A damaged fourth-stage blade row does not subtract one-ninth of this turbine's output. The number that matters is the enthalpy converted by that row, followed by the loss in that conversion and the way the altered flow redistributes pressure and loading through the remaining stages. On this nine-stage, 6,000 rpm generator drive, the first stage is Curtis and stages 2 through 9 are reported as approximately 50% reaction. That distinction moves the screening estimate away from a simple impulse-stage assumption.
Aerodynamic loss appears as reduced shaft output and additional irreversible heating, while missing, cracked, or distorted material creates mechanical loads that an efficiency calculation cannot qualify. This is heat and load redistribution, not logic: repair selection starts with steam-path integrity and only then moves to lost kilowatts.
Wrong fixes and their consequences
| Proposed action | Why it looks attractive | Why it can fail |
|---|---|---|
| Infer the loss as one stage out of nine | It provides an immediate percentage. | Stages do not absorb equal enthalpy. The Curtis stage and exhaust-end stages can carry disproportionate energy, and downstream stages recover some available energy through the reheat effect. |
| Use the impulse-stage estimate without correction | A cleaned-up impulse row was estimated at a 5% to 20% stage-efficiency reduction and a 0.25% to 1.5% total-performance effect. | Stages 2 through 9 are reported as about 50% reaction. A moving-row defect has a larger effect when appreciable pressure drop occurs across that row. |
| Remove only the damaged rotating row | It eliminates suspect moving blades. | The retained stage-four nozzle restriction no longer feeds a functioning rotor row, creating an unfavorable pressure and velocity distribution. |
| Remove the rotating row and diaphragm as a simple bypass | Removing both can produce less aerodynamic loss than retaining the diaphragm alone. | Stage five has a larger nozzle area than stage four. The reduced downstream restriction can increase stage-three pressure drop, blade loading, and diaphragm deflection. |
| Dress visible damage and return to service | Profile cleanup can preserve more performance than deleting a row. | Surface appearance does not establish remaining section, crack status, attachment integrity, balance, or fatigue life. |
| Rely on an open-valve test | It is a familiar capacity comparison. | The steam supply is below turbine rating, instrumentation is limited, and process-driven passout changes the heat balance. Those conditions can conceal a small efficiency change. |
Energy distribution and reaction-stage sensitivity
An impulse stage develops most of its pressure drop in the stationary nozzles; the moving blades primarily turn and decelerate the resulting high-velocity jet. A reaction stage divides pressure drop between the stationary and moving rows. Damage to a reaction moving row therefore disrupts both velocity turning and expansion inside the rotor passage.
The reported arrangement is a Curtis first stage followed by stages that are approximately 50% reaction. The provisional damaged-row identification is stage four and must be confirmed during inspection. For a reasonably restored impulse row, a 5% to 20% reduction in that stage's efficiency was proposed as an initial screening range, corresponding to roughly 0.25% to 1.5% overall performance loss. That case is useful as a lower-sensitivity comparison, not as the selected estimate for the reported stage-four design.
For an approximately 50% reaction stage with blades that cannot be cleaned to a satisfactory profile, the stated screening range is 2% to 3% of total output. Applied arithmetically, that equals 64.5 to 96.75 kW at the 3,225 kW(e) MCR point, or 41.7 to 62.55 kW at the 2,085 kW(e) MER point. At a nominal 3,000 kW operating point, the same range is 60 to 90 kW. These are output-loss screens; they are not blade-stress limits or guarantees of repaired performance.
Quantities, limits, and measurement locations
| Quantity | Reported value or case | Why it matters | Where to read or establish it |
|---|---|---|---|
| Rated electrical output | MCR 3225 kW(e) |
Reference point for the 2% to 3% screening calculation | Nameplate and generator records |
| Alternate electrical output rating | MER 2085 kW(e) |
Second stated reference point; the acronym's plant-specific meaning should be taken from site documentation | Nameplate and operating manual |
| Turbine speed | 6000 rpm |
Sets centrifugal loading and makes mass or profile asymmetry a mechanical concern | Speed instrumentation and turbine documentation |
| Inlet steam | 415 psig at 390 °C (734 °F) |
Defines the admission state for a heat balance | Calibrated inlet pressure and temperature instruments |
| Exhaust vacuum | 28.05 inHg |
Backpressure changes available expansion and output | Condenser vacuum measurement, with the plant's reference convention recorded |
| Passout condition | Details not reported | Flow, pressure, temperature, and extraction location determine how much energy leaves before the exhaust | Passout flowmeter and pressure/temperature instruments |
| Damaged stage | Provisionally stage four | Stage location and reaction govern loss and load redistribution | Physical stage identification against turbine drawings |
| Stage-four output screen | About 2% to 3% for the reported reaction case and poor profile recovery | Provides an economic comparison among repair choices | Calculated from a valid operating baseline or a turbine heat-balance model |
Damage mechanism and inspection boundary
Severe damage isolated to a middle stage demands an explanation before repair. A foreign object can circulate between stator and rotor, damage moving blades, strike stationary-vane trailing edges, and bend or close nozzle passages. Restricted passages then alter local mass flow, incidence, reaction, and circumferential loading even if adjacent rotating rows initially look intact.
Inspect beyond the visibly damaged row. Confirm the stage number, map every damaged blade, and examine stage-four stationary-vane trailing edges for deformation or closed passages. Extend the inspection upstream and downstream to find impact marks, displaced material, rubbing, deposits, erosion, loose components, and a credible entry path. Check rotor attachments, diaphragm condition, clearances, and evidence of contact. A qualified turbine repair organization should select the examination methods for the blade material, attachment geometry, and accessible surfaces.
