Refinery Neutralizer Control During Startup, Shutdown and Low-Rate Operation

July 28, 2026

A crude-unit overhead neutralizer program is usually designed around normal operation.

The crude rate is established. The tower pressure and top temperature are reasonably stable. Water wash is available. The neutralizer pump operates inside a familiar range. Samples arrive on schedule. Operators understand how the accumulator pH normally responds to chloride, ammonia and chemical dosage.

Startup, shutdown and restart remove that stability.

During a transient, temperatures move faster than chemical inventories. Vapor flow changes before sampling catches up. Water can appear, disappear or condense in a different location. A dosing pump may be running while the carrier flow is too low to distribute the chemical. A water wash that is protective at full throughput may quench, maldistribute or flood equipment at low rate. Residual salts from the previous campaign may redissolve, migrate and concentrate somewhere else. The first accumulator sample may be collected only after the most vulnerable condensation period has already passed.

This is why refinery startup neutralizer control should not be treated as a reduced version of the normal operating program.

It is a separate operating mode with different chemistry, different measurements, different limits and different decision authority.

The central question is not:

“What neutralizer rate do we use at normal throughput?”

It is:

How do we preserve corrosion and salt-control margin while the overhead system is moving through states for which steady-state assumptions are temporarily invalid?

Transient operation creates a timing problem before it creates a dosage problem

Most unsuccessful startup responses begin by focusing on the chemical rate too early.

The neutralizer rate matters, but the first technical challenge is timing.

Four clocks are moving at once:

  • the process clock;
  • the condensation clock;
  • the chemical-delivery clock;
  • the measurement clock.

The process clock

The process clock describes changes in feed, heat input, vapor generation, pressure, reflux, overhead loading and throughput.

During startup, the process may move from steam or circulation service to hydrocarbon feed, then through low-rate operation and finally toward the target crude rate. During shutdown, the sequence reverses, but retained liquids, deposits and local cold spots make the chemical path different from a simple backward startup.

The condensation clock

The condensation clock describes where water first becomes available and how the wet region moves as temperature and pressure change.

The vulnerable zone is not fixed. A line section that remains dry during normal operation may become the first wet surface during startup. A condenser that normally receives a well-mixed vapor stream may temporarily see uneven liquid distribution. A low point may collect acidic water before a representative accumulator sample exists.

The chemical-delivery clock

The chemical-delivery clock begins when the product leaves the tank, not when an operator presses the pump start button.

Line fill, pump priming, pulsation, check-valve opening, carrier flow, injector wetting, droplet formation and travel through the process all create delay. At low volumetric rate, that delay can be longer than expected.

The measurement clock

The measurement clock includes sample transport, sample conditioning, analyzer stabilization, laboratory turnaround and the time required for corrosion indicators to respond.

A pH result may represent chemistry from an earlier process state. Iron may rise after the acid event has ended. A corrosion probe may respond faster than a manual coupon but still cannot identify every localized mechanism.

A strong refinery transient corrosion control plan aligns these four clocks before it adjusts dosage aggressively.

Define transient states instead of using one generic startup procedure

State-based refinery startup control sequence from mechanical readiness through stable operation

“Startup” is too broad to be a useful corrosion-control instruction.

The overhead passes through several chemically different states, and each state needs its own objective.

State 0 — Mechanical readiness

The unit is not yet processing crude, but the corrosion-control system must already be ready.

Confirm:

  • neutralizer storage level and product identity;
  • tank temperature and product pumpability;
  • pump calibration and minimum reliable output;
  • injection-line integrity;
  • quill or atomizer condition;
  • carrier steam, water or other motive-fluid availability;
  • water-wash availability and nozzle condition;
  • sampling-system cleanliness;
  • online analyzer calibration;
  • laboratory coverage;
  • fallback chemical inventory.

Mechanical readiness must be confirmed before the process begins to create an acidic overhead environment.

State 1 — Dry heat-up or circulation

Equipment temperature increases, but normal crude-vapor chemistry may not yet exist.

