7 Signs Your Tower Internals Need Replacement Before Your Next Turnaround
Don’t Wait for a Shutdown to Find Out Your Column Has Been Telling You for Months
A refinery turnaround is expensive by design. Planned downtime, coordinated crews, replacement hardware staged and ready — that’s the cost of doing it right. What’s far more expensive is arriving at a turnaround unprepared, finding internal damage that wasn’t anticipated, and discovering that the column has been degrading for the last two years in ways that operating data was clearly signaling.
Most tower internal failures don’t announce themselves suddenly. They build. The column continues to run — just less efficiently, less reliably, and at increasing cost — until the next scheduled plant shutdown reveals what the data was already suggesting.
The engineers and turnaround managers who consistently get the most out of their columns know how to read those signals before they’re standing inside a vessel looking at the damage. Here are seven of the most reliable ones.
1. Product Purity Has Drifted Without a Clear Operating Cause
The most direct indicator of internal degradation is the one that shows up in product quality data: fractionation that was previously sharp is now less so.
When overhead or side-draw specifications start drifting — when the gap between light and heavy key components in adjacent cuts begins to narrow, or when operators are pushing reflux ratios higher to maintain spec — the column is producing less theoretical separation per physical tray or packing stage than it was designed to.
Before attributing this to feed composition changes or crude slate variation, rule out internal condition. Fouled trays reduce effective vapor-liquid contact. A degraded liquid distributor creates maldistribution in a packed bed that collapses separation efficiency across the entire section. Damaged valve caps or worn tray decks reduce the active bubbling area that drives mass transfer.
If your operations team has been incrementally adjusting to maintain product quality over the last operating cycle — more reflux, more reboiler duty, tighter draw control — that’s not optimization. That’s compensation for internal wear.
What to look for: Increasing energy consumption to hold the same product spec, creeping changes to draw temperatures or reflux ratios that don’t correlate with feed changes, and product quality that has gradually moved toward its specification limit rather than comfortably within it.
2. Column Pressure Drop Is Higher Than Historical Baseline
Pressure drop across a distillation column or a packed bed section is one of the most sensitive indicators of hydraulic condition. It reflects what’s actually happening with vapor-liquid flow on the internals — and when it increases without a corresponding increase in throughput, something has changed inside the column.
Fouling is the most common cause. Solids deposition on trays, packing, or distributors partially blocks vapor flow paths, increasing resistance and raising pressure drop. Corrosion products can accumulate on tray decks and in packing beds. Polymerization or coking in reactive services progressively plugs internals from the feed zone outward.
Less obvious causes include structural deformation — trays that have sagged, warped, or shifted from high vapor velocities or thermal cycling — and liquid distributor malfunction that causes uneven liquid loading on packed beds, which can produce localized flooding and apparent pressure drop increases.
What to look for: Pressure drop that has trended upward over successive operating months without a corresponding change in feed rate or operating conditions, or pressure drop that spikes at throughputs that were previously handled without constraint. Compare against the baseline established immediately after the last turnaround, not against the operating average.
3. Throughput Is Limited Below Historical Operating Capacity
If your column is flooding at rates it previously handled without issue, the hydraulic capacity of the internals has changed. Flooding — the point at which vapor entrains excessive liquid upward or liquid backs up on trays — occurs at a predictable vapor and liquid loading for a given internal design. When that threshold drops, the internals are not performing as designed.
Fouled or corroded tray decks reduce the open area available for vapor flow, effectively lowering the vapor capacity that causes flooding. Damaged or missing valve caps change the tray hydraulics. Collapsed or shifted packing beds reduce the void fraction available for vapor passage.
In practice, this shows up as column flooding at feed rates that operators know from experience the column should handle — and it often gets attributed to process conditions rather than internal condition because the column ran fine at those rates during the previous cycle.
What to look for: Flooding onset at lower-than-historical feed rates, operational limits that have crept downward from one operating cycle to the next, or differential pressure readings that indicate flooding in a specific section at throughputs that didn’t cause problems previously.
