Rotec AI

Tools

Bearing SelectorPick the right bearing, step by stepVerify a BearingLife, safety factor and fitsLoad CalculatorLoads from your machineWorked templatesFan, gear and shaft load templatesCross-ReferenceSame size from other makersBearing AutopsyNewUpload a photo, get the likely causesFailure AnalysisRead damage, find the causeLubricationGreases, oils, cleanliness, oil lifeEquipment AnatomyMotors, pumps, fans, gearboxesFAQ and quick checksSearch the basics in plain languageFind a DistributorSuppliers by brand and city

Learn

Free calculatorsLife, grease, oil, loads and moreFind a bearingSize, rating and equivalents by designationFailure modesDamage types, causes and fixesKnowledge libraryLubrication and reliability articlesBearings by machineMotors, pumps, mills, kilns and moreHow-to guidesBearing life, grease, failures, numbersField casesFailure investigations and monitoring cases (licence)Lubrication guidesGreases, oil systems, oil life and costQuick answersPlain-language answers to common questions

More

PricingPartnershipsContact
Case lessons

A key and keyway unbalance: when chamfer size changed the vibration

A high-speed coupling hub test pointed at unbalance. The real source was a few grams of missing material in the keyway.

The machine

A steam turbine drove a compressor through a gearbox, around 3300 hp over 6,370 to 7,495 rpm, with hydraulically fitted hubs on the compressor side and single-key fits elsewhere. The gearbox maker found high vibration on a solo run, added field balance weights to finish the test, and shipped it on. Later in the string test the turbine showed 18 micrometre and the gear 29 micrometre vibration.

The wrong suspect

Everyone assumed the coupling was unbalanced. Hubs were checked on mandrels with dummy keys. The first results looked high: 408 g-mm at one hub and 479 g-mm at the other, against limits of 9.8 and 13.25 g-mm per plane in the standard. The dummy keys, however, did not represent the mass of the missing material. Recalculated with a 3D model, correct dummy keys brought both hubs within the limits.

The real source

The installed key had been turned down to the shaft diameter but its chamfer was almost four times too big. A chamfer of 3 mm instead of the 0.6 to 0.8 mm recommended removes about 7 g from the key. At the keyway radius of about 56 mm that is roughly 392 g-mm of unbalance, more than the whole allowable balance of the hub.

ItemValue
Unbalance from key chamfer difference392 g-mm
Mass of the difference7 g
Chamfer used vs recommendedabout 3 mm vs 0.6 to 0.8 mm
Vibration after correct keyabout 13 micrometre, same as solo run of the turbine

Why it matters

Hubs are balanced before the keyway is cut, so the key has to replace the removed material accurately. The heavy turbine rotor hid the same error; the lighter gear pinion was sensitive. A rotor analysis only predicts the response if the amount and location of unbalance fed into it are right.

Takeaway. When a coupling looks unbalanced and checks fine on the mandrel, look at the key: a few grams at the keyway radius can exceed the whole balance tolerance.
Put it to work in the tool
All knowledge articlesBearing failure modesGrease selector