The main risk is assuming that a pump should run simply because the normal program uses that chemical.

A neutralizer added into an inadequately flowing line can collect near the injector, wet a local surface, react with residual contamination or create a concentrated pocket that later enters the process as a slug.

The objective is to verify distribution conditions, not to chase an accumulator pH that may not yet represent normal service.

State 2 — Initial hydrocarbon and acid precursor entry

Crude or hydrocarbon feed begins entering the system. Chloride, sulfur species, organic acids, ammonia and background amines can begin moving through the unit before stable overhead water is available.

This state requires close attention to feed history, desalter condition, residual wash water and expected acid release.

State 3 — Approach to first effective condensation

This is often the most important startup state.

Temperature and pressure approach the point where water begins to condense on real equipment surfaces. The first liquid can be highly acidic because a small amount of water absorbs a relatively concentrated acid inventory.

The first effective condensation point may not be the same as the normal accumulator sampling location.

State 4 — Low-rate wet operation

Water is present, but vapor velocity, mixing, water wash, chemical atomization and residence time may all differ substantially from normal operation.

The key question becomes whether the chemical is reaching the entire vulnerable cross-section or only part of the flow.

State 5 — Rate ramp and control transfer

Throughput increases, temperatures shift and the system moves toward steady-state control.

The challenge is deciding when startup limits can be replaced by the normal operating envelope.

State 6 — Stable operation

The unit has met the defined stability criteria. Only now should normal optimization logic fully replace startup conservatism.

Build the startup plan backward from the first wet metal surface

Crude-unit overhead piping and condenser network where condensation location shifts during startup

A good startup plan does not begin with the accumulator.

It begins with the first location where corrosive water can contact metal.

Identify the expected condensation path

Review:

  • tower-top temperature trajectory;
  • overhead pressure trajectory;
  • condenser duty;
  • ambient conditions;
  • steam and hydrocarbon composition;
  • line-wall temperatures;
  • cold spots and low points;
  • water-wash injection location;
  • sample locations.

The purpose is not to predict one perfect dew-point temperature. It is to identify the equipment sequence through which wetting is likely to develop.

Separate water dew point from salt risk

Water condensation and salt formation are related but different boundaries.

An amine hydrochloride or ammonium chloride salt can become stable before bulk water is available to dissolve and transport it. The site's guide to amine salt deposition in refinery overhead systems explains why a normal downstream pH does not prove that upstream surfaces remained salt-free.

During startup, both boundaries move.

A conservative plan therefore tracks:

  • predicted water dew point;
  • predicted salt point;
  • wall-temperature margin;
  • location of effective wash water;
  • chemical distribution.

Define a no-blind-zone rule

The refinery should know which measurement or operating limit protects each critical section.

If no valid sample, model, corrosion monitor or temperature measurement represents a section during startup, that section is a blind zone.

A blind zone should trigger conservative operating limits, additional temporary monitoring or delayed progression to the next startup state.

Neutralizer should arrive before acid damage, but not before delivery conditions exist

This timing balance is central to crude tower overhead startup.

Starting too late can expose the first condensate to severe acidity.

Starting too early can place concentrated chemical in a system that cannot transport or mix it correctly.

Define the chemical-start permissives

A startup procedure can require several conditions before neutralizer injection begins:

  • minimum process or carrier flow established;
  • injection line primed;
  • pump discharge pressure confirmed;
  • flow indication valid;
  • injector temperature within its approved range;
  • expected acid precursor entry confirmed;
  • operator communication complete.

The exact permissives depend on the unit and injection design.

Use a verified initial dose, not an improvised guess

The initial rate should come from the approved startup basis.

That basis may use:

  • minimum controllable pump flow;
  • expected acid loading;
  • product active concentration;
  • carrier ratio;
  • startup temperature profile;
  • credible chloride case;
  • salt-margin limit.

The initial rate is not necessarily the final rate. It is the starting point for a controlled response.