4. Temperature Profile Has Shifted Across the Column
The temperature profile along a distillation column — the temperature at each tray level from bottom to top — is a fingerprint of the separation taking place inside. When that profile shifts without a change in feed composition or operating targets, the separation is occurring differently than it was designed to.
A flattened temperature profile in a section of the column suggests poor separation in that section — fewer effective theoretical stages than design. This can indicate fouled or damaged trays in a specific region, maldistribution in a packed bed section, or feed inlet problems that are disrupting vapor-liquid equilibrium near the feed zone.
A shifted profile — where the same temperature gradient has moved up or down the column relative to historical position — suggests that the feed point is not working as intended, that draw compositions have changed due to reduced fractionation, or that vapor-liquid traffic patterns have been disrupted by internal damage.
What to look for: Temperature readings at specific tray levels that have trended away from historical values, or temperature spans across a section that have compressed over successive months. Distributed temperature monitoring makes this analysis tractable; columns with sparse thermocouple coverage may require inference from product quality data.
5. You’re Seeing Unexpected Liquid or Vapor Maldistribution Symptoms
Maldistribution — uneven distribution of liquid or vapor across the column cross-section — is one of the most efficiency-destructive conditions that can develop in a distillation column, and one of the hardest to diagnose from outside the vessel.
In packed columns, liquid distributor performance is the critical variable. A distributor that was functioning correctly at commissioning can degrade through corrosion, plugging of distribution points, or physical damage. Even partial maldistribution — liquid concentration at one side of the column cross-section — can reduce packed bed efficiency to a fraction of design capacity, because the vapor rising through the dry side of the bed contacts no liquid and performs no separation.
In trayed columns, vapor maldistribution from damaged downcomers, warped tray decks, or shifted seal pans produces similar effects — active separation in some areas, bypass in others.
What to look for: Asymmetric temperature readings at the same elevation (if multiple thermocouples exist at a given tray level), product quality inconsistency that doesn’t track with operating variable changes, and efficiency calculations that suggest far fewer theoretical stages than the physical tray count would predict.
6. Corrosion History or Service Conditions Suggest You’re Past Design Life
All tower internals have a service life that is determined by the corrosion environment, the operating temperature and pressure, the mechanical stresses from thermal cycling and hydraulic loads, and the quality of the original installation.
In aggressive services — amine treating, sour water stripping, crude units with high chloride or sulfur content, or columns handling corrosive overhead environments — that service life may be significantly shorter than in cleaner applications. Internals that were specified for one service and have been operating in a modified service may have degraded faster than historical experience would suggest.
If your column’s operating history includes a known corrosive excursion, a period of operation outside design conditions, or a service change that wasn’t followed by an internal reassessment, those are flags for earlier-than-scheduled inspection.
What to look for: Known corrosion history in the service, previous turnaround findings that showed accelerated degradation, or service conditions that have changed since the internals were originally specified. Carbon steel internals in a service that has become more corrosive than original design conditions are a particular concern.
7. It’s Been More Than Five Years Since a Full Internal Inspection
There is no substitute for direct inspection. Operating data can identify that something has changed inside a column. It cannot tell you the physical condition of the internals — whether tray decks are corroded through, whether valve caps have worn to the point of losing seal, whether a liquid distributor has shifted or is partially plugged, whether packing has been crushed or contaminated in ways that reduce its effective surface area.
Five years is a common turnaround interval for process columns in moderate service. For columns in more aggressive service — fouling-prone feeds, corrosive overhead environments, high-temperature applications — that interval may be too long. For columns that are showing operational symptoms consistent with internal degradation, waiting for the next scheduled turnaround to confirm internal condition is a choice to continue operating below design efficiency, at elevated energy cost, for however long remains until that turnaround.
The cost of an unplanned shutdown — forced by a sudden internal failure, a flooding event that damages downstream equipment, or a product quality excursion that triggers a regulatory response — substantially exceeds the cost of adding scope to a planned turnaround based on what the data is telling you.
What to look for: An extended run since the last full internal inspection, operational symptoms from any of the preceding six signs, and a turnaround planning process that doesn’t include internal condition assessment as a starting point.
