Design Concerns Propulsion Shafting Alignment

8/19/2019 Design Concerns Propulsion Shafting Alignment

http://slidepdf.com/reader/full/design-concerns-propulsion-shafting-alignment 1/14

ABS TECHNICAL PAPERS 2003

Design Concerns in Propulsion Shafting Alignment 1

DESIGN CONCERNS IN PROPULSION

SHAFTING ALIGNMENT

Davor Sverko,

Associate Member, American Bureau of Shipping

ABSTRACT

Propulsion shafting alignment of the modern merchant vessels is highly sensitive to small disturbances in bearing offset.This sensitivity creates difficulties in calculating alignment and ensuring the alignment is conducted as designed. It istherefore important that the shaft alignment analysis provides the necessary (accurate and applicable) data to support

each stage of the actual alignment process. The alignment procedure itself is carried out in several stages of the vesselconstruction, from the sighting through before the shafting installation, through the Sag and Gap prior to the shaftingassembly, and to the final verification by measurements after the construction completion. When analyses are not

conducted properly the alignment procedure may not comply with analytical requirements. The following design relatedshaft alignment problems are specifically addressed in this article: the tail shaft bearing problems, the consequences of

intermediate shaft bearing offset adjustment, crankshaft modeling, bearing clearance, gear meshes, hull deflections,alignment optimization, and the alignment acceptability criteria.

Key Words: Shaft alignment, Hull deflections, Diesel engine bedplate sag, Slope boring, Prescribed displacements.

1. INTRODUCTION

The frequency of shaft alignment related bearing damageshas increased significantly in recent years. The alignment

related damages may be attributed to the design of thevessel, shipyard practices, lack of regulations, andinadequate analyses.

Figure 1Directly coupled propulsion shafting – example

Propulsion systems of the modern vessels (Figure 1) aremostly diesel engine driven direct-coupled installations,the design of which results in increased disparity between

structural flexibility of the hull and the shafting. Ingeneral, ship hulls have become more flexible due toscantling optimization and increased ship’s length, while

the demand for higher powers has resulted in largerdiameter; i.e. stiffer shafting systems. Consequently, thealignment of the propulsion system becomes moresensitive to hull deflections, resulting in difficulties inanalyzing the alignment and conducting the alignment procedure.

Alignment procedure is not uniformly applied in theindustry. The procedure is dependent on shipbuilders’

practices and experiences. Some shipyards have lowconfidence in alignment analysis, considering it more as a

reference than as a document to follow. They rather rely

on their experience and confidence in their ability torectify potential problems as they arise. However, thismay not be an acceptable procedure from the point of

view of the Regulatory Bodies.

Analyses do not always represent propulsion systems

accurately, and may not always provide sufficientinformation to ensure “error free” alignment procedure.The intention of this paper is to particularly address problems that may arise from incompatibility between theanalyses and the alignment procedures.

2. ANALYSIS VS. PROCEDURE

The shaft alignment is a static procedure where applied

loads are static forces and moments. Dynamic factors arenot normally considered because the primary purpose ofthe analysis is to support an alignment procedure that is

conducted under static conditions.

It is important to know the alignment condition as

accurately as possible. However, if vessel is not in the drydock it is difficult to establish a reference line againstwhich alignment condition may be verified. Furthermore,

measurements often do not provide complete informationneeded for alignment verification.

The mismatch between the alignment analysis and the procedure itself starts at the very beginning of thealignment process. While the vessel is in the dry dock, the

alignment condition is easily controlled; the vessel is oneven keel, and the sighting trough and bearing pre-

positioning is conducted very accurately. However, oncethe vessel is afloat, the control over procedure’s accuracy




2 Design Concerns in Propulsion Shafting Alignment

may be completely lost. The problem is that hull

deflections are very difficult to predict. Consequently,without knowledge of offset change with hull deflections,the alignment process may not be verified with the desired

accuracy.

In order to protect diesel engine from damages due to possible inadequate alignment, engine designers normallyrequire application of the engine’s bedplate sagging after

vessel is afloat. The bedplate sagging is supposed torectify and diminish hull deflection influence on theengine alignment, as it primarily annihilates deflections

which occur as vessel afloat, and further it reducesinfluence of hull’s hogging as it get loaded. However, thisresults in inconsistent alignment procedure. Theestablished dry dock referenced line is now changed only

in the section below the main engine (M/E), and the restof the propulsion system remains affected by afloatedvessel hull deflections. The shafting and the engine are

now aligned to the different base lines (one which weknew initially which is defined in the dry dock, and the

other one, essentially unknown), established for enginerealignment and bedplate sagging after the vessel islaunched. The consequence is the shafting affected by hull

deflections on one side, and the M/E on the other siderectified to the even keel condition.

What essentially happened is: we gained control over theengine alignment (engine reference line is now known), but we practically lost possibility of accurate control of

the alignment condition of the shafting as well as the sterntube bearing.

