not either one. Safety factor of two and a qualified person, joined by "and." Two things still get lost in the field. Per employee attached.
Two people on the same beam clamp is a 10,000 pound question, not a 5,000 pound one. Crews rig that way constantly, because the anchor got identified for one person and then three people needed it. The engineered path is not a loophole for "it looks strong enough." It says designed and installed, not just used, and it names a qualified person. Subpart M does not define that term.
The definition is in 1926.32(m): "Qualified means one who, by possession of a recognized degree, certificate, or professional standing, or who by extensive knowledge, training, and experience, has successfully demonstrated his ability to solve or resolve problems relating to the subject matter, the work, or the project." Degree, certificate, professional standing, or demonstrated ability. Not seniority.
If nobody on the project fits that and can produce the calculation, the project is on the 5,000 pound path whether it wants to be or not. The numbers that decide whether an anchor works Anchorage strength is only half of it. The rest of 1926.502(d) sets what the system has to do when it catches somebody, and those numbers are what make one anchor usable and another useless.
A personal fall arrest system has to limit the maximum arresting force on an employee to 1,800 pounds when used with a body harness. And under (d)(16)(iii) it has to "be rigged such that an employee can neither free fall more than 6 feet (1.8 m), nor contact any lower level." Two prongs, joined by "nor." Clearing the free fall number is only half of that sentence.
Read it against a real anchor point. An anchor at foot level with a six foot lanyard is not a six foot fall. 1926.500(b) defines free fall distance as the vertical displacement of the attachment point on the harness, from the onset of the fall to just before the system begins to apply force.
Under (d)(17) that attachment point sits in the center of the wearer's back near shoulder level, or above the head. So the D-ring travels the lanyard length plus however far it sits above the anchor, which on most people puts you past ten feet before the energy absorber has done anything. Overhead, that same lanyard can be rigged inside the six foot limit. At the feet it cannot.
Nothing about the hardware changed. Only the geometry did. There is also a second "two" in the standard, and it is not the same two. (d)(16)(v) requires the system to "have sufficient strength to withstand twice the potential impact energy of an employee free falling a distance of 6 feet (1.8 m), or the free fall distance permitted by the system, whichever is less."
That applies no matter which compliance path you are on. The safety factor of two in (d)(15)(i) is a condition on the engineered anchorage path only. One more number worth keeping straight. The 6 feet in 1926.501(b)(1) is the height that triggers fall protection in construction. The 6 feet in 1926.502(d)(16)(iii) is a limit on free fall distance. Same number, measuring two different things.
Where the plan usually runs out In our experience the written fall protection plan is rarely the problem. The problem is that the plan describes the building as designed, and the crew is working on the building as it exists this week. The patterns repeat. The anchorage disappeared with the sequence. The steel that was going to be the tie-off point got decked over.
The overhead structure the plan assumed is now below the work. The anchor was never rated, just chosen. Conduit, a sprinkler line, a guardrail post, a roof vent. The guardrail post is the easy one, because the standard names it. 1926.502(d)(23) says personal fall arrest systems shall not be attached to guardrail systems. The rest are not named anywhere, which is the problem.
Nothing blesses them, and none of them will carry 5,000 pounds or survive the engineered path, so whoever hooked onto one has to explain which of the two compliance paths they were on. The window between guardrail and tie-off. Leading edge work, decking, guardrail installation itself.
The gap where permanent protection is not in place yet and the temporary system has to carry the work is exactly where planning is thinnest. Nobody owns the answer. The crew finds out the plan does not fit at the moment they need it, eight feet up, with a schedule behind them. Worst possible time, worst possible person to solve an engineering question.
Planning it before the crew is up there Plan by work area and by phase, not by building. Answer it for this elevation, this week, this trade. A single site-wide answer is usually wrong by the second phase. Write down what the connection actually is. Not "tie off at height."
The specific anchor, its rating or the basis for it, how many people it carries, and the connecting device it was planned around. Check the clearance math, not just the strength. Free fall distance, deceleration distance, which (d)(16)(iv) caps at 3.5 feet, lifeline and lanyard elongation, which the definitions in 1926.500(b) leave out of both of those numbers so you have to add it yourself, the worker's height below the D-ring, and the distance to the next level down.
