TL;DR
This crane rail survey Port Hedland case study follows a 120 t maintenance crane in an iron ore car-dump workshop that had begun skewing and chewing through wheel flanges. Working inside a 36-hour scheduled outage, ISS measured both runways to ±1 mm with a Leica MS60 MultiStation against AS 1418.18, found the drive-end rail had crept 14 mm out of span over a column-settlement zone, and handed the maintenance crew shim-by-shim adjustment values before the bridge was re-energised on schedule.
Key takeaways
- A 14 mm span error and a 9 mm cross-level difference — both well outside AS 1418.18 — were causing the crane to crab, overloading the drive-end wheels and tripping the long-travel motor on hot Pilbara afternoons.
- The runway had never been surveyed since the workshop was commissioned in 2014; tape measurements during a prior repair had missed the error because they were taken at only three cross-sections, not the 22 ISS measured.
- ISS captured 44 rail-head points per runway with a Leica MS60 (±1 mm + 1.5 ppm), referenced to a four-point local control net rather than GDA2020/MGA2020 zone 50 site coordinates, because rail alignment is a relative-geometry problem.
- Field acquisition took 6.5 hours of the outage; processed deviation plots and adjustment values were issued to the boilermakers within the same shift so shimming could begin immediately.
- Post-adjustment verification brought span to within ±2 mm and cross-level to 3 mm across the full 38 m runway, clearing AS 1418.18 and the client's tightened ±3 mm span specification. Total fee was AUD 7,400 including mobilisation from Karratha.
The challenge: a crane that would not track straight
The client operates an iron ore rail-receival and car-dump facility near Port Hedland, the world's largest bulk-export port. Inside the maintenance workshop, a 120 t double-girder overhead travelling crane lifts dump-cell rotors, drive assemblies and locomotive components — the heavy parts that keep the ore stream moving onto Capesize vessels.
Over several months the crane had started to "crab": the bridge pulled diagonally as it travelled, the drive-end long-travel motor tripped on thermal overload during the hottest part of the day, and the maintenance superintendent had replaced two drive-end wheel sets inside eighteen months. Flange wear on the drive-end wheels was roughly three times that on the idle end — the classic signature of a skewing crane forcing itself along a runway that is no longer parallel.
The workshop had been commissioned in 2014 and the runway had never had a survey-grade alignment check. A contractor had taken steel-tape span measurements during the first wheel replacement, found nothing alarming, and moved on. The problem: tape work over a 38 m runway at three cross-sections cannot resolve a slowly developing geometric error, and it cannot measure cross-level or straightness at all.
The constraint was the calendar. The only window to access rail level safely — bridge isolated and parked clear — was a 36-hour mechanical outage already booked for unrelated dump-cell work. Miss it, and the next opportunity was three months away.
The approach: relative geometry, measured fast
ISS treated this as a troubleshooting survey against AS 1418.18:2018 (runways and monorails) and AS 2550.1 (safe use, which mandates annual runway inspection). The brief was simple and demanding: find the misalignment, quantify it against the standard, and produce adjustment values the boilermakers could act on before the outage closed.
A deliberate decision was made early. The runway alignment did not need to be tied to the site's GDA2020 / MGA2020 zone 50 grid or to AHD heights, because crane rail alignment is a question of relative geometry — is rail A parallel to rail B, is each rail straight, are the two rail heads at the same level. Forcing the work onto the national datum would have added control-transfer time inside a precious outage for no measurement benefit. Instead, ISS established a stable four-point local control network on the workshop columns and slab and worked entirely in that local frame.
The team mobilised from the ISS Karratha base the evening before the outage, so the moment the lockout was confirmed, surveying began without a separate mobilisation eating into the window.
Equipment and method
| Element | Detail |
|---|---|
| Primary instrument | Leica Nova MS60 MultiStation (angle/distance ±1 mm + 1.5 ppm, ATR auto-targeting) |
| Verification scan | Leica RTC360 terrestrial laser scanner for rail-head profile and flange-wear capture |
| Control | Four-point local network on columns and slab; runway baseline defined from idle-end rail |
| Standard | AS 1418.18:2018; AS 2550.1; client spec tightening span to ±3 mm |
| Datum | Local runway coordinate frame (no MGA2020/AHD tie required) |
| Rail-head points | 44 per runway at ~1.8 m spacing, plus every joint and support |
| Access | Elevated work platform (client-supplied, certified before mobilisation) |
The MS60 was set up at two stations to maintain clear sight lines down both 38 m runways. With the bridge parked at the idle end, the team measured the rail-head centreline and top surface at each marked point on both rails, capturing span at 22 matched cross-sections, horizontal straightness along each rail, and the cross-level (elevation difference) between rails. The RTC360 then captured a fast point cloud of the drive-end rail and the worn wheel flanges, giving a continuous rail profile and documenting wear that discrete points alone would miss. Total field time inside the outage was 6.5 hours.
