Author; Brett Weintz. Updated 21 April 2021.
What is the procedure for ship propeller and intermediate shaft alignment?
If a propeller shaft special survey or stern tube bearing repair is planned with a shipyard, then bearing load measurements with the vessel afloat before entering dry dock provides valuable information for decisions regarding any alignment adjustments.
The objective of ship propeller shaft alignment measurement is a satisfactory load distribution between all, and across each, bearing(s) supporting the complete propulsion shaftline. Optimised bearing loads is a sound basis for maximising bearing life (reliability). This is particularly important for the sterntube aft bearing.
Ship sterntube/ A-bracket bearings are important given the combination of bearing length and shaft bending generated by the mass/ hydrodynamic propulsive forces acting on the propeller. A key objective of measurements conducted in support of a docking involving work on the propulsion alignment is to determine the loading conditions on the sterntube aft bearing.
You can find the following Brabon Engineering Services site pages and case studies regarding ship propulsion shaft bearings:
- Shaft alignment measurements Brabon Engineering Services undertake (this page).
- Shaft alignment measurement details discussion page. Includes factors affecting derived results/ uncertainty.
- Shaft alignment principles discussion page.
- Shaft alignment measurements case study.
- Strain gauge shaft alignment measurement case study.
- Sterntube aft bearing machining offset accuracy case study.
- Fair curve alignment design example.
- Sterntube bearing problem case study 1.
- Sterntube bearing problem case study 2.
- Ship sterntube aft bearing operations and problems discussion page.
- Hydrodynamic bearing design and damage discussion page.
- Ship vibration (shaft dynamics) discussion page.
How Brabon Engineering Services check ship propeller shaft alignment?
- Bearing loads by shaft jack-up measurements using strain gauge load cell and displacement transducer with computer data logging. Automated analysis of jacking data.
- Main engine, main bearing loads by crankshaft jack-up measurements.
- Bearing loads by strain gauge shaft bending measurements. Advanced analysis techniques are applied to derive bearing loads beyond the accessible spans.
- Bore alignment measurements (in the sterntube or A-bracket) using the Taylor-Hobson micro-alignment telescope and targets. A bearing damage survey and assessment of likely cause(s) is typically offered with measurements.
- Engine crank web deflections.
- Alignment/ offset measurements of elastic couplings and cardan shafts. Gap and sag of uncoupled (intermediate) shaft sections (preliminary alignment).
- Assessment of intermediate shaft and propeller shaft straightness.
- Modelling, design calculations and bearing load assessment/ interpretations as well as bearing adjustment (calculations and implementation).
Bearing load by shaft jacking measurement
Brabon Engineering Services can undertake bearing jacking measurements to derive the shaft static/ stopped bearing loads. Innovative data logging and analysis software is used to provide bearing loads immediately after the measurement. There is no delay until the next day.
Measurement involves a hydraulic jack, a strain gauge load cell (sensor) and a displacement transducer for accuracy, see Figure 1 below. The sensors are connected to an analog-to-digital device (strain digitizer) for simultaneous measurement and data display/ logging on a computer.
The shaft is raised and lowered through the bearing clearance. A plot of the measured load and lift is examined to determine the jack load at zero lift, i.e. as if the bearing was not present, and hence the shaft load that would be borne at the jack location, see example Figure 2 below. Multiple jacking measurements are generally undertaken to give a more accurate averaged value of the load.
At the sterntube forward bearing as well as bearings adjacent to a reduction gearing thrust collar, the jacking measurement needs to be conducted especially slowly for accurate results. This is to allow for displacement of oil around the shaft journal clearance and drag of the thrust collar/ bearing respectively.
Brabon Engineering Services also apply the shaft jacking process to derive the main engine, main bearing loads. The temporary support beam used for main bearing inspection/ replacement is used for the measurement. Care is needed in the jacking diagram assessment due to unloading of adjacent bearings. Similarly, the jack correction factor applied is dependent on the loaded condition of the adjacent bearings.
Bearing loads by shaft bending strain gauge measurement
Brabon Engineering Services can derive propulsion shaft alignment bearing loads by undertaking strain gauge bending measurement on a shaftline. This provides bearing loads for the shaft stopped/ static condition (as with bearing jacking). With innovative analysis software the derived bearing load results are available a short while after the measurement.
Shaft bending strain is measured using electrical resistance strain gauges at a number of axial locations in the accessible spans of the shaft. Each station has two strain gauges bonded to the shaft in diametrically opposite positions and connected in a half Wheatstone bridge to a strain meter. While half bridges provide satisfactory results, full bridges can be used for greater accuracy. Static bending strain readings (with the shaft stationary) are recorded at 90° rotational intervals as the shaft is progressively turned in the ahead direction using the turning gear. This provides bending strain in the vertical and horizontal planes. Rotating the strain gauges (shaft) through 180° provides the (total) bending strain in the shaft, rather than deviations in strains.
