How to Inspect a Harness and What Takes It Out of Service

Why this matters

Inspect on the ground, in decent light, before the harness goes on, and never at height. And settle the ending before you start: a harness that fails any criterion below does not go back in the bag "to look at later." It gets cut through the webbing in several places with a sharp blade on a bench, cutting away from your body, and then it goes in the bin. Every harness that killed somebody after a failed inspection was one that got put back in a bag.

Harness inspection is not a walk-around. It is a search for two different things at once: evidence that this harness has already been loaded, and degradation whose mechanism you cannot see directly. Those need different eyes, and organising the inspection by part - shoulder straps, then buckles, then D-ring - reliably misses the second kind, because heat, ultraviolet and chemical attack do not respect the boundaries between components. Organise it by mechanism instead.

Fit is a separate check with separate pass criteria, and What a Body Harness Does That a Body Belt Cannot owns it. This card is about condition.

Two inspections, two owners

Pre-use, by the person wearing it, every time. For construction, 29 CFR 1926.502(d) requires personal fall arrest systems to be inspected prior to each use for wear, damage and other deterioration, with defective components removed from service. General industry carries a matching duty at 29 CFR 1910.140, which frames it as before initial use each work shift. This one is not a formality and it is not delegable to whoever loaded the van.

Periodic, by a competent person, documented. ANSI/ASSP Z359.2 sets a formal inspection by a competent person at intervals not exceeding one year, or more often where the manufacturer requires it. That consensus standard binds you through your own written fall protection program, through a contract that calls for it, or through a jurisdiction that has adopted it by reference, not on its own. The manufacturer's instructions may set a shorter interval, and where the two differ the shorter one governs.

A third trigger sits outside both: any component subjected to impact loading comes out immediately under 29 CFR 1926.502(d) and does not go back until a competent person has determined it undamaged and suitable for reuse. What that determination actually looks like is in What Happens to a Fall Arrest System After It Arrests a Fall.

The technique that finds what looking cannot

Webbing hides damage when it is flat, because the weave closes over it. Bend the strap into an inverted U over a 6 to 8 inch length and walk that bend hand over hand along the entire strap, both faces and both edges. The bend spreads the weave and brings cuts, broken fibres, pulled threads, embedded grit and glazing to the surface where they are obvious rather than arguable.

Do that on every load-bearing strap. On a standard full-body harness that is six: both shoulder straps, both leg straps, the sub-pelvic strap and the chest strap. It takes a couple of minutes and it is the difference between an inspection and a glance.

Organise it by mechanism, not by part

Load evidence. A deployed impact indicator, wherever the manufacturer put it. Stitching that has pulled or drawn at a strap junction. Webbing that is locally thinned, stretched, or noticeably narrower than the same strap elsewhere. A D-ring that is no longer flat or no longer round.

Cutting and abrasion. Cut edges, nicks, broken fibres, fuzzing concentrated in one spot rather than general wear. Concentrated fuzzing means something was rubbing there, so find out what.

Heat. Fibres that look glazed, shiny, hardened, browned or melted, and webbing that has gone stiff. Heat damage is common on any harness worn near torch work, welding, or an unshielded flue.

Chemical. Discoloration, brittleness, a crunchy feel, or transverse cracks when the strap is bent. And the important part: some chemical attack leaves no visible mark at all, so a known exposure removes the harness on history rather than on appearance. Battery acid in a van, a solvent spill, drain chemicals, concrete admixtures, a cleaning product nobody read the label on.

Ultraviolet. Faded colour plus a chalky or fuzzy surface. UV takes strength out of synthetic webbing in a way you cannot see, feel, or test in the field, and a harness that lives on a dash or hangs on a fence in the sun is aging faster than the one beside it in a bag.

Hardware. D-rings, buckles, tongues, rollers, grommets and keepers. Distortion, cracks, sharp edges, corrosion pitting, a roller that will not turn, a tongue that will not seat, a keeper that has gone slack.

Labels. Read them. If you cannot read the model, the standards, the date and the instructions, the harness comes out of service. This is a real criterion and it is the one people argue about, because the webbing looks fine. An unreadable label means you cannot establish what the device is rated for, and you cannot inspect against instructions you cannot read.

What is a criterion and what is only a finding

The removal criteria are binary. One instance ends it, there is no severity scale, and there is no field repair. Do not re-stitch, do not swap a buckle, do not shorten a strap. Repair is the manufacturer's or nobody's.

