Fan blade damage and a deterioration trend on a high-cycle 737-800
A Boeing 737-800 on a high-cycle regional network is borescoped during a transit at Hong Kong after the crew reports slight vibration. The CFM56-7B fan shows one blade with a 4 mm trailing-edge tear at the root fillet and several leading-edge nicks. The AMM permits blending up to 0.8 mm depth for certain fan blade damage, with reprofiling and re-inspection at 500 cycles, but the blend allowance is location-specific and the root fillet area is excluded. Trend data shows EGT margin down from 22 degrees C to 14 degrees C over the last 350 cycles, fuel flow up 1.4 per cent, vibration at 2.3 units against a 4.0 limit, and clean magnetic chip detectors. A spare engine is nine hours away, the aircraft has 14 sectors in the next three days, and the shift supervisor asks whether to blend and release or keep the aircraft on the ground.
1. Blend or remove? Show the decision logic and the data you would use.
Read the worked answer
The tear cannot be blended, so the question becomes how to restore a compliant blade.
My first move is to read the limit as written. The blend allowance is location-specific: damage at the trailing-edge root fillet sits in the highest-stress region of the blade, where the manual excludes blending because material removal there concentrates stress and can initiate a crack under centrifugal and vibratory loading. A generic 0.8 mm allowance elsewhere on the aerofoil does not transfer to that location. Blending a crack-like feature at a fillet to make it disappear is the same category of error as blending fuselage skin to make a reading look better.
The nicks may be blendable if they fall inside the permitted zones, but that is a separate assessment, blade by blade.
The performance trend is a second, independent issue. Loss of 8 degrees C EGT margin in 350 cycles is faster than a healthy fleet rate, and the vibration is elevated but below limit. That points to deterioration elsewhere, possibly in the high-pressure compressor or turbine, or to fan imbalance. Blending one blade changes the fan balance state and could make vibration worse if the moment weight is not restored. With clean chip detectors there is no metal distress signal, but the trend still argues for a borescope of the compressor and turbine stages before I would sign the engine back to a 14-sector programme.
Decision: no blend at the fillet, no release. Options are a matched replacement blade if the AMM permits single-blade replacement with correct moment weight, or an engine change.
2. Suppose the fillet damage is confirmed and an approved route allows blade replacement on wing. What documentation, checks and limitations would you impose before release?
Read the worked answer
Data before work. Confirm the AMM task reference, the blade part number, and the moment weight of the removed and replacement blades, since fan blades are matched sets. If the replacement cannot be matched within the manual's limit, the set is affected and the job escalates.
Workmanship controls. Root fillet and dovetail cleanliness, lubrication, correct seating, torque and locking, and a recorded visual and dimensional check of the adjacent blades. Any other blade found outside limits is dispositioned against the manual, not against the disposition I would like.
Post-work verification. An engine run or taxi check with vibration survey, comparison against the pre-work figure, and a check that the EGT indication and fuel flow are consistent with the trend. Confirm no new exceedance and that the engine monitoring programme has been updated with a manual reference for the blade change.
Limitations and records. The e-tech log and engine record must show the blade serial number, moment weight, and a clear statement that the fillet damage was removed with the blade, not blended. If the manual imposes a re-inspection or a reduced interval, it goes into the maintenance programme with a due date, not a note in the log.
Operational briefing. Tell the crew the vibration baseline and that an increasing trend should be reported, and set an internal trigger: any vibration rise above the previous reading or a further EGT margin loss at the next check returns the engine for review. The trend, not just the blade, is the thing being managed.
3. What evidence would change your decision towards releasing the aircraft, and what would push you the other way?
Read the worked answer
Towards release: the manual demonstrably permits the specific damage location and depth; the blend can be done within limits with a defined reprofile and inspection interval; a matched replacement blade is available with correct moment weight; the compressor and turbine borescope is clean; the EGT trend is explained, for example by a known fleet-wide effect or a recent wash, and is consistent with the fleet average for that cycle count; and the vibration returns to the pre-event baseline after the work.
Away from release: any manual exclusion at the damage location; evidence of crack growth such as a dye-penetrant indication at the fillet after blending; a trend that continues after blade work; vibration that rises after replacement; oil analysis showing elevated iron or nickel; or the discovery of similar tears on other blades, which suggests a foreign object or a fleet-wide issue and shifts the problem to the operator's reliability programme and the type certificate holder.
The pressure test. The presence of 14 sectors and a distant spare engine is not evidence. The decision rests on approved data and measured condition. I would also resist the framing that the choice is blend-and-release or ground the aircraft: a third option, replacing a blade on wing from stock, may satisfy both airworthiness and schedule, and it is my job to put that on the table before anyone reaches for a blend tool.