THE WHOLE PLANT IMPERATIVE • PART 5
The one place on the plant where deferral doesn’t buy you time. It buys you a deadline somebody else sets.
The day a notice of violation arrives, the plant is usually running fine. Nothing tripped. Nothing alarmed. Availability for the month was excellent. The letter isn’t about how the plant ran. It’s about what the plant can prove, and about a decision somebody made eighteen months ago and has long since forgotten.
That is what separates emissions systems from everything else in this series. A tube leak costs you an outage. A fouled condenser costs you megawatts in August. Both are expensive, and both leave the plant yours to run and the timing of the fix yours to choose. Catalyst and CEMS don’t offer that arrangement. When they fall far enough behind, what’s at risk isn’t output. It’s permission.
Catalyst doesn’t fail. It expires.
SCR catalyst never fails the way the rest of the plant fails. There is no event, no trip, no root cause to write up. It declines — steadily, along a curve you can plot years ahead. Thermal sintering closes off surface area. Trace contaminants poison active sites. Particulate plugs the passages the gas has to move through. Every one of those mechanisms is slow, well characterized, and entirely expected.
Which makes the way it usually gets managed hard to explain. The common pattern is to run until NOx starts crowding the limit, then scramble. The emergency order competes with everybody else’s emergency. It arrives on the vendor’s schedule instead of yours. It installs in an outage you didn’t plan and aren’t being paid for. And every dollar of the premium bought speed and nothing else.
The alternative isn’t sophisticated. Baseline the catalyst at commissioning or at the next changeout. Trend conversion efficiency and ammonia slip quarterly. At roughly 10 to 15 percent below baseline, the curve tells you when you cross the permit limit — usually 12 to 18 months out. That is enough runway to write the replacement into the next planned outage scope instead of reacting to it. Operators who work this way report total lifecycle costs 20 to 30 percent lower. They also don’t have violations, which is the part that never shows up in the cost comparison.
The question was never whether the catalyst degrades. It’s whether the replacement lands in an outage you scoped — or in a letter you didn’t.
The catalyst was sized for a plant you no longer run
Most of the catalyst in the U.S. fleet was specified for baseload duty. SCR works inside a narrow temperature band — roughly 570 to 750°F — and outside it, conversion efficiency falls away fast. A unit that came up in the morning and held flat until night spent nearly every hour of its life inside that band.
That unit doesn’t exist anymore. Plants cycle daily, sit at low load, and start and stop far more often than the original design basis ever contemplated. Every transition is time spent at the edge of the window, where the catalyst is weakest and the emissions are hardest to hold. Low load is also where CO risk peaks, which leans hard on oxidation catalyst that was sized back when low load was an exception rather than a Tuesday.
Then comes the compensation. As the catalyst ages, the reflex is to inject more ammonia to keep NOx where the permit wants it. It works, for a while. It also raises ammonia slip — which fouls downstream surfaces with ammonium bisulfate, can put a plume on the stack, and is a permitted limit in its own right. Injecting your way out of a degraded catalyst is how one maintenance problem quietly becomes two compliance problems.
The one asset that produces evidence instead of power
Everything else on site exists to make power or to protect the equipment that makes it. CEMS exists to prove something. That is a different job, and it fails in a different way.
Under 40 CFR Part 75, data availability has to stay above 90 percent, and most state permits say something similar. The operators with clean records don’t run at the line. They hold above 95, keep redundant analyzers on the critical measurement points, and maintain a written protocol that puts a technician on a failed analyzer the same day rather than at the next convenient window.
Here is the part that’s easy to miss. A plant can run perfectly clean and still have a compliance problem, because missing data is not read as good behavior. The gaps get filled by substitution rules written to be conservative, and they accumulate quietly until somebody totals the year. CEMS maintenance is the easiest thing on the plant to defer for exactly that reason — deferring it appears to cost nothing at all, right up until it is the only thing anyone wants to discuss.
The stack and the ductwork sit in the same blind spot. Liner damage, expansion joint failure, and acid attack from flue gas condensation are not glamorous line items, but they open bypass paths around controls you have already paid for, and they take away the access required to run mandated testing. A plant that can’t safely reach its test ports has a compliance problem before anybody measures anything.
An inspection is an audit of decisions you’ve forgotten
Title V renewal, deviation reporting, and whatever follows an inspection all share one feature: they look backward. They audit choices made 12, 24, and 36 months ago. The quarter the catalyst trend review got skipped. The analyzer left down for three weeks in June. The outage where the stack liner inspection came out of scope to protect the schedule.
Not one of those felt consequential the day it was made. That is the entire mechanism. The damage was never unpredictable — it was only unexamined — and the bill doesn’t arrive when the deferral happens. It arrives when someone with authority reads the record.
Every other system in this series lets you trade money against risk and live with the answer. Emissions systems don’t offer the trade. The counterparty isn’t the market, and the terms were written down before you started.
The next piece goes somewhere with no permit attached to it at all — which is exactly how it gets away with costing so much.
For more information on how Groome can help make a difference, visit our Gas Turbine Plant Performance Solutions page.
Read Part 6 → THE GAS TURBINE INLET.
Start from the beginning → YOUR TURBINE IS FINE. THE ASSETS AROUND IT MAY NOT BE.
