Ageing management

For the last weeks of her rotation, Malik gives Hana a capstone task. She is to draft the outline of an ageing management plan for the fuel channels and feeders of a refurbished Darlington unit, covering the period from now to the end of its planned operating life, and present it to the group. "Everything you have learned in this course comes together here," he says. "Which mechanisms matter, how fast they act, how we will see them coming, and what we will do about them. The tubes are new, but the clock started the day the unit returned to service."

What ageing management is

Ageing management is the set of engineering, operations and maintenance activities that keep the degradation of structures, systems and components within acceptable limits throughout a plant's life. It recognises that every component ages, by the mechanisms described in this course and others such as cable insulation degradation and concrete deterioration, and that safety depends on understanding, monitoring and controlling that ageing rather than assuming components stay as built.

An effective ageing management program typically follows a cycle:

The Canadian framework

In Canada, REGDOC-2.6.3, Aging Management, sets the CNSC's expectations. Licensees must have an integrated ageing management framework that covers all structures, systems and components important to safety, with component-specific programs for the most critical ones, such as fuel channels, feeders, steam generators and the reactor containment structure. Standards in the CSA N285 series set specific requirements: CSA N285.4 for periodic inspection of CANDU components and CSA N285.8 for fitness-for-service evaluation of pressure tubes. Periodic safety reviews, carried out roughly every ten years and in support of licence renewal and refurbishment, assess whether ageing is being managed and whether the plant remains safe compared with modern standards.

The elements of a program

Regulators in both countries describe an acceptable ageing management program in similar terms. A common formulation has ten elements:

Prioritising by risk

A plant has thousands of components, and not all deserve the same attention. Programs prioritise by combining the likelihood of significant degradation with its consequences. Likelihood depends on the mechanism, its rate and the margin remaining; consequence depends on the safety function and on whether degradation could be detected and corrected in time. A simple scoring scheme multiplies the two on scales of one to five and maps the score to an inspection interval. Real programs use more detailed methods, including risk-informed inspection approaches, but the logic is the same.

# Ageing management: rank components by risk and set inspection intervals.
# Score = likelihood (1-5) x consequence (1-5); interval shortens as score rises.
comps = [
    ("pressure tubes: hydrogen equivalent and flaws", 3, 5),
    ("feeder pipes: wall thinning", 3, 4),
    ("steam generator tubes: cracking", 2, 4),
    ("reactor vessel beltline: embrittlement", 2, 5),
    ("Alloy 600 nozzle welds: stress corrosion cracking", 3, 4),
    ("buried service water piping: corrosion", 3, 2),
    ("electrical cables in containment: insulation ageing", 2, 3),
]
def interval_years(score):
    if score >= 15: return 2
    if score >= 10: return 4
    if score >= 6: return 6
    return 10
for name, L, Cq in sorted(comps, key=lambda c: -c[1] * c[2]):
    s = L * Cq
    print(f"{name:52s} score {s:2d} -> inspect every {interval_years(s)} years")
# Remaining-margin trend: hydrogen equivalent in a pressure tube rising at a steady rate
Heq_now, rate, limit = 32.0, 1.6, 120.0     # ppm, ppm per year, example acceptance value
print(f"years until example hydrogen equivalent limit is reached: {(limit - Heq_now) / rate:.0f}")

pressure tubes: hydrogen equivalent and flaws        score 15 -> inspect every 2 years
feeder pipes: wall thinning                          score 12 -> inspect every 4 years
Alloy 600 nozzle welds: stress corrosion cracking    score 12 -> inspect every 4 years
reactor vessel beltline: embrittlement               score 10 -> inspect every 4 years
steam generator tubes: cracking                      score  8 -> inspect every 6 years
buried service water piping: corrosion               score  6 -> inspect every 6 years
electrical cables in containment: insulation ageing  score  6 -> inspect every 6 years
years until example hydrogen equivalent limit is reached: 55

Capstone: fuel channels and feeders

Hana's draft plan for the fuel channels of the refurbished unit follows the ten elements. Its scope covers all 480 pressure tubes, their calandria tubes, garter springs and end fittings, and the feeders. Preventive actions include coolant chemistry control to limit corrosion and deuterium pickup, limits on heat-up and cool-down rates when hydrogen equivalent is high, and avoiding fuelling-machine damage. Monitored parameters are diameter, wall thickness, length, sag, pressure-to-calandria-tube gap, garter spring positions, hydrogen equivalent from scrape samples, and flaws, measured with the channel inspection tools of Lesson 7 on a sampling plan meeting CSA N285.4. Trending compares each channel with creep and hydrogen models, and acceptance criteria come from CSA N285.8 fitness-for-service rules and the licensing basis. Corrective actions include channel repositioning, garter spring repositioning, flaw evaluation and, ultimately, replacement. For feeders, the plan adds wall-thickness measurements at the bends most prone to flow-accelerated corrosion and crack inspections at the most stressed locations, even though the replacement feeders use higher-chromium steel.

