Suppose a 6 MW steam turbine is stripped for a planned six-week outage. The inspection finds two diaphragms beyond repair, and the OEM quotes 34 weeks for replacements. The machine then sits open for months while the plant runs on rented gensets at a burn rate that exceeds the cost of the parts several times over.
Nothing in that sequence was a technical failure. The overhaul was correct, the inspection was correct, the diagnosis was correct. What was missing was that turbine spare parts lead time in Nigeria had never been treated as a design input to the outage plan. It was discovered after the casing was open, which is the one moment when it cannot be managed.
Why turbine spare parts lead time in Nigeria is longer than the catalogue says
An OEM quoting from Europe or Asia gives you an ex-works date. That number is the start of the clock, not the end of it. Between it and a part in your stores sit several stages that plants routinely underestimate.
Manufacturing itself is often the smallest share. Blades, diaphragms, nozzle segments and bearing shells for older frames are frequently made to order rather than held, because the population of machines still running them is small. That alone can be 12 to 30 weeks.
Then comes freight. Sea freight to Lagos or Onne runs 5 to 9 weeks door to port for most origins, and air freight, while faster, is limited by dimension and weight for anything rotor sized. After arrival, customs clearance, inspection and haulage inland add time that varies widely by port, documentation quality and whether the importer is experienced with the relevant HS classifications. The Nigeria Customs Service publishes current procedures, and getting classification right at the first attempt is the single largest controllable variable in this stage.
Finally there is foreign exchange. Securing and settling FX for a large parts order is a real scheduling item, not an administrative footnote, and a delay here stops the order before manufacturing even begins.
If you want a spares position assessed against your actual machine population rather than a generic list, request a technical proposal.
The four stages, with realistic ranges
The table below reflects planning ranges we work to for Nigerian plants. Treat them as planning bands, not quotations. Your figures will move with origin, machine age, part class and how well your documentation is prepared.
| Stage | Typical range | What moves it most |
|---|---|---|
| FX sourcing and order settlement | 2 to 10 weeks | Bank relationship, order size, documentation readiness |
| Manufacture or refurbishment | 4 to 30 weeks | Whether the part is stocked, made to order, or obsolete |
| Freight to Nigerian port | 5 to 9 weeks sea, 1 to 2 weeks air | Origin, dimensions, whether air freight is even feasible |
| Clearance and inland haulage | 1 to 6 weeks | HS classification accuracy, port, abnormal load permits |
| Total, order to store | 12 to 40+ weeks | Obsolescence is the biggest single driver |
The practical consequence is that a part discovered as needed on the day the casing opens will not be fitted in that outage. Planning has to happen before disassembly, which means it has to happen on the basis of predicted findings rather than confirmed ones.
Classifying criticality: which parts you hold and which you order
Holding every spare is capital sitting on a shelf. Holding none is the scenario at the top of this article, a machine open and waiting on parts for months. The resolution is classification, and it turns on two questions per part: what does its absence cost per day, and how long does it take to obtain.
Class A, hold on site. Long lead time and high consequence. Journal and thrust bearing shells, governor and control valve components, main lube oil pump internals, critical instrumentation. These are the parts that turn a two-day repair into a two-quarter outage.
Class B, hold regionally or under a framework agreement. Moderate lead time, moderate consequence. Seals, couplings, common valve internals, filter elements in bulk. A supply agreement with agreed call-off beats both holding and ad hoc ordering.
Class C, order as needed. Short lead time, low consequence, or genuinely commoditised. Fasteners, gaskets, standard filters, consumables.
The classification exercise needs the machine history, not just the parts manual. A frame that has shown repeat bearing distress belongs in a different class than an identical frame that has not. That history comes out of a proper condition review, which is what a power plant audit is for, and it feeds directly into the scope discussion covered in turbine overhaul scope of work.
Obsolescence and the reverse engineering option
The hardest case is the part that cannot be bought at any lead time. Machines installed in the 1970s and 1980s remain in service across Nigerian industry, and for many of them the OEM has withdrawn support or the OEM itself no longer exists in its original form.