The shaftline static vertical bearing loads are derived by comparing the bending moments calculated from the strain measurements with the bending moments and bearing reaction forces calculated from a mathematical model of a reference alignment condition. The derived bearing loads are, thus, dependent on the calculated weight of the modeled complete shaftline assembly. As bearing loads from strain measurements are independent of the jacked loads, the two sets of results are generally correlated/ compared.
The shaftline static transverse bearing loads are obtained directly from the strain measurements recorded with the gauges in the horizontal plane.
Brabon Engineering Services can use the bending strain with other measurements to derive loads at additional shaftline bearings (beyond those accessible for jacking measurements). This includes the load distribution between the aft and forward ends of the stern tube aft bearing as well as the propulsion reduction gearing main wheel aft to forward bearing load bias/ distribution (if applicable). Note that various combinations of sterntube aft bearing load bias/ distribution and shaftline bearing offsets can generate the same loads at the stern tube forward bearing and intermediate bearing. That is, the combination of jacked loads do not correlate to one unique shaft alignment condition. Thus, bending strain measurement can provide important additional information of the shaft alignment condition.
Calculation methods involving ‘reverse engineering’ and/ or matrix inversion are sometimes employed by other organizations. However, results stated for all the shaftline bearings may be unreliable as the measured bending in the accessible spans typically provides insufficient data for a specific solution (all bearing loads). For example, a deviation in shaft bending adjacent to the sterntube may be generated by a load/ offset change at either the stern tube aft bearing (either end) or the stern tube forward bearing.
I have undertaken propulsion shaft alignment measurement projects on numerous vessels. This has included new construction as well as dry docking propeller shaft stern tube bearing replacement/ repairs. A successful shaft alignment project requires an engineer having practical understanding as well as academic knowledge. That is, practical understanding of shaftline characteristics and problems as well as academic knowledge to undertake calculations for assessments and adjustments. I also have the perseverance and grit to identify and assist in returning the vessel to service in minimum time.
Stern tube bore alignment measurement
Brabon Engineering Services can undertake (propeller shaft bearing) bore alignment (offset) measurements, as well as assessments and any recommended bearing offset adjustments.
Experience shows the Taylor-Hobson micro-alignment telescope and targets is superior as a complete measurement system to laser alignment. A reference line is established between the centres of the aft and forward stern seal recesses located at each end of the stern tube, see Figure 3 below. Offsets of sterntube/ A-bracket bearing (surface) centres are measured at various axial locations relative to the reference line. Offsets are measured with the vertical/ horizontal direction micrometers of the alignment telescope. Post-processing image analysis/ scaling assures an accuracy of better than 0.01mm.
Offsets are measured at multiple axial locations through the sterntube aft bearing as well as the housing (after bearing extraction) in order to derive the relative slope of both items. An adverse (large) slope mis-match between the bearing and propeller shaft journal due to a forward-down slope of the bearing housing (relative to the reference line), can result in stern tube aft bearing damage.
Offsets of the bearing housings are also measured (after the necessary bearing extraction). This allows for the machining offsets on the replacement bearing to compensate for any error in the slope of the housing as well as the slope of the propeller shaft aft journal. A forward-down slope error of the stern tube aft bearing housing can arise during shipyard vessel construction due to sag in the boring bar if the housing machining operation is conducted with the sternframe block horizontal (machined sternframe casting construction) or due to sag in a ‘flexitube’ before being resin chocked.
Offsets of bearing/ shaft centres forward of the sterntube can also be measured, e.g. intermediate bearings and main engine output flange. However, the results need to be carefully considered due to hull deformations when the vessel is supported on dry dock blocks (which are typically unknown and/ or not quantified).
Ship shaft alignment measurements – case study
The case involved an 57,000 tonne DWT bulk carrier having a diesel direct-drive, single screw propulsion plant. Main propulsion was by a 6-cylinder, two-stroke engine having an mcr of about 9,400kW at 127rpm. The vessel was in the final stages of reassembly in a shipyard following replacement of the stern tube aft bearing. The original sterntube aft bearing had been damaged as a result of the propeller shaft becoming fouled by a rope. A problem arose with the shaft alignment.
The shaftline comprised a propeller shaft 6.5m long x 500mm in diameter, intermediate shaft 6.0m long x 420mm in diameter and engine crankshaft. The shaftline was supported by an oil lubricated, 1,000mm long, whitemetal lined, stern tube aft bearing, intermediate bearing and main engine, main bearings (eight off). There was no sterntube forward bearing fitted.