• Any cut, nick, broken fibre or pulled thread in load-bearing webbing.
• Any pulled, cut, burned or missing stitch at a load-bearing joint.
• Any heat, chemical or UV signature from the list above.
• Any distorted, cracked, corroded or non-functioning hardware.
• A deployed impact indicator, or any history of impact loading.
• Illegible or missing labels.
• Unknown history. A harness that turns up with no owner and no record is a harness whose exposures you cannot rule out.

Dirt is a finding, not a criterion. It is also the commonest reason a real criterion gets missed, because grit and dust hide exactly what the inversion technique is meant to reveal. Clean per the manufacturer's instructions, which for most synthetic harnesses means water and a mild detergent, then air dry away from heat and out of direct sunlight, then re-inspect. Never use solvents, bleach or a pressure washer, all of which attack synthetic webbing, and never dry a harness on or near a heater, because that is the heat mechanism above applied deliberately.

Three records off one Monday morning

Three harnesses out of one van, and a different reason behind each result.

HARNESS H-07   pre-use, tech on site, on the ground
  Inversion, both shoulder straps     LEFT: at about
    14 in below the dorsal pad, edge nick with
    broken fibres visible when the weave opens
  Inversion, both leg straps clear
  Inversion, sub-pelvic and chest clear
  Stitching at all junctions intact
  Hardware, D-ring flat and free yes
  Impact indicator not deployed
  Labels legible yes
  RESULT                              REMOVED

One finding ended it and nothing else on the record mattered. That is the point of a binary criterion: the tech did not have to judge how bad an edge nick is, and did not have to defend the call to anyone. The harness was cut through in three places at the bench and binned the same morning, and the entry stayed in the log with the reason on it.

HARNESS H-11   pre-use, same van
  Right leg strap heavy dust,
    strap feels stiff along most of its length
  All other straps, inversion clear
  Stitching, hardware, indicator clear
  Labels legible yes
  ACTION                              clean per mfr
    instructions: water and mild detergent, air
    dry away from heat and sunlight, re-inspect
  RE-INSPECTION                       stiffness gone,
    webbing supple, inversion clear
  RESULT                              IN SERVICE

Read the fork on H-11 carefully, because it is the one judgement call in this card. Stiffness that washes out was dirt. Stiffness that survives a proper cleaning is glazing, and glazing is heat or chemical, and either of those is a removal. The cleaning step is not a way to pass a harness. It is the test that tells you which of two very different things you were looking at.

HARNESS H-04   periodic, competent person, shop
  Webbing, all six straps, inversion no visible
    damage anywhere
  Stitching, hardware, indicator clear
  Labels legible yes
  History note from the log was stored in
    the van compartment where a battery leaked
    six weeks ago; harness was not visibly wet
  RESULT                              REMOVED

H-04 is the record that justifies keeping a log at all. Nothing on it could be seen. Sulphuric acid attack on synthetic webbing can leave strength loss with no reliable visual signature, so a documented exposure is the evidence, and appearance is not a rebuttal. A shop with no history field on its inspection record has no way to reach that conclusion, and would have put H-04 back on a roof.

What the record catches that a walk-around does not

Keep four things on the sheet and the inspection starts finding what the eye cannot.

A unique ID per harness. Without it there is no history, and without history there is no H-04 finding.

A history field, not just a condition field. Exposures, near-misses, storage incidents, who used it and where. This is the only field that catches invisible damage.

The in-service date, and the date the labels stopped being legible if that ever happens. Age alone is not a criterion in most manufacturers' instructions, but age plus UV exposure plus an unreadable label is three findings that arrived together.

The disposition, in words. "Removed and destroyed" beats a blank line, because a blank line is how a failed harness gets back into a bag.

The failure mode this whole procedure exists to prevent is quiet and it happens in shops that inspect diligently: a tech finds something marginal at seven in the morning with a crew waiting, cannot decide whether it counts, and decides to keep an eye on it. There is no version of a harness where keeping an eye on it is a control. The criteria are binary so that the decision costs the tech nothing to make, and the destruction step exists so the decision cannot be un-made by whoever finds it in the bin.

References

• 29 CFR 1926.502(d), personal fall arrest system criteria for construction, including inspection prior to each use, removal of defective components, and immediate removal of components subjected to impact loading.
• 29 CFR 1910.140, personal fall protection system criteria for general industry, including inspection before use each work shift.
• ANSI/ASSP Z359.2, in the edition your written fall protection program or contract adopts, for the documented competent-person inspection interval.
• Manufacturer instructions for the specific harness, which own the cleaning method, the inspection interval and any service life limit.
• See related: What a Body Harness Does That a Body Belt Cannot, What Happens to a Fall Arrest System After It Arrests a Fall.