At that point the options narrow to three. Refurbish the existing component if the base material allows it. Source from a machine being decommissioned elsewhere, which is opportunistic and unreliable as a strategy. Or reverse engineer and manufacture locally.
Reverse engineering is legitimate engineering when done properly, and it is not a shortcut. It requires dimensional capture from the original part, material verification by analysis rather than assumption, confirmation of heat treatment and surface condition, and a manufacturing route with inspection at each stage. For a rotating component it also requires balancing and, depending on the duty, non destructive examination of the finished item.
Done to that standard it can turn a 40 week problem into a 10 to 14 week one, and it removes the FX and freight stages entirely. Done casually, by copying a worn part without material verification, it produces a component that fits and then fails, which is worse than having no part at all. This is central to our rotating equipment services work, and the failure modes it has to design against are the ones described in turbine bearing babbitt failure.
Planning the outage around lead time rather than despite it
The practical method reverses the usual sequence. Instead of scheduling the outage and then discovering what is needed, you predict what is likely to be needed and let that set the schedule.
Start with a pre outage inspection that does not require full disassembly. Borescope inspection, lube oil analysis, vibration history review and operating data trending will predict a large share of findings before anything is opened. Lube oil analysis in particular gives early warning of bearing distress, and the practice is covered in turbine lube oil analysis.
From that evidence, build a probability weighted parts list: parts almost certain to be needed, parts plausibly needed, and parts needed only in a worst case. Order the first group outright. Place the second group on conditional or call off terms where the supplier will accept it. Accept explicit risk on the third.
Then set the outage date so that the long lead items land before it, with float. A plant that opens a machine knowing the diaphragms are already in the store has an entirely different outage from one that opens it hoping they will not be needed.
This is also where the economics become visible. Consequence has to be quantified in the same units as the parts, which is what plant downtime cost per hour sets out, and it is usually the argument that justifies the Class A holding.
Inspection intervals interact with all of this, because a longer interval increases the probability of severe findings and therefore the parts exposure per outage. That relationship is covered in turbine inspection intervals.
What to require from a supplier
Ask for lead time broken into the four stages rather than as a single number, because a single number hides where the risk sits and gives you nothing to manage. Ask what is held in stock versus made to order, in writing. Ask for the country of manufacture and the HS classification, since both drive the clearance stage.
For any critical part, ask what the supplier will provide if the quoted date slips: a loan item, a refurbished interim, or nothing. The answer tells you whether the date is a commitment or an estimate.
For regulatory context on generation and captive power obligations that may affect how long a plant can stay down, the Nigerian Electricity Regulatory Commission is the reference point.
We build spares strategies alongside turbine overhaul and generator maintenance programmes rather than as a separate exercise, because the classification only works when it is informed by the condition of your specific machines. Book a plant assessment to start from your machine list and history.
Frequently Asked Questions
How far ahead should we order spares for a planned turbine outage?
Work backwards from the longest lead item in your probability weighted list and add at least six weeks of float. For most Nigerian plants that means placing the first orders 9 to 12 months before a major outage. Minor outages with a narrow expected scope can work to 4 to 6 months, provided the Class A items are already on site.
Is air freight worth the premium for turbine parts?
For small high value items with high consequence, usually yes, because the freight premium is small against a day of lost production. For rotor sized or heavy components it is often not feasible at all due to dimension and weight limits, so it should never be assumed as a recovery option in the plan.
Can reverse engineered parts be trusted in a critical turbine application?
Yes, when the process includes material analysis rather than assumption, verified heat treatment, dimensional inspection against the original, and appropriate non destructive examination and balancing of the finished component. The risk lies entirely in shortcuts. A part copied dimensionally from a worn original without material verification should not go into a critical application.
What is the single most common planning mistake with spares?
Treating the OEM ex-works date as the delivery date. It excludes FX settlement, freight, clearance and inland haulage, which together frequently exceed the manufacturing time itself. Plans built on the ex-works number are typically short by two to four months.