Name who decides when the plan does not fit. Someone with authority to stop the task and get an answer, reachable during the shift. If that person is not named and not reachable, the crew will improvise, and improvising is how conduit becomes an anchor point. Re-walk it when the sequence changes. Anchorage is the first casualty of a resequence, and resequences do not usually get routed through safety.
If it helps to run this with a crew rather than read it, our working at heights material and the toolbox talk library cover the same ground in a format you can hand a foreman. The question worth asking on your own project Pick a crew working at height tomorrow and ask one person what they are supposed to connect to and who told them.
If the answer comes back fast and specific, the planning happened. If it is a shrug or a point at something nearby, the gear on that crew is doing less than everyone assumes. Harnesses are the visible part, so they are the part that gets bought and audited.
The anchorage is what makes them a fall arrest system instead of a uniform, and it is almost always decided on paper weeks before anybody climbs. Related reading: Fall protection tops OSHA's Top 10 for the 16th straight year, the day your fall prevention stand-down actually changes something, and a trench box protects the people inside it, nobody else.
TriCore Safety is a construction and industrial safety consulting firm founded by Chris Fredette, CHST. We write and review fall protection programs, and we walk them in the field to see whether they match the work. Tell us about the job.
Walk enough jobsites and you start seeing the same picture. A crew is working at height, every one of them in a full body harness, lanyards hanging. Nothing is connected to anything. It is easy to read that as a discipline problem. Usually it is not.
The harnesses got bought, the training got done, and nobody ever answered the one question that makes the gear work: what, specifically, does a person connect to in this work area, at this stage of the job? That question is the anchorage plan. It is the part of fall protection that gets skipped most often, and it is the part the standard is most specific about.
The connecting device is the part that gets inspected every morning. What it hooks to usually got decided months earlier, by somebody who never came to the site. What 1926.502(d)(15) actually says It is one sentence, with an "or" sitting in the middle of it: "Anchorages used for attachment of personal fall arrest equipment shall be independent of any anchorage being used to support or suspend platforms and capable of supporting at least 5,000 pounds (22.2 kN) per employee attached, or shall be designed, installed, and used as follows: (i) as part of a complete personal fall arrest system which maintains a safety factor of at least two; and (ii) under the supervision of a qualified person."
So there are two ways to comply, not one rule and an exception. Either the anchor holds 5,000 pounds per person tied to it, or the whole system is designed, installed and used as an engineered system. The second path is both of those subparagraphs, not either one. Safety factor of two and a qualified person, joined by "and." Two things still get lost in the field. Per employee attached.
Two people on the same beam clamp is a 10,000 pound question, not a 5,000 pound one. Crews rig that way constantly, because the anchor got identified for one person and then three people needed it. The engineered path is not a loophole for "it looks strong enough." It says designed and installed, not just used, and it names a qualified person. Subpart M does not define that term.
The definition is in 1926.32(m): "Qualified means one who, by possession of a recognized degree, certificate, or professional standing, or who by extensive knowledge, training, and experience, has successfully demonstrated his ability to solve or resolve problems relating to the subject matter, the work, or the project." Degree, certificate, professional standing, or demonstrated ability. Not seniority.
If nobody on the project fits that and can produce the calculation, the project is on the 5,000 pound path whether it wants to be or not. The numbers that decide whether an anchor works Anchorage strength is only half of it. The rest of 1926.502(d) sets what the system has to do when it catches somebody, and those numbers are what make one anchor usable and another useless.
A personal fall arrest system has to limit the maximum arresting force on an employee to 1,800 pounds when used with a body harness. And under (d)(16)(iii) it has to "be rigged such that an employee can neither free fall more than 6 feet (1.8 m), nor contact any lower level." Two prongs, joined by "nor." Clearing the free fall number is only half of that sentence.