Confirm actual delivery

Refinery operator verifying overhead process conditions and neutralizer delivery during startup

Commanded pump rate is not enough.

Confirm, where available:

  • tank-level movement;
  • calibrated flow;
  • pump stroke or speed;
  • discharge pressure;
  • return-flow status;
  • line temperature;
  • injector differential pressure.

The engineering principles in neutralizer injection, atomization and mixing become more important at low rate because the margin for poor delivery is smaller.

Low throughput can be harder than full throughput

A refinery may assume that lower crude rate creates lower corrosion risk because the total acid load is lower.

That conclusion can be wrong.

Low throughput neutralizing amine performance is controlled by distribution as well as total demand.

Pump turndown may be inadequate

The calculated chemical demand can fall below the pump's reliable operating range.

Possible results include:

  • erratic stroke delivery;
  • large dose pulses;
  • poor flow verification;
  • check-valve chatter;
  • loss of calibration accuracy.

Atomization can deteriorate

Lower chemical and carrier flow can produce larger droplets, unstable spray or wall wetting near the injector.

The bulk pH may eventually respond while part of the pipe circumference remains poorly treated.

Vapor mixing can weaken

Reduced velocity can change how droplets and vapors distribute through elbows, tees and exchanger inlets.

A chemically correct dose can still produce an uneven field application.

Residence time can increase

Longer local residence time can increase the opportunity for condensation, deposition or reaction in cold sections.

Control-loop gain can change

A dose adjustment that creates a small pH movement at full rate may create a large movement at low water inventory.

Startup tuning should therefore use smaller approved steps and longer observation windows unless validated high-frequency measurements justify faster action.

Water wash during startup is a hydraulic decision and a chemistry decision

Startup water-wash strategy showing wetting, salt dissolution and continuous water-phase transport

Startup water wash strategy cannot be reduced to “turn on early” or “use the normal percentage.”

The wash must reach the right surfaces with enough water to dissolve soluble salts without creating unacceptable quench, maldistribution, erosion, flooding or unstable operation.

Define the purpose of the wash in each state

During one state, the purpose may be to wet a vulnerable exchanger inlet.

During another, it may be to dissolve salts already present.

During a third, it may be to establish a continuous water phase for transport.

The required flow and injection pattern can differ.

Do not assume the normal wash percentage scales linearly

At low overhead flow, a fixed percentage can produce a water quantity below nozzle distribution requirements.

A fixed absolute rate may provide better coverage but create excessive cooling.

The procedure should define both:

  • minimum hydraulic coverage;
  • maximum process impact.

Verify water quality

Startup wash water may come from a temporary or recently restored source.

Review:

  • chloride;
  • ammonia;
  • amine contamination;
  • dissolved oxygen;
  • pH;
  • solids;
  • temperature;
  • compatibility with the chemical program.

A contaminated wash can add the species the startup plan is trying to control.

Coordinate neutralizer and wash timing

Adding neutralizer without effective water can increase salt inventory on dry surfaces.

Adding water before adequate neutralization can create highly acidic first condensate.

The correct sequence depends on the modeled and measured unit conditions.

The layered roles of caustic, neutralizer, water wash and filming inhibitor are described in the refinery overhead layered corrosion-control guide.

Do not let startup pH become a false master variable

pH is essential, but startup makes pH easier to misinterpret.

The sample may represent the wrong time

A sample collected at 10:00 may represent liquid that entered the sample system several minutes earlier. If the unit was ramping quickly, the process state may already be different.

The sample may represent the wrong location

Accumulator pH describes accumulator water.

It does not directly describe the first wet surface, an upper exchanger inlet or a poorly washed branch.

The water inventory may be too small

A small amount of condensed water can produce an unstable or unrepresentative result. Sampling may disturb the local inventory.

Multiple bases may be present

Residual neutralizer, ammonia, tramp amines and startup chemicals can all contribute to the measured value.

The article on tramp amines and H2S scavenger contamination explains why the measured base inventory can differ from the neutralizer pump rate.