Shipyard alignment work
During shaftline reassembly, the main engine service technician had directed the shipyard to adjust the intermediate bearing upwards. This has apparently been in order to centre the propeller shaft in the stern tube forward seal location recess. However, this was not based on any calculations or the original design. Subsequent bearing jacking measurements by the shipyard found the main engine, aftmost main bearing was unloaded. However, there were no measurement or calculations of the load distribution on the sterntube aft bearing. In addition, there had been no bearing load or bore alignment measurements with the damaged stern tube bearing, i.e. to understand the original alignment condition. Note that given the intermediate bearing had been raised, a corresponding decrease in the main engine, aftmost main bearing load would naturally be anticipated.
The main engine service technician had then suggested the engine should be realigned to the shafting so as to correct the engine bearing load distribution. This would be a significant task with costs in the order of USD 1M (not including the additional lost charter time). Preparation for the task involves releasing the main engine holding-down bolts, raising the engine about 5mm, breaking and removing the existing chock, then lowering the engine to the original position. There follows an iterative process of alignment measurements and repositioning the engine with the engine being chocked once a satisfactory alignment is achieved. The vessel Managers were concerned about the potential escalation in scope of work and sought a second opinion.
It was suggested that the intermediate bearing was reinstated at its original vertical offset. Bearing load measurements by the strain gauge and shaft jacking methods were then conducted. From the bending strains it was found there was a heavy-aft load distribution on the stern tube aft bearing. (This would have been significantly greater with the upwards adjustment.) The bearing jacking results showed a 4.1 tonne load on the intermediate bearing (which correlated with the load derived from the bending strains) and a 2.1 tonne load on the main engine, aftmost main bearing. Main engine crank web deflections were satisfactory. The vessel was in a ballast condition with the engine cool.
A further adjustment downwards of the intermediate bearing was suggested to reduce the heavy-aft load bias on the stern tube aft bearing without excessively increasing the main engine, aftmost main bearing load. Subsequent repeat bearing load measurements following the sea trial found that it had been possible to satisfy the alignment requirements for the sterntube aft bearing and the main engine without realignment of the engine.
In engineering, details are the crux of the situation.
Shaft alignment is not a complex process, but interpretation of data and assessments require care and experience as well as understanding of all factors affecting the indicated results.
Brabon Engineering Services can help by conducting meticulous shaft alignment measurements and assessments.
Call for a discussion: +353 87 383 5043
email for a proposal: email@example.com
Send a message via our contact page
Connect on LinkedIn
Engine crank web deflection measurements
Brabon Engineering Services can undertake crank web deflection measurements using the latest technology. Crank web deflections show the vertical and horizontal angular difference between each adjacent main journal by the variations in distance between crank web faces as the shaft is rotated. Normal convention is for positive deflection with an opening crank web gap as well as positive curvature with sagging in the vertical plane and (sagging) towards the exhaust side.
Engine deflections vary with engine/ foundation temperature and main bearing wear. Suspected main bearing wear should be confirmed by measurements by depth/ bridge gauge or Swedish feeler gauges. Deflections of the aftmost three crank webs are also affected by the combined alignment condition of the crankshaft and connected shaftline, i.e. mainly the (forward) intermediate bearing. It is also possible to apply the Hudson method (1974) and estimate the crankshaft offset curve from the measured deflections.
Flexible coupling alignment measurements
Brabon Engineering Services can undertake measurements of flange-to-flange alignment across flexible couplings/ cardan shafts. This may be required to confirm that the coupling is subject to less than the maximum allowable misalignment. Flexible couplings typically have a specified maximum continuous radial and angular misalignment, e.g. to limit the heating of the elastic elements, as well as a (greater) maximum allowable transient misalignment to allow for starting, stopping and dynamic events.
Measurements should be conducted after the machinery has been running and the housing/ foundations have attained the normal temperature gradient. Note that horizontally offset gearing units have thermal growth in the horizontal and vertical directions. Only measurements in the cool and warm conditions can confirm if the horizontal direction thermal growth is centred around the main wheel, or near (one of) the input pinion(s). If the shaftline includes a gearing set, then ideally any input clutch should be engaged. There is generally no need to remove the flexible coupling if the system is assembled. Both shafts should be rotated together.
Measurements should involve the shaftline being continuously rotated through a full revolution in the ahead direction, i.e. without stopping. Measurements involving rotation through only 180° may be liable to errors. For the angular misalignment measurement, two dial indicators should be used in order to eliminate errors due to axial displacement of one, or both, shaft during the revolution. It is also generally prudent to conduct a repeat measurement of the aftmost crank web in the astern direction to quantify any deviation due to shaft deflection by forces from the turning gear as well as to confirm repeatability of the results.
