PAS 128 is the British Standards Institution publicly available specification for underground utility detection, verification and location. It defines four Quality Levels — QL-D, QL-C, QL-B and QL-A — that classify how a buried service was located and how confident a designer or contractor can be in its position on the drawing. Every line on a PAS 128 utility survey deliverable carries a Quality Level label, and a survey is only as useful as the QL it was scoped to.

This guide explains the four Quality Levels in plain English, the detection methods behind each one, the accuracy you can expect, and the procurement mistakes that cost projects time and money. If you need a survey delivered, see our PAS 128 utility survey service.

Why PAS 128 exists

Before PAS 128 was published in 2014, “utility surveys” varied wildly in quality. One firm might mark a water main from a desk-based records search; another might trace it with ground-penetrating radar and confirm depth with a trial pit. Both deliverables looked similar on paper — a single line on a CAD drawing — but the risk to the next person who put a digger in the ground was completely different.

PAS 128 fixed this by mandating that every detected utility on a deliverable drawing is labelled with the method that found it. A line marked QL-D means “from records only — could be metres off, could be missing entirely.” A line marked QL-A means “physically exposed and surveyed in place — accurate to within ±50 mm.” Designers, contractors and asset owners can read the drawing and immediately understand the residual risk per service per location.

PAS 128 is now the de facto standard for utility surveying in the UK. Network Rail, Highways England, HS2, Crossrail, Thames Tideway, and most Tier 1 contractors specify it as the minimum requirement for any works requiring possession or land entry. On large rural sites — see our Anglian Water 70 km² reservoir DCO survey case study — utility records reconciliation and surface evidence capture (QL-C) are routinely done alongside the aerial topographic capture as part of a single integrated brief.

The four Quality Levels

The table below summarises the four QLs and shows how each maps to the RICS Measured Surveys of Land, Buildings and Utilities (3rd edition) accuracy bands — the framework every Angell Surveys deliverable references. Plan accuracies are quoted at 1 sigma (68% confidence); the RICS document also publishes 2 sigma (95% confidence) values which are exactly twice the 1 sigma figure.

QLMethodPlan accuracy (1σ)Depth accuracyRICS BandWhen to use
DStatutory undertaker records (desktop)±500 mm or worseNoneI / JEarly feasibility, planning gates
CSite reconnaissance + records reconciliation±500 mmNoneIPre-design risk register
B1Geophysical detection (GPR + EML + sonde)±100–150 mm±15% of depthGDesign coordination, CDM evidence
AIntrusive verification (trial pit / vacuum exc.)±25 mm±25 mmEPre-construction, clash-critical points

The RICS bands and the PAS 128 Quality Levels are complementary, not competing. PAS 128 tells you how the utility was located; RICS tells you the positional confidence in the resulting line on a drawing. Where a designer or contractor specifies “Band E QL-A at all pile locations” they’re asking for both the verification method (intrusive) and the absolute accuracy class (RICS E, ±25 mm at 1 sigma).

QL-D — Records only

A QL-D survey is a desktop exercise. The surveyor requests utility records from every statutory undertaker that owns infrastructure in the area — water, gas, electricity, telecoms, fibre, district heating, drainage — and overlays them on a site plan. No site work is involved.

QL-D is useful for early feasibility work where you need to know roughly what’s there before committing to a site visit. It is not suitable for design coordination or excavation planning. Statutory records are notoriously inaccurate — water mains laid in the 1960s may be mapped 5 m away from their actual position, and many private services (gas mains across private estates, private fibre runs, abandoned cables) won’t be on any record at all.

A QL-D study typically takes 2–5 working days, depending on how quickly the undertakers respond.

QL-C — Site reconnaissance

A QL-C survey adds a site visit to the records-only QL-D study. The surveyor walks the site, photographs and notes visible utility evidence — chambers, valves, marker posts, kerb-mounted indicators, recent reinstatement scars — and reconciles what they see with the QL-D records. Any service that can be confirmed visually is upgraded from QL-D to QL-C.

QL-C is a useful early-stage check, but it relies entirely on visible surface evidence. A buried service with no surface furniture (a fibre run, a private water supply) remains at QL-D.

QL-B — Geophysical detection

QL-B is the workhorse of PAS 128. A surveyor with calibrated geophysical equipment systematically sweeps the site and physically detects buried services. Three core methods are used:

  • Ground Penetrating Radar (GPR) — a radar antenna sends pulses into the ground and records reflections from buried features. GPR detects most buried objects regardless of material — metallic and non-metallic. It also measures depth (calibrated against site soil conditions). Two-frequency GPR (high-frequency for shallow detail, low-frequency for deep penetration) is standard for QL-B work.
  • Electromagnetic Locator (EML) — also called a Cable Avoidance Tool (CAT) and Signal Generator (Genny). EML detects metallic services that carry an alternating electromagnetic signal. Live electricity cables and signal-clamped metal pipework are picked up reliably.
  • Sonde / signal-clamp tracing — a small transmitter is inserted into a pipe via a chamber, valve or rodding access, and the surveyor traces its signal from the surface. Used for non-metallic ducted services that EML can’t see directly.

QL-B is the minimum standard for design coordination and for evidencing the CDM 2015 Regulation 11 designer’s risk assessment. Network Rail and Highways England typically specify QL-B as the minimum for any survey on their estates.

Plan accuracy is ±150 mm. Depth accuracy is ±15% of depth (so a service at 1 m depth is located to ±150 mm in plan and ±150 mm in depth). The surveyor labels each utility on the deliverable with both the detection method and the achieved Quality Level — so a line might read “W · QL-B · GPR + EML” or “G · QL-B · GPR (no EML signal).”