Use a startup evidence panel

Startup evidence panel tracking pH, chloride, neutralizer flow, water wash, pressure and temperature

Instead of one pH target, review a panel that can include:

  • pH trend and data validity;
  • chloride or salt burden;
  • neutralizer verified flow;
  • water-wash verified flow;
  • tower-top temperature and pressure;
  • critical wall temperatures;
  • corrosion monitor response;
  • iron when meaningful;
  • pressure drop;
  • crude and desalter status.

Use state-based control authority

Digital refinery control room coordinating state-based neutralizer control during transient operation

Closed-loop control can be valuable during startup, but it should not automatically receive the same authority it has during validated steady operation.

Monitor mode

The system records measurements and calculations without changing dosage.

This mode is appropriate when sensors are stabilizing or the process has not entered the qualified control state.

Advisory mode

The system recommends a change, but the operator or startup lead approves it.

This creates speed without allowing one transient measurement to drive an uncontrolled dose change.

Bounded automatic mode

Automatic adjustment is allowed only after:

  • required analyzers are valid;
  • process state is recognized;
  • minimum and maximum flows are established;
  • salt and wash limits are satisfied;
  • the response curve is inside the approved range.

Escalation mode

When a boundary is exceeded, the controller stops optimizing and triggers investigation.

The site's guide to online monitoring and closed-loop neutralizer control provides the broader hierarchy; startup adds state recognition and stricter authority limits.

Define the conditions for transferring to normal control

Refinery operator reviewing stability criteria before transferring startup to normal control

The startup program should not end because the unit has reached a scheduled time.

It should end when stability criteria are met.

Possible transfer criteria

  • crude rate inside the defined band;
  • tower-top temperature and pressure stable;
  • overhead water inventory established;
  • water wash stable and verified;
  • neutralizer flow stable and verified;
  • pH inside range for the required duration;
  • no critical corrosion escalation;
  • salt-margin requirement satisfied;
  • sample and analyzer agreement acceptable;
  • no unresolved mechanical abnormality.

Transfer should be recorded so later event analysis can distinguish startup control from normal control.

Crude transitions during restart can change the startup basis

The first crude after a turnaround or shutdown may not match the historical startup crude.

Inventory availability, logistics and economics can place a different blend into the unit.

Review the actual feed, not the planned feed

Before acid precursor entry, confirm:

  • tank identity;
  • blend composition;
  • raw salt;
  • water and solids;
  • TAN where relevant;
  • H2S treatment history;
  • tramp-amine risk;
  • desalter readiness.

Update chemical demand and salt cases

A startup neutralizer rate approved for one crude should not be treated as universal.

The site's guide to neutralizing amine selection for changing crude slates explains why crude variability can change acid burden, base inventory and salt risk.

Use temporary crude restrictions when uncertainty is high

The refinery may choose to:

  • limit the difficult crude percentage;
  • delay a tank switch;
  • hold throughput;
  • increase wash or monitoring;
  • require additional laboratory confirmation.

The purpose is not to eliminate every variable. It is to keep the startup inside an understood response envelope.

Startup after a turnaround requires a clean-system assumption check

A freshly opened and cleaned overhead system may behave differently from the pre-turnaround unit.

Deposits may have been removed

Removing deposits can improve heat transfer and pressure drop, but it can also remove surfaces that previously absorbed or buffered chemistry.

Equipment geometry may have changed

New exchanger bundles, piping modifications, nozzles, quills, insulation or drains can change condensation and distribution.

Instruments may be new or relocated

A new analyzer may not reproduce the old trend because the sample location, conditioning system or response time is different.

Residual cleaning chemicals may remain

Water, alkaline cleaners, acids, passivation chemicals or debris can temporarily alter startup samples.

Do not force new data to match old expectations

Use validated reference methods and inspect the complete event sequence before changing chemical rates merely to recreate a historical number.