Malik's main comment on her draft is about time. A program must look far enough ahead to act before limits are reached, and it must be able to tell when its predictions are wrong. For each mechanism, Hana adds a line showing the margin remaining, the current rate, and the year at which the margin would be used up if the rate continued. Where that year falls within the planned life, the plan includes either more frequent inspection or a mitigation step now.

Beyond metals

Not all ageing is in metals. Electrical cables in containment age by heat and radiation, which embrittle their polymer insulation; they are managed by environmental qualification and condition monitoring, such as measuring insulation elongation or indenter modulus. Concrete structures age by alkali-silica reaction, corrosion of reinforcing steel and, in containments, loss of prestressing force; they are managed by inspection and tendon testing. Obsolescence of instrumentation and control equipment is a form of ageing too, managed by replacement planning. A complete program covers all of them.

Hana presents her plan at the group meeting. The discussion runs long, mostly about inspection intervals and how much margin to keep. Afterwards, Malik tells her that is exactly what a good ageing management meeting looks like: not a recital of procedures but an argument, grounded in mechanisms and data, about how to keep a plant safe for decades.

Repair, replace or run

When degradation approaches a limit, a plant has three broad options: repair or mitigate the component, replace it, or accept a shorter life for the plant. The choice is partly technical and partly economic. Small components such as valves or sections of piping are routinely replaced. Large components, such as steam generators and reactor vessel heads, have been replaced at many plants. A few, such as PWR vessels, are effectively irreplaceable, so their ageing management aims to keep them within limits for the whole planned life.

CANDU reactors are unusual in that their most life-limiting core components, the pressure tubes and feeders, can be replaced. Refurbishment replaces all of them, together with other major components, after roughly 25 to 30 years of operation, giving the unit a further 30 or more years. At Darlington, refurbishment of all four units, including replacement of 1,920 fuel channels and 3,840 feeder pipes, was completed in 2026. Ageing management of the new channels began on the day each unit returned to service, using the same mechanisms and models as before, informed by forty years of data on the original tubes.

Inspection and monitoring technology

Ageing management depends on being able to see degradation. Inspection technology has improved steadily: phased-array ultrasonic probes give images of welds and can size cracks a few millimetres deep; eddy current arrays inspect steam generator tubes quickly; guided wave methods screen long runs of buried or insulated pipe; remote and robotic tools inspect locations that people cannot reach because of dose or access. Online monitoring, such as acoustic leak detection, vibration monitoring, fatigue monitoring of transients and chemistry sensors, gives continuous information between outages. Each inspection method has a probability of detection that depends on flaw size and type, and programs must use methods qualified for the flaws they are looking for.

Projecting margins: a worked timeline

Hana's plan includes a table for each mechanism with three numbers: today's value, the rate of change, and the acceptance value. Dividing the remaining margin by the rate gives the year when action is needed if nothing changes. For hydrogen equivalent in the lead pressure tubes, for example, an illustrative value of 32 ppm rising at 1.6 ppm per year against an example limit of 120 ppm gives about 55 years, comfortably beyond the planned life. For a feeder bend thinning at 0.08 mm per year with 1.5 mm of allowance, the figure is about 19 years at the measured rate, or about 9 years with a safety factor of 2 on the rate, which falls inside the planned life and therefore needs a decision now: either more frequent measurements to confirm the rate, or planned replacement.

Such projections are only as reliable as the rates behind them. Rates change when operating conditions, chemistry or materials change, and they are uncertain when based on few measurements. Good programs therefore update projections after every inspection, use conservative rates where data are sparse, and record the assumptions so that a later engineer can see why a decision was made.

Knowledge management

Ageing management spans longer than a career. The original design data, fabrication records, inspection results and the reasoning behind past decisions must remain available and understandable to engineers decades later. Records of material heats, weld procedures and surveillance results are kept as quality records. Equally important is the knowledge in people's heads, which plants capture through mentoring, documented technical bases and programs like the rotation Hana has just completed. Malik ends the course by giving her a binder of his own notes on pressure tubes and vessel steels, saying that the best ageing management program in the world fails if the next generation of engineers does not understand why it is built the way it is.