Read it against a real anchor point. An anchor at foot level with a six foot lanyard is not a six foot fall. 1926.500(b) defines free fall distance as the vertical displacement of the attachment point on the harness, from the onset of the fall to just before the system begins to apply force.
Under (d)(17) that attachment point sits in the center of the wearer's back near shoulder level, or above the head. So the D-ring travels the lanyard length plus however far it sits above the anchor, which on most people puts you past ten feet before the energy absorber has done anything. Overhead, that same lanyard can be rigged inside the six foot limit. At the feet it cannot.
Nothing about the hardware changed. Only the geometry did. There is also a second "two" in the standard, and it is not the same two. (d)(16)(v) requires the system to "have sufficient strength to withstand twice the potential impact energy of an employee free falling a distance of 6 feet (1.8 m), or the free fall distance permitted by the system, whichever is less."
That applies no matter which compliance path you are on. The safety factor of two in (d)(15)(i) is a condition on the engineered anchorage path only. One more number worth keeping straight. The 6 feet in 1926.501(b)(1) is the height that triggers fall protection in construction. The 6 feet in 1926.502(d)(16)(iii) is a limit on free fall distance. Same number, measuring two different things.
Where the plan usually runs out In our experience the written fall protection plan is rarely the problem. The problem is that the plan describes the building as designed, and the crew is working on the building as it exists this week. The patterns repeat. The anchorage disappeared with the sequence. The steel that was going to be the tie-off point got decked over.
The overhead structure the plan assumed is now below the work. The anchor was never rated, just chosen. Conduit, a sprinkler line, a guardrail post, a roof vent. The guardrail post is the easy one, because the standard names it. 1926.502(d)(23) says personal fall arrest systems shall not be attached to guardrail systems. The rest are not named anywhere, which is the problem.
Nothing blesses them, and none of them will carry 5,000 pounds or survive the engineered path, so whoever hooked onto one has to explain which of the two compliance paths they were on. The window between guardrail and tie-off. Leading edge work, decking, guardrail installation itself.
The gap where permanent protection is not in place yet and the temporary system has to carry the work is exactly where planning is thinnest. Nobody owns the answer. The crew finds out the plan does not fit at the moment they need it, eight feet up, with a schedule behind them. Worst possible time, worst possible person to solve an engineering question.
Planning it before the crew is up there Plan by work area and by phase, not by building. Answer it for this elevation, this week, this trade. A single site-wide answer is usually wrong by the second phase. Write down what the connection actually is. Not "tie off at height."
The specific anchor, its rating or the basis for it, how many people it carries, and the connecting device it was planned around. Check the clearance math, not just the strength. Free fall distance, deceleration distance, which (d)(16)(iv) caps at 3.5 feet, lifeline and lanyard elongation, which the definitions in 1926.500(b) leave out of both of those numbers so you have to add it yourself, the worker's height below the D-ring, and the distance to the next level down.
Name who decides when the plan does not fit. Someone with authority to stop the task and get an answer, reachable during the shift. If that person is not named and not reachable, the crew will improvise, and improvising is how conduit becomes an anchor point. Re-walk it when the sequence changes. Anchorage is the first casualty of a resequence, and resequences do not usually get routed through safety.
If it helps to run this with a crew rather than read it, our working at heights material and the toolbox talk library cover the same ground in a format you can hand a foreman. The question worth asking on your own project Pick a crew working at height tomorrow and ask one person what they are supposed to connect to and who told them.
If the answer comes back fast and specific, the planning happened. If it is a shrug or a point at something nearby, the gear on that crew is doing less than everyone assumes. Harnesses are the visible part, so they are the part that gets bought and audited.
The anchorage is what makes them a fall arrest system instead of a uniform, and it is almost always decided on paper weeks before anybody climbs. Related reading: Fall protection tops OSHA's Top 10 for the 16th straight year, the day your fall prevention stand-down actually changes something, and a trench box protects the people inside it, nobody else.
TriCore Safety is a construction and industrial safety consulting firm founded by Chris Fredette, CHST. We write and review fall protection programs, and we walk them in the field to see whether they match the work. Tell us about the job.

Comments