Shutdown is not startup in reverse

Crude distillation unit shutdown and preservation procedure highlighting residual corrosion and deposit risk

Crude unit shutdown corrosion requires its own state map.

During shutdown, total flow falls while equipment surfaces cool. Residual acid, salts, water and hydrocarbons remain in the system. Drains and low points become more important. Chemical and wash systems may be stopped while the equipment is still passing through corrosive temperatures.

State A — Controlled rate reduction

As throughput decreases, reassess pump turndown, atomization and wash distribution.

Do not assume the normal dose can simply be reduced in proportion to crude rate.

State B — Feed removal and residual vapor processing

Acidic and basic species can continue moving after fresh crude feed stops.

The neutralizer stop point should therefore be linked to chemistry and flow conditions, not only to feed closure.

State C — Cooling through condensation zones

Cooling can create new wet regions and dissolve existing salts into concentrated local water.

This period can expose low points, deadlegs and exchanger sections that were not the controlling locations during normal operation.

State D — Wash, flush and drain

Water addition may be needed to remove soluble deposits, but it should be engineered for flow path, drainage and corrosion control.

Water that cannot drain can become a long-term corrosion source during standby.

State E — Standby preservation

Define whether equipment will remain wet, dry, inerted, chemically preserved or open for maintenance.

The preservation plan should be compatible with metallurgy, expected duration, future inspection and restart preparation.

Decide when neutralizer should stop during shutdown

Stopping too early can expose late condensate to acidity.

Stopping too late can leave concentrated amine or salts in stagnant equipment.

Use stop permissives

Possible conditions include:

  • acid precursor flow below the defined threshold;
  • carrier or process flow no longer adequate for distribution;
  • water-wash phase complete;
  • equipment drained or flushed as required;
  • shutdown chemistry verified;
  • preservation plan active.

Document retained inventory

Record chemical remaining in:

  • tank;
  • pump;
  • injection line;
  • quill;
  • temporary hoses;
  • flush system.

This prevents unknown chemical slugs during the next restart.

Steam-out and hot circulation can relocate contamination

Maintenance preparation can move salts and chemicals into places where they were not previously concentrated.

Steam can heat and transport deposits

Soluble or partially soluble material can migrate with condensate.

Condensate can collect in low points

If drains are blocked, closed or poorly positioned, acidic or salt-rich water can remain against metal.

Hot circulation can change partitioning

Residual amines and acids can redistribute between hydrocarbon and water phases as temperature changes.

Sampling during cleaning service needs interpretation

A high or low pH result during steam-out may reflect cleaning chemistry rather than the normal overhead corrosion mechanism.

The procedure should identify which results are operational controls and which are only diagnostic observations.

Use event-based troubleshooting when startup pH does not respond

Refinery control room responding to a critical low-pH alarm during crude-unit startup

A common reaction to low startup pH is to increase neutralizer immediately.

That response may be correct, but it should follow a rapid diagnostic branch.

Branch 1 — Is the measurement valid?

Check:

  • sample flow;
  • sample temperature;
  • analyzer calibration;
  • reference sample;
  • timestamp alignment.

Branch 2 — Is the chemical reaching the process?

Check:

  • tank inventory;
  • pump operation;
  • line blockage;
  • check valve;
  • carrier flow;
  • injector condition.

Branch 3 — Did acid demand change?

Check:

  • actual crude;
  • desalter performance;
  • chloride;
  • water wash;
  • temperature and pressure;
  • recycled streams.

Branch 4 — Is another base or salt mechanism present?

Review ammonia, tramp amines, residual chemicals and salt-point risk.

Branch 5 — Is the vulnerable location upstream of the sample?

A normal accumulator pH cannot clear an upstream salt or first-condensate problem.

The structured logic in the crude overhead low-pH root-cause guide is especially useful during startup because the sequence of changes matters more than the final alarm.

Create startup and shutdown stop rules

A transient procedure needs boundaries that prevent the team from continuing simply because the schedule is under pressure.