Profile restoration has aerodynamic priorities, but the priorities depend on stage type. For an impulse rotor, the leading edge and convex surface are especially influential. In the reported reaction section, both expansion and turning occur through the moving passage, so fixed- and moving-row throat condition, surface contour, trailing edges, and passage-to-passage uniformity all enter the decision.
Repair decision path
New blades provide the strongest path toward recovering both geometry and mechanical margin. OEM-supplied blades are the primary choice; accurately engineered replacements are the secondary choice when the original supply path is unavailable. Either route still requires material, attachment, dimensional, balance, and clearance acceptance appropriate to the machine.
Dressing may retain better performance than removing a row when enough sound material remains to restore a smooth, repeatable profile. Its acceptance cannot rest on predicted efficiency. The engineering disposition must address remaining blade section, crack indications, attachment condition, frequency and stress implications, rotor balance, and variation among blades.
If the rotating row must be removed, evaluate the stationary diaphragm as part of the same flow-path change. Removing the diaphragm can reduce the performance penalty relative to retaining its restriction, but stage five's larger nozzle area exposes stage three to a greater pressure drop. The assessment must calculate or otherwise qualify stage-three moving-blade load, stage-three diaphragm load and downstream deflection, axial thrust effects, clearances, and the smaller load increases expected across stages five through nine.
Do not return the turbine to service on an efficiency estimate alone. Blade release, diaphragm contact, or overload can turn a modest kilowatt penalty into major rotor and casing damage.
Engineering procedure
- Freeze the operating reference. Record inlet pressure and temperature, exhaust vacuum, speed, generator output, control-valve position, and every available passout pressure, temperature, and flow. Mark unavailable or unreliable measurements rather than substituting assumed values.
- Confirm the steam-path arrangement. Match physical rows to drawings and verify that the damaged moving row is stage four. Confirm the Curtis first stage and the approximate reaction assigned to the later stages.
- Establish the damage cause and extent. Inspect the damaged row, stage-four stationary-vane trailing edges, neighboring rows, diaphragms, attachments, seals, and potential foreign-object paths. Account for missing material.
- Screen the performance penalty. Use the 2% to 3% total-output range only for the stated case: an approximately 50% reaction stage with poor profile recovery. Use 0.25% to 1.5% only as the separate impulse-stage comparison tied to a 5% to 20% stage-efficiency reduction.
- Develop repair configurations. Compare replacement blades, qualified dressing, rotating-row removal with the diaphragm retained, and rotating-row removal with the diaphragm removed. Reject configurations that fail mechanical-integrity or load-redistribution checks before comparing efficiency.
- Model the selected flow path. Use stage pressures, nozzle areas, reaction, clearances, passout location, and steam conditions from drawings and measurements. For a deletion case, focus on the increased stage-three expansion and loading caused by flow into the larger stage-five nozzle area.
- Set acceptance criteria before assembly. Define profile and throat tolerances, crack acceptance, attachment condition, rotor balance, clearances, diaphragm position, and operating surveillance with the responsible turbine engineer or repair organization.
- Commission against a repeatable point. Stabilize inlet state, exhaust vacuum, passout demand, speed, and generator conditions as closely as the process permits. Trend thermal and mechanical behavior through the permitted loading range.
Performance and mechanical verification
A small output change cannot be separated from normal process variation by generator kilowatts alone. Passout is the dominant missing variable: a change in extracted mass flow or enthalpy can move electrical output without any change in stage efficiency. If the passout point lies downstream of the damage and remains superheated, compare passout temperature at matched inlet, exhaust, flow, and load conditions before and after the damage or repair. A temperature shift can expose changed expansion through the damaged section.
Build the best available heat balance from measured inlet flow and state, passout flow and state, exhaust condition, and electrical output. Where instrumentation cannot support that balance, use repeated matched-condition trends and label the result as a relative comparison. An open-valve test below the turbine's steam-supply rating cannot demonstrate full capacity.
Mechanical verification is separate from the heat balance. During run-up and loading, track vibration, phase behavior where available, bearing temperatures, axial position or thrust indication, speed stability, casing behavior, abnormal noise, and evidence of rubbing. Compare each signal with the machine's documented baseline and trip or alarm criteria; no generic numerical threshold can replace those machine-specific limits.
Frequently asked questions
How do I estimate output loss from one damaged turbine stage?
Identify the stage type and its share of expansion first. For this approximately 50% reaction stage with poor profile recovery, use 2% to 3% of total output as a screening range: 64.5 to 96.75 kW at 3,225 kW(e).
How do I decide between dressing and replacing damaged blades?
Choose replacement when remaining section, cracking, attachment integrity, balance, or fatigue margin cannot be accepted. Dressing is an aerodynamic option only when inspection and engineering analysis also qualify the blade mechanically.
How do I check whether the stationary vanes were damaged?
Inspect the stage-four vane trailing edges, throats, and passage-to-passage opening for impact deformation or closure. Also trace missing material and inspect adjacent rows because a foreign object can circulate between the stator and rotor.
How do I verify a small efficiency loss in a passout turbine?
Compare matched inlet state, exhaust vacuum, passout flow and state, speed, and electrical load. If passout downstream of the damage remains superheated, its temperature history can help identify a changed expansion pattern.
When should I stop the repair assessment and escalate?
Stop when blade integrity, missing-material location, stage-three overload after row deletion, diaphragm deflection, balance, or the damage mechanism remains unresolved. Escalate through the OEM or documented successor's official support channel, or engage a qualified turbine consultant before authorizing assembly or operation.