What the survey found
The data was unambiguous. The drive-end rail had crept outward, widening the span from a nominal 28,000 mm to 28,014 mm over a 6 m zone roughly two-thirds of the way along the runway — a 14 mm error against an AS 1418.18 allowance of ±10 mm for spans over 30 m, and well outside the client's tightened ±3 mm specification.
| Parameter | Measured (worst case) | AS 1418.18 limit | Client spec | Result |
|---|---|---|---|---|
| Span deviation | +14 mm | ±10 mm | ±3 mm | Fail |
| Cross-level (elevation difference) | 9 mm | 10 mm | — | Marginal pass |
| Horizontal straightness (over 10 m) | 6 mm | 3 mm | — | Fail |
| Straightness (full 38 m) | 11 mm | 15 mm | — | Pass |
| Rail joint vertical step | 1.5 mm | 2 mm | — | Pass |
The span error and the localised straightness error sat directly above a workshop column that the deviation plot showed had settled — consistent with the soft coastal sediments common around the Port Hedland tidal flats. The settled column had pulled its runway beam, and the rail with it, outward and slightly down. The crane was forced to crab through that zone every travel cycle, explaining the drive-end wheel wear, the flange contact captured in the RTC360 scan, and the long-travel motor tripping under the extra rolling resistance on hot days when thermal expansion narrowed clearances further.
This was the value of measuring 22 cross-sections rather than three: the error was a local feature, not a uniform one, and only dense measurement located it precisely enough to fix.
The result: fixed inside the window
Because ISS issued processed deviation plots and point-by-point adjustment values within the same shift, the boilermaking crew did not wait for a report. They received specific shim thicknesses and lateral correction values for the affected fastening positions, re-set the drive-end rail clips, shimmed the cross-level back toward zero, and ground a minor high point at one joint.
ISS then re-measured the corrected zone before the crane was re-energised. The as-adjusted runway came back to within ±2 mm on span and 3 mm on cross-level across the full 38 m — comfortably inside AS 1418.18 and meeting the client's tightened ±3 mm span target everywhere except a single section held at +2 mm by the as-built fastening pattern, which the client accepted. The bridge was handed back, tracked straight on its first test travel, and the crane returned to service on schedule.
The outcome: a one-survey payback
The survey cost AUD 7,400, including mobilisation from Karratha, two days of field and processing time, the RTC360 verification scan, and the final report. Against that, the client had been replacing drive-end wheel sets at roughly AUD 6,000 each plus workshop downtime — one avoided failure paid for the survey. The underlying column-settlement issue was flagged for structural monitoring before it could worsen.
ISS also recommended bringing the runway into an AS 2550.1 annual survey cycle and establishing the baseline alignment dataset the workshop had lacked since 2014, so future deviation is caught as a trend rather than through wheel failure. The client adopted both and added the runway to a 12-month survey schedule timed to its outage calendar.
Frequently asked questions
Why not survey the crane while it was running?
Crane rail surveying needs safe access to the full runway at rail level, which is impossible while the bridge is travelling. The crane must be isolated and parked clear. Here, ISS deliberately scheduled the work inside an existing 36-hour mechanical outage so the alignment survey added no extra downtime — the bridge was already locked out for dump-cell work.
Why use a local datum instead of GDA2020 / MGA2020?
Rail alignment is a relative-geometry problem: whether the two rails are parallel, straight and level with respect to each other. Tying the survey to the national MGA2020 grid or AHD heights would have consumed control-transfer time during the outage without improving a single alignment measurement. A stable local control network referenced to the runway baseline gives the precision the job needs.
How did the original tape measurements miss a 14 mm error?
The earlier check measured span at only three cross-sections with a steel tape. The error was a local feature concentrated over one settled column, so three widely spaced readings stepped straight over it. It also could not measure cross-level or straightness at all. ISS measured 22 matched cross-sections to ±1 mm, which resolves and locates a developing error that sparse tape work cannot see.
What does a crane rail survey like this cost?
This job was AUD 7,400 including Karratha mobilisation, total station and laser scanning, processing, and reporting. Most crane rail surveys fall between AUD 3,000 for a simple indoor runway and AUD 8,000 for complex or remote systems. Mobilisation distance, runway length, access, and turnaround drive the figure — scheduling inside an existing outage and providing certified access keeps it down.
How quickly can adjustment values be provided?
For troubleshooting surveys inside an outage, ISS processes on site and issues deviation plots and point-by-point shim and correction values within the same shift, so the crew can adjust immediately and ISS can verify the result before re-energising. A formal report follows within five business days.
Request a quote
If a crane in your workshop or process building is crabbing, wearing wheels unevenly, or tripping its travel motor, the cause is almost always measurable runway misalignment — and it is far cheaper to find than to keep replacing wheel sets. ISS surveys crane rails across the Pilbara and nationally with Leica total stations and 3D laser scanning, works inside your outage windows, and delivers AS 1418.18 compliance data with practical, point-by-point adjustment values. Call us on 0407 057 015 to scope a crane rail survey or to build a baseline alignment dataset for your next planned shutdown.