Possible startup hold triggers

  • neutralizer delivery cannot be verified;
  • water wash unavailable when required;
  • critical analyzer invalid;
  • pH below the approved limit for the defined duration;
  • salt-margin limit violated;
  • critical wall temperature below the approved boundary;
  • corrosion monitor escalating rapidly;
  • unexpected pressure-drop growth;
  • actual crude outside the startup basis.

Possible shutdown hold triggers

  • equipment cannot drain;
  • wash or flush path is blocked;
  • stagnant acidic water identified;
  • chemical inventory cannot be isolated;
  • preservation condition cannot be established;
  • unexpected deposit release occurs.

Define who can release the hold

Release authority should be named in advance.

The decision may require operations, process engineering, corrosion engineering and maintenance input rather than one individual acting under schedule pressure.

Build a transient operating matrix

Operating State Primary Risk Required Evidence Control Posture
Mechanical readiness Chemical system unavailable when acid entry begins Pump, line, injector, wash and analyzer checks No progression without readiness sign-off
Dry heat-up Local chemical accumulation and poor transport Minimum carrier and process flow Manual or monitor mode
Initial feed Unknown acid and background amine loading Actual crude, desalter and chloride information Verified initial dose with strict bounds
First condensation Highly acidic first water and salt-before-water risk Dew-point path, wall temperature, wash readiness Highest monitoring intensity
Low-rate wet operation Poor atomization, maldistribution and unstable pH response Verified flow, sample validity and corrosion trends Advisory or bounded control
Rate ramp Moving chemical demand and condensation location Trend alignment and operating-state recognition Controlled step changes
Stable operation Normal process variability Transfer criteria complete Normal approved control
Shutdown cooling New wet zones, concentrated deposits and stagnant water Drain, flush, wash and chemistry status State-specific manual control
Standby Retained water, oxygen ingress and preservation failure Preservation inspection and documentation No routine neutralizer control

Turn each startup into a learning cycle

Industrial startup learning cycle for planning, checking and updating corrosion-control procedures

A restart should produce more than a successful return to production.

It should improve the next startup.

Preserve the event timeline

Record:

  • state transitions;
  • feed changes;
  • chemical start and stop times;
  • verified flow;
  • wash events;
  • analyzer validity;
  • pH and chloride response;
  • corrosion indicators;
  • operator actions;
  • holds and releases.

Compare predicted and actual boundaries

Did water appear where expected?

Did pH respond within the expected delay?

Did the pump remain controllable?

Did the water wash provide the intended coverage?

Did any cold point or branch behave differently?

Update the startup basis

Changes may include:

  • new initial dose;
  • revised chemical-start permissive;
  • new hold point;
  • additional temperature measurement;
  • different sample frequency;
  • pump or injector modification;
  • revised crude restriction.

Use management of change when the procedure changes

A successful improvisation should not become an undocumented tradition.

Convert proven learning into approved procedures, training, equipment records and alarm limits.

The best transient program protects the sequence, not only the endpoint

Normal operation is often judged by whether the unit reaches target throughput with acceptable pH, corrosion and fouling performance.

Transient operation must be judged differently.

The endpoint can look successful even if equipment passed through a damaging condition on the way there.

A startup can finish with pH in range after the first condensate already attacked a cold section.

A shutdown can finish with the unit safely offline while salt-rich water remains trapped in a low point.

A restart can achieve full rate while a poorly atomized chemical deposit begins forming near the injector.

Therefore, the best overhead corrosion during restart program verifies the sequence:

readiness → transport → acid entry → first condensation → wet low-rate control → rate ramp → stable transfer.

The best shutdown program verifies another sequence:

rate reduction → residual chemistry control → cooling → wash or flush → drain → preservation.

Neutralizer is one important tool inside those sequences.

It cannot compensate for missing water, poor distribution, invalid measurements, contaminated wash water, incorrect crude assumptions or uncontrolled cooling.

The mature refinery does not ask the chemical program to rescue an undefined transient.