How to Act on These Signals Before the Turnaround
Recognizing that your column’s internals may be degraded is the starting point. What happens next determines whether you address the problem efficiently or expend significantly more resources fixing it reactively.
AMACS’s Hardware Needs Assessment is designed for exactly this situation. It works through your column’s operating history, current performance data, and last inspection findings to identify what’s likely failing, what internals should be replaced versus cleaned and returned, and what hardware needs to be staged for your upcoming turnaround.
The objective is to arrive at the turnaround with decisions already made — not to spend the first days of the shutdown diagnosing what the operating data was already suggesting for months.
For columns showing severe symptoms or unexpected operating constraints, AMACS also provides emergency 24/7 support for situations that can’t wait for a scheduled turnaround window.
If your column is showing any of these signs and a turnaround is on the horizon, contact AMACS to discuss what the data suggests and what replacement scope makes sense before you open the vessel.
Request a Hardware Needs Assessment →
Frequently Asked Questions
Tower internals are the components inside a distillation or separation column that create the vapor-liquid contact environment where separation occurs — including trays, packing, liquid distributors, feed inlet devices, and mist eliminators. They degrade over time through corrosion, fouling, mechanical wear, and thermal stress. When internals degrade below a functional threshold, they reduce separation efficiency, limit throughput capacity, and increase energy costs. Replacement restores the column to design performance.
The most reliable indicators are operating data: product purity drift that doesn’t correlate with feed changes, increasing pressure drop without throughput increases, throughput limitations that have shifted downward from historical levels, and temperature profile changes along the column. None of these is definitive on its own, but a pattern across multiple indicators is strong evidence of internal degradation. Correlation with known service history — fouling-prone feed, corrosive environment, extended run without inspection — increases confidence.
A refinery turnaround is a planned shutdown of a process unit for inspection, maintenance, and equipment replacement. It is the primary opportunity to inspect and service tower internals, which can only be accessed when the column is offline, depressurized, and cleaned. Tower internals should be inspected at every turnaround, with replacement decisions based on measured condition, service history, and performance data from the operating cycle. Columns showing performance degradation should be prioritized for full internal inspection at the next available turnaround window.
Degraded internals continue to reduce separation efficiency, requiring more energy input to maintain product quality. Throughput limits decrease as hydraulic capacity drops. In severe cases, flooding, carryover, or structural failure of internal components can force an unplanned shutdown — with significantly higher costs than planned replacement scope. In regulated industries, product quality excursions from degraded separation can have compliance implications beyond the immediate operational cost.
AMACS provides hardware needs assessment services that work through column operating history, current performance data, and previous inspection findings to determine what internal replacement scope is appropriate for an upcoming turnaround. The goal is to arrive at the turnaround with hardware decisions already made and replacement internals staged, rather than making those decisions under time pressure after the column is opened. AMACS manufactures replacement internals including trays, packing, distributors, and mist eliminators, and provides field support through the installation.
A turnaround typically refers to a planned maintenance event for a specific process unit — a single column, a reactor, or a defined unit boundary. A plant shutdown refers to a broader cessation of operations across multiple units, often for safety, regulatory, or major capital project reasons. Tower internal inspection and replacement work can occur during either event, but turnaround planning for column internals should begin well in advance of either type of scheduled outage to ensure replacement hardware is engineered, manufactured, and available.
In some cases, yes. The decision to clean and return versus replace depends on the physical condition of the internals — the degree of corrosion, structural integrity, and whether cleaning can restore functional performance. Trays with minor fouling and no structural damage can often be cleaned in place. Severely corroded, warped, or mechanically damaged internals require replacement. A hardware needs assessment or direct inspection is the basis for making that determination — not a default assumption in either direction.
For a standard turnaround with anticipated internal replacement scope, planning should begin six to twelve months in advance. Custom-engineered internals require lead time for design, fabrication, and quality review. For columns showing significant performance degradation or where previous turnaround findings indicated near-term replacement needs, earlier engagement allows for more complete engineering review and avoids fabrication schedule pressure. Emergency replacement capability exists for situations that can’t wait, but planned procurement is consistently less expensive and produces better engineering outcomes.