QL-A — Intrusive verification

QL-A is the gold standard. The buried service is physically exposed — usually by vacuum excavation or a small trial pit — and surveyed in place at the exposed surface. The exact position, depth, diameter, material and condition are recorded directly. There is no remaining geophysical uncertainty.

QL-A is specified for clash-critical points: where a proposed pile, manhole, or trench would land within ±300 mm of a detected utility, or where the consequence of striking the service is severe (a high-pressure gas main, a fibre trunk, a 132 kV cable). It is not specified across the whole site — only at the points where the cost of being wrong is high.

QL-A work requires coordination with the principal contractor (vacuum excavation is itself a controlled activity under CDM 2015) and access to a statutory undertaker permit where the service is third-party owned. Lead time is typically 1–3 weeks per QL-A point.

How to specify Quality Levels in a brief

A common procurement mistake is to ask for “a PAS 128 utility survey” without specifying the Quality Level. The result is usually QL-D — a desktop study — because that’s the cheapest deliverable a surveyor can issue while still using the PAS 128 label.

A useful brief should specify:

  1. The minimum QL for the whole site — typically QL-B for any site with planned excavation or piling.
  2. The points or zones requiring QL-A upgrade — for example, “QL-A at each proposed pile location within 1 m of a detected utility, and at all crossings of the existing 33 kV cable.”
  3. The deliverable format — AutoCAD DWG with utilities on layers by service type AND Quality Level, plus a PAS 128 report including the methodology, equipment, statutory undertaker correspondence and observation log.
  4. The reference system — usually OSGB36 / Newlyn ODN, but contractor-defined site grids are common on rail and highway projects.
  5. The PAS 128 report — this is the audit trail. It contains the equipment used, calibration dates, weather/soil conditions, and any survey limitations. Without it, the drawing is not PAS 128 compliant.

Under the Construction (Design and Management) Regulations 2015, Regulation 11, designers are required to “eliminate, so far as is reasonably practicable, foreseeable risks to the health or safety” of construction workers. Buried services are a primary CDM 2015 risk: striking a live electricity cable is a recognised cause of fatal injury, and a struck gas main can trigger an evacuation, criminal investigation and seven-figure remediation cost.

A PAS 128 QL-B utility survey is now the standard route for designers to evidence that they have addressed buried-service risk. The HSE has not formally adopted PAS 128 as a legal requirement, but it is the recognised standard of care. A designer who specified less than QL-B on a site where excavation was planned, and a buried service was struck, would be in a weak position in any subsequent enforcement action.

What you should expect on a deliverable

A complete PAS 128 deliverable includes:

  • AutoCAD DWG, utilities on layers by service type AND Quality Level (one layer for “Water QL-B,” another for “Water QL-A,” etc.), referenced to OSGB36 / Newlyn ODN.
  • PAS 128 report (PDF) — methodology, equipment list with calibration dates, statutory undertaker correspondence, site observation log, photographs, drawing list, surveyor’s CV.
  • Preview PDF at a stated scale (typically 1:200 or 1:500 for site work, 1:1000 for corridor work).
  • Optional outputs — a 3D BIM utility model in IFC for design-coordination clash detection, an ESRI shapefile or GeoJSON export for GIS, and a separate Quality-Level-A point register listing every intrusive verification point.

If any of these are missing — particularly the report — the deliverable is not PAS 128 compliant, regardless of what the cover sheet says.

Common procurement mistakes

  • Asking for “a utility survey” without specifying QL — defaults to QL-D, which is useless for design coordination.
  • Specifying QL-A across the whole site — wildly over-scoped and unaffordable. QL-A is for clash points only.
  • Forgetting the statutory undertaker permits for QL-A — adds weeks of unplanned lead time when discovered late.
  • Not requesting the PAS 128 report — leaves no audit trail, undermining the CDM 2015 evidence value.
  • Mixing QL labels into a single drawing layer — destroys the per-service-per-location risk signal that PAS 128 was designed to provide.

Frequently asked questions

How long does a PAS 128 QL-B utility survey take? A standard 0.5–1 ha development site takes one to two days on site, with the deliverable issued within 5–10 working days. Linear corridors (highway, rail, utility crossing) are scoped against worked length and access window.

Is QL-B detection accurate enough for piling design? QL-B alone is not accurate enough for clash-critical piling. QL-B identifies where services run; QL-A is required at any pile location within ±300 mm of a detected service.

Can you survey across live track or carriageway? Yes — Network Rail possessions and Highways England Traffic Management closures are the standard route. Survey teams hold the necessary PTS, Sentinel, NRSWA and traffic-management qualifications.

Do you reconcile with statutory undertaker records? Yes — QL-B work begins with the QL-D records request and QL-C reconciliation. Any discrepancy between records and geophysics is annotated on the deliverable.

What if a service has no GPR or EML signature? Some non-metallic, non-ducted services (older private supplies, abandoned-but-not-removed lines) are undetectable from the surface. These remain at QL-D unless physically exposed at a QL-A point. This is a known limitation of all geophysical methods and is recorded in the PAS 128 report.

Need a PAS 128 utility survey on a UK project? Our utility survey service covers QL-D, QL-C, QL-B and QL-A delivery across England, Wales, Scotland and Northern Ireland, with AutoCAD DWG and PAS 128 report as standard outputs. We’re regulated by RICS and hold the safety, training and insurance for Network Rail, Highways England, water-utility and nuclear-licensed sites.