How often should you inspect a manhole? ITU-T L.330 gives numbers

When did you last inspect the manholes, poles and ducts in your network? And on what criterion did you decide it still wasn’t time? For most networks the honest answer is that inspection happens when someone is on site for something else: a fault, a build. That is not a schedule, it is a side effect.

ITU-T Recommendation L.330 (10/2020), “Telecommunication infrastructure facility management”, approved on 29 October 2020 by Study Group 15, answers that blind spot. The Summary explains why: “Extensive outside telecommunication infrastructure facilities that support information technology continue to deteriorate due to aging.” And who it is for: “its intended users are not only operators who need to improve life-cycle management, but also developers who consider applying rapidly progressing technologies.”

Three categories, and the one that always goes missing

Table 1 (clause 6.2) sets out three kinds of inspection. Daily inspection: “Inspection performed visually or with simple tools when working on-site” — the visual check made while on site for something else, which almost everyone already does without calling it that. Periodic inspection: “Inspection performed at planned intervals. Some inspection items may include quantitative records” — the one that is almost always missing: it needs a calendar. Precise inspection: “Inspections performed to determine risk rating when more detailed assessments are required after result of the daily/periodic inspection” — triggered when the periodic check finds something a glance can’t explain.

Daily inspection happens on its own, as a by-product of field work. Periodic inspection doesn’t: if it isn’t written into a plan with a date, it simply doesn’t happen — until precise inspection is left to discover it, on an asset already in trouble.

How often

Table 5 (clause 8.1), “Typical inspection frequencies”, sets the intervals by asset category:

Concrete pole, steel pole or glass-reinforced plastic (GRP) pole: ≤ 5–10 years
Wooden pole: ≤ 5 years
Guy-line and suspension wire: ≤ 5–10 years
Aerial cable and drop cable: daily inspection
Open-cut tunnel and shield tunnel: ≤ 5 years
Manhole cover: ≤ 5–10 years
Concrete frame and polymer concrete frame of handholes: ≤ 10 years
Bridge-supported conduit: ≤ 5–10 years
Cabinet or pillar: < 2–3 years
Tower: < 2–3 years
Prefabricated shelter: < 2–3 years

A note in the table: “Inspection frequency may be different in accordance with the types of terrain in which the facility is installed” — terrain shifts the calendar; the table only gives the starting point.

Two figures stand out. Aerial and drop cables get daily inspection, while a manhole cover can go ten years: not an inconsistency — the aerial cable faces wind and traffic every day, the cover doesn’t — but proof that “more visible” isn’t “more inspected.” And street cabinets carry the tightest frequency of all, under two to three years: anyone planning around what is seen, rather than what deteriorates, is looking at the wrong criterion.

Yes/no, or a number

This is where L.330 draws the distinction the whole piece rests on, clause 8.2: “Most check points for the inspection of outside plant facilities can be examined preliminarily in terms of the presence or absence of the phenomenon, which are designated ‘Y/N’. Some check points require a quantitative measurement to evaluate the risk rating or degree of deterioration, which is designated using the minimum required unit for the measurement.”

In practice: some checks only need a yes or a no — the crack is there, or it isn’t. Others demand a number with its unit — the crack is 0.2 mm wide, or 0.4 mm. That isn’t a matter of form: Y/N produces an opinion, tied to whoever happened to look that day; a measurement produces data, comparable over time — it says not just that the crack exists, but whether it’s growing, and how fast. The difference isn’t one of precision, it’s one of kind: one carries the observer with it, the other carries only the facts.

The manhole, item by item

Table 10, “Inspection of manhole and handhole”, applies that distinction to an asset every reader recognises.

Manhole cover:

abrasion: Y/N in daily inspection, 0.1 mm in periodic inspection;
crack: Y/N in both;
rattle: Y/N in both;
level difference: Y/N in daily, 0.1 mm in periodic.

Concrete frame:

crack: Y/N in daily, 0.1 mm in periodic;
water infiltration: Y/N in both;
exposed reinforcement: measured by length, unit 1 cm;
surface defects in the concrete: unit 1 cm²;
in precise inspection, four further measurements: cover depth 1 mm, carbonation depth 1 mm, concrete strength 1 N/mm², chloride content 0.1 kg/m².

Polymer concrete frame:

crack: Y/N in daily, 0.1 mm in periodic;
water infiltration: Y/N in both.

Two notes. Note 1 points to ISO 16311: “The crack pattern should be recorded for large crack widths (e.g., no less than 0.3 mm)” — the crack pattern must be recorded once the width reaches 0.3 mm. Note 2, for the polymer frame: “The areas where cracks occur (top/side/bottom surface) should be carefully recorded” — it must be written down whether the crack is on the top, side or bottom surface.

How we check it

On site we apply the same distinction: where L.330 asks for Y/N we log a point check, where it asks for a measurement we log it with the right unit — millimetres, square centimetres, newtons per square millimetre. The first session produces a dated inspection plan: what gets checked, how often, on what criterion and unit, and the list of assets — poles, manholes, conduits, cabinets — for which no date of last inspection currently exists. The plan stays with the client whether or not the work continues.

See the service · Talk to an engineer

The four risk ratings

Table 2 (clause 6.3) turns every finding into a risk class, on four levels:

I, normal: “No action is required”;
II, low: “Follow-up inspection is required. Defects have no immediate effect, but are expected to move to a higher rating in the long term”;
III, high: “Planned repair is required. Defects will progress quickly and can be expected to move to the higher rating”;
IV, emergency: “Immediate repair is required. Defect induced accidents may occur.”

Why a scale is needed: without one, “needs repair” is the opinion of whoever happened to look that day, and two engineers can rate the same defect differently. With the scale, the finding becomes a level — I, II, III or IV — that feeds into the maintenance plan and decides on its own whether and when to act.

What it isn’t

L.330 is an international technical recommendation, not a law: compliance is voluntary, and the frequencies in Table 5 are explicitly stated as “typical”, not a calendar obligation imposed by a regulator. The value lies in naming it in a specification or a maintenance plan: only there does a “typical” frequency become a verifiable commitment, with a date and an owner.

Two threads, applied

First thread: inspection categories, frequencies by asset, risk ratings and units of measurement become verifiable lines in the maintenance specification and acceptance checkpoints, with the record — date, category, finding, unit — the client can produce.

Second thread: inspection findings — asset, date, category, risk rating, measurement — don’t stay on scattered sheets. With CSIDIA, the group’s other company, they become a single map of the network on which an AI runs the diagnosis and the crew closes the fault: with a risk rating and a date on every asset, the order of interventions gets calculated — as already happens with the as-built record and the splice re-entry log. Within the client’s perimeter: on-premise, on standalone machines with no deep integration, or a dedicated cloud with a data centre in Italy, always with shared management.

Could you say, right now, for how many of your manholes there is a date of last inspection and a written finding? Talk to an engineer: the site visit is at no cost, and the inspection plan that comes out of it stays yours whatever you decide next.