Shaft straightness measurements
Examples of scenarios where a propeller shaft or intermediate shaft can become bent include:
- A casualty event, such as the vessel grounding or the propeller encountering an underwater obstruction.
- The propeller becoming fouled and stopped abruptly by a rope.
- Asymmetric heating of a shaft journal as typically associated with a bearing failure.
- A shaft may also naturally develop a slight bend over time due to relaxation of forging stresses with normal operating temperature cycling.
- An apparent bend can also be the result of angular misalignment being ‘locked’ into a joint when a tapered sleeve coupling is tightened.
- Note that an otherwise straight shaft directly coupled to an engine will exhibit a run-out due to the mass of running gear of the aftmost cylinder and the variation in crankshaft stiffness (as a shaft beam) with rotational angle.
A total run-out (TIR) of say 0.20mm (0.008″) in an intermediate shaft span would likely be within normal engineering tolerances. The total run-out of the propeller shaft taper/ flange adjacent to the outboard end of the stern tube/ A-bracket should be less than say 0.08mm (0.003″) given the relatively short free length from the bearing.
A significant shaft bend can have potential adverse effects on the bearings as well as machinery coupled to the shafting, e.g. reduction gearing. While the associated load variation on the bearing(s) typically will not cause bearing damage, a bend could cause an (alternating) misalignment and edge loading between the shaft journal and the bearing. For example, a bend in the propeller shaft is unlikely to affect the propeller operation, but given the sterntube aft bearing length, a large bend may result in bearing damage around the aft end 6 o’clock position, particularly during manoeuvring. Where a shaft is coupled to reduction gearing a bend may cause an adverse variation in the tooth contact pattern, e.g. contact area oscillating between the axial centre and the edge of the teeth.
Dismantling the shaftline and transferring the suspect shaft section to a workshop involves significant disruption and cost. Alternatively, it is possible to investigate the extent of bend as well as any consequential effects to inform any decision to take the vessel out of service. The amplitude of the shaft bend can be identified by measurement using a dial indicator at a series of axial locations. The shape of the bend can be found by marking the maxima and minima points relative to a datum, say one of the coupling bolts. This can show if the bend is in one plane, or in a spiral. If it is suspected that the shaft has a complex bend, e.g. in two planes or a spiral, then two dial indicators at each measurement location arranged 90° apart can show if the shaft progression is a circle or other orbit shape, e.g. Lissajous figure.
Bearing jacking measurements at multiple rotational positions of the shaft can quantify the load variation at the bearing(s). Using the load variations it is also possible to estimate the bend in the shaft. That is, as if the assembled shaft was free of constraint by the bearings.
If a direct drive, diesel engine is suspected to have a problem, then crank web deflection measurements should be conducted to confirm satisfactory condition. These could be compared to results of previous normal periodic condition monitoring measurements. In severe cases, intermediate bearing running surfaces and foundations should be inspected. Where reduction gearing is involved, tooth contact checks can be made using the turning gear and soft blue applied at say four rotational positions of the main wheel. The tooth contact under load can be checked using hard lacquer applied in four sets of bands (minimum) around the main wheel circumference. Inspections of lacquer removal patterns are then conducted after a period of normal running.
It should be noted that for an intermediate shaft rigidly connected to a diesel engine crankshaft there will be shaft bending as well as load variations versus rotational position at the intermediate bearing and the engine main bearings that are inherently due to the crankshaft shape and running gear masses. The associated load variations can be significant, e.g. more than 10% of the mean load. Accordingly, care needs to be taken with analysis.
Assessment of ship shaft alignment measurement results
(How to compare measured propeller shaft bearing loads against the optimum alignment?)
Assessment of a propulsion shaft alignment condition is by consideration of the derived bearing load measurement results versus calculated/ design values. Sterntube/ A-bracket bearing bore alignment results are generally examined in a shaftline system mathematical model to estimate, then assess, the load distribution on the stern tube aft bearing. Any assessment should include observations from the sterntube aft bearing visual examination (direct knowledge of bearing condition is of great value).
Bearing loads that are within 20% of design values for the given vessel condition are typically assessed as satisfactory. Bearing load results with the ship afloat are compared against design. Measurements conducted in dry dock tend to be used for comparison of before versus after reassembly. For slow-speed, diesel direct drive propulsion, the aft-most main bearing of the main engine may register a negligible load when measured at the light ballast and machinery cool condition. However, this may be acceptable given the particular class of vessel where the aft main bearing load tends to be sensitive to vessel draught aft and engine temperature.