THE WHOLE PLANT IMPERATIVE • PART 3
No moving parts, a fraction of the attention, and more forced outages to its name than anything else on site.
Ask a plant manager to name the riskiest asset on site and you’ll usually hear “the turbine.” Pull five years of forced-outage records and you’ll find something else entirely.
The heat recovery steam generator has no rotor, no combustion, no OEM service rep on speed dial. It is, in the plainest terms, a very large box of tubes. It is also the single most common source of forced outages in a combined-cycle plant — the place where availability is won or lost, quietly, over years.
A tube leak doesn’t announce itself the way a compressor surge does. It starts as a hairline flaw that often gets missed, in a location nobody inspected, from a mechanism nobody was tracking. It grows on a clock nobody is watching. Then, on the day the grid needs you most, it takes the unit offline.
And the repair is almost never the expensive part. Plugging a tube is a modest piece of work. The outage wrapped around it — the derate, the replacement power, the capacity penalty, the start you couldn’t make on the hottest afternoon of the year — is what lands on the P&L. As Part 1 argued, the cost of the failure long ago came unhooked from the cost of the fix. Nowhere on the plant is that gap wider than here.
It fails on a schedule. Just not yours.
Almost every HRSG failure mode is slow, predictable, and set in motion by decisions made years earlier.
Flow-accelerated corrosion thins tube walls from the inside, in economizers and low-pressure evaporators, wherever water chemistry and flow conditions conspire. Under-deposit corrosion works beneath scale you can’t see without opening the drum. Creep-fatigue accumulates in headers and tube-to-header welds with every start, stop, and ramp — which, as Part 1 argued, you are now doing far more often than the designers ever assumed. Thermal shock from a fast start distorts what was straight. Attemperator spray hits hot metal and cracks it.
Not one of these is a mystery. Every one of them is measurable before it becomes a failure. So when an HRSG takes a unit down, the honest word for it is rarely “unforeseen.” It is “unlooked-for.” The damage was there, following a documented mechanism, in a place the industry has understood for decades. That is the uncomfortable part of the whole-plant argument: this failure was available to be found. The problem isn’t that HRSG damage is unpredictable — it’s that it is invisible unless you go looking, and going looking costs an outage window and a budget line that is very easy to defer one more year.
Water chemistry is the cheapest reliability you can buy
If there is one place where a small, boring, well-funded program pays for itself in orders of magnitude, it is water chemistry.
Chemistry determines whether your tubes thin, whether deposits form, whether corrosion gets a foothold. A drifting chemistry program does no visible damage this quarter — which is exactly what makes it so easy to trim. It writes checks that come due as tube failures three, five, seven years out, by which time nobody remembers the sampling that lapsed or the treatment that got value-engineered. The cost of doing chemistry properly is small, predictable, and knowable. The cost of not doing it arrives as a forced outage on a hot August afternoon, and it is none of those three things.
Cycling has made this more urgent, not less. Every cold start and every deep turndown puts the water side through conditions it was never designed to see — and the plants cycling hardest are usually the ones whose chemistry programs were written for baseload operation and never revisited.
The alloy problem nobody wants to talk about
There is one more clock running, and this one is specific to the modern fleet.
The high-temperature sections of many HRSGs are built from creep-strength-enhanced ferritic steels — Grade 91 and its relatives — chosen for exactly the strength and thin-wall economics that made today’s units possible. These alloys are unforgiving. Get the fabrication or the post-weld heat treatment even slightly wrong, or run them through enough thermal cycles, and they develop Type IV cracking in the heat-affected zone beside the weld: damage that grows for years with almost no outward sign, and then stops growing, because it has failed.
Much of the U.S. combined-cycle fleet was built in a narrow window around the turn of the century. Those units are now old enough, and have cycled enough, to be arriving at exactly this problem — at the same moment they are being asked to run another decade or two past the life anyone planned for them. This is not a 2040 problem for somebody else to inherit. It is a condition-assessment decision for this outage season.
Look before it looks for you
The whole-plant answer to the HRSG is not “inspect everything.” Nobody can afford that, and it wouldn’t help — HRSGs have hundreds of tubes and you can’t look at all of them.
So the real question was never what the inspection costs. It is which damage mechanism you are choosing not to look for this year — and what it will cost you in the year it finds you.
The answer is to inspect by consequence and by mechanism. Know which circuits your chemistry and your operating profile put at risk. Know where your alloy transitions are and where the welds sit. Sample tubes where the damage mechanisms say the damage will be, not where the ladder happens to reach. Trend it over time, so you are watching a rate and not a snapshot. And treat every planned outage as the cheapest inspection window you will ever get — because it is.
Do this and an HRSG tube failure becomes what it should be: a repair you schedule. Skip it and the HRSG schedules you. The tubes are keeping score whether or not you’re watching.
The HRSG is the biggest single lever in a combined-cycle plant. It is not the only one. Next we take up the system that steals megawatts precisely when they are worth the most — and does it so quietly that most plants never see the bill.
For more information on how Groome can help make a difference, visit our Gas Turbine Plant Performance Solutions page.
Read Part 4 → COOLING AND THE SUMMER PEAK.
Start from the beginning → YOUR TURBINE IS FINE. THE ASSETS AROUND IT MAY NOT BE.