It defines the transient, assigns evidence to each state and allows the neutralizer to operate only inside conditions where delivery, chemistry and monitoring can be trusted.

Focused FAQ

When should neutralizer injection begin during crude-unit startup?

Refinery startup neutralizer control should begin according to approved state-based permissives, not only a clock time. The process should have sufficient transport or carrier flow, verified injection-system readiness and a credible expectation of acid precursor entry before the chemical is introduced.

Why is startup overhead corrosion risk different from normal operation?

During startup, temperature, pressure, vapor flow, water condensation, chemical delivery and sample response change at different speeds. The first wet metal surface may be upstream of the normal sample point, and the first condensate can be more acidic than later bulk water.

Can the normal neutralizer dosage be scaled directly with crude rate?

Not always. Low throughput neutralizing amine performance can be limited by pump turndown, pulse delivery, atomization, carrier flow and vapor mixing. A lower total acid load does not guarantee uniform chemical distribution.

Why is water wash difficult during startup?

Startup water wash strategy must balance salt dissolution and surface coverage against quench, maldistribution, erosion and flooding. Normal full-rate wash ratios may not provide the right hydraulic behavior at low overhead flow.

Is accumulator pH enough to control startup corrosion?

No. Accumulator pH is important, but it describes sampled downstream water. It may not represent first condensate, an upstream salt-deposition zone, a cold branch or a poorly washed exchanger inlet. Startup decisions should use a broader evidence panel.

What is the most dangerous startup period?

The approach to first effective water condensation is often a critical period because a small water inventory can absorb a concentrated acid load. The actual risk depends on unit design, temperature path, chloride loading, background bases, water wash and chemical distribution.

Can closed-loop neutralizer control be used during startup?

Yes, but authority should be state-based. Early startup may require monitor or advisory mode. Bounded automatic control should begin only after analyzers, process state, chemical delivery, water wash and operating limits are validated.

What causes low pH during startup even when the neutralizer pump is running?

Possible causes include invalid sampling, an unprimed or blocked injection line, inadequate carrier flow, poor atomization, higher-than-expected chloride, desalter problems, contaminated crude, missing water wash or an upstream acidic zone not represented by the sample.

When can startup control transfer to normal operation?

Transfer should occur after defined stability criteria are met, such as stable throughput, temperature and pressure; verified wash and neutralizer flow; valid measurements; acceptable pH duration; adequate salt margin; and no unresolved corrosion or mechanical warning.

Why is shutdown not simply startup in reverse?

Crude unit shutdown corrosion involves cooling equipment, residual acids and bases, existing deposits, draining limitations and new wet zones. Chemical stop timing, flushing, drainage and preservation create a different sequence from startup.

When should neutralizer injection stop during shutdown?

The stop point should be linked to residual acid flow, distribution capability, wash or flush completion and the preservation plan. Stopping too early can expose late condensate to acidity, while stopping too late can leave concentrated chemical or salts in stagnant equipment.

How should a refinery investigate corrosion after restart?

Reconstruct the timeline across feed, temperature, water condensation, chemical flow, wash flow, sampling, pH, chloride and corrosion indicators. Overhead corrosion during restart is best investigated by identifying which variable changed first rather than focusing only on the final alarm.

What should be included in a neutralizer commissioning checklist?

A neutralizer commissioning checklist should cover product identity, storage condition, pump calibration, line priming, injector condition, carrier flow, water-wash readiness, analyzer validation, laboratory coverage, startup dose, operating limits, hold points and fallback actions.

How can startup experience improve future corrosion control?

Preserve the event timeline, compare predicted and actual condensation and chemical-response behavior, identify blind zones, update limits and use formal management of change to revise procedures, training, monitoring and hardware.

#RefineryStartup #NeutralizerControl #CrudeUnit #OverheadCorrosion #RefineryShutdown #LowRateOperation #WaterWash #AmineSalt #ProcessChemistry #CorrosionControl #RefineryRestart #AssetIntegrity