The solution to the analysis-procedure inconsistency problem may be in conducting the whole shaftingalignment procedure with vessel in the dry dock where we

can accurately verify the procedure against the analysis.In order to align the system in the drydock, we must beable to:

• estimate hull deflections, as the alignment needs to besatisfactory with vessel afloat,

• define the optimal set of prescribed bearingdisplacements to ensure a robust alignment which isrelatively unaffected by hull deflections when vessel

is afloat,• conduct the whole alignment procedure in the dry

dock.

In order to allow some provision for corrections, the

shaftline bearings as well as the diesel engine, or gearboxshould not be chocked until bearing condition issatisfactory in vessel’s afloat condition.

Below, we addressed some of the most importantalignment-design related issues on several alignment

sensitive installations. A very large crude oil carriers(VLCC) with directly coupled 6-cylinder diesel engine isused in Sections 3, 4 and 5 to investigate the following

issues:

• tail shaft bearing problems,

• consequences of the intermediate shaft bearing

offset adjustment,

• crankshaft modeling effect on accuracy of the

analysis,

• bearing clearance,

Gear meshes are investigated in Section 8 on a turbine

driven VLCC with directly coupled gearbox. Further, inSection 9 we evaluated the effect of the hull deflectionson a large container vessel. We also addressed the

alignment optimization, and alignment acceptabilitycriteria in Sections 9 and 10.

The analyses presented below are all conducted applying

the ABS shaft alignment software.

3. TAIL SHAFT BEARING

Tail shaft bearing contact is very much dependent on theapproach taken in the shafting alignment design. Thelarger the area of static contact between the shaft and the

bearing, the smaller the contact stress exerted on the bearing from the shaft, and the faster the oil film willdevelop; having as a consequence extended bearing life.

In the analyses below, we first compared three alignmentdesigns in an example with diesel engine driven VLCC

installation, and synthetic tail shaft bearings. Next, weaddressed different approaches in bearing contact

modeling, i.e. single point contact vs. two-point contact between the bearing and the shaft.

Figure 2 below compares the impact of each of the three

different alignment designs on the bearing-shaft contactarea:

Zero Offset Alignment : is provided for reference. Zerooffset presumes that all bearings’ bottom shellsare positioned on a straight line. Results of this

analysis are not satisfactory, as the forward sterntube bearing is fully unloaded. However, the aftstern tube bearing loading condition is very good

with relatively large area of contact between theshaft journal and the bearing.

Positive Offset Alignment : although it results in acceptable bearing reactions, the contact area between the bearing and the shaft is not very good. Positive

Offset design is the solution where the mainengine and the intermediate shaft bearings areraised above the zero offset line (Figure 2 - row

2, Figure 3). Relative misalignment slope between the shaft and the bearing is estimated to be 0.855 [mrad].

Negative Offset Alignment : is the desired approach. Themain engine and intermediate shaft bearing are





Z E R O

O F F S E T

P O S I T I V

E

O F F S E T

N E G A T I V E O

F F S E T

P R E S C R I B E D D

I S P L A C E M

E N T S

B E A R I N G R

E A C T I

O N S

N O D A L S

L O

P E S

T / S B E A R I N G C

O N T A C T

C O N D I T I O N

F i g u r e 2 T a i l s h a f t b e a r i n g c o n t a c t a s a f u n c t i o n o f a l i g n m e n t d e s i g n







Contact condition evaluation using ABS Tail Shaft Bearing software:

Figure 3 Bearing loading condition - Positive Offset

Figure 4 Bearing loading condition - Negative Offset





4. INTERMEDIATE SHAFT BEARING OFFSET

ADJUSTMENT

In Section 3, several approaches to the alignment

problem are presented. In this section, the solution with Negative-Offset is further investigated for its sensitivity

to changes in the intermediate shaft bearing offset(Figure 5 and Figure 6).

It is a normal practice in shipyards to adjust theintermediate shaft bearing offset when alignment doesnot comply with calculated values. Often the problem is

main engine bearing loading condition that can berectified by changing the vertical offset of theintermediate bearing.

Although, adjusting the intermediate shaft bearingoffset may help in resolving the engine bearing problem, it may eventually adversely affect the

condition of the bearings located aft of the intermediate bearing. The misalignment condition of the after stern

tube bearing supporting the propeller is of particularconcern here. Vessels with short, stiff shafts and single

lineshaft bearings such as VLCC’s, ULCC’s and large bulk carriers are particularly affected.The problem is investigated under two different propulsion shafting designs:

• with forward stern tube bearing,

• without forward stern tube bearing.

The adjustment of intermediate bearing offset of 0.1,0.2, 0.5 and 1.0 [mm] upward and downward, from the

initially prescribed baseline, is investigated. The offsetchange influence on the following bearings isevaluated:

• diesel engine aft-most bearing (M/E Brg. 1) -reaction change,

• diesel engine second aft-most bearing (M/EBrg. 2) - reaction change,

• tail shaft bearing - change in misalignmentslope.

The bearing offset change is plotted on the x-axis, and the two y-axes contain information on engine bearing reaction change and tail shaft bearingmisalignment.

Figure 5 System sensitivity to intermediate bearing offset change – with forward stern tube bearing

System With Forward S/T Bearing (Figure 5): by adjusting the intermediate shaft bearing offset we achieved desiredinfluence on the M/E Brgs. 1 and 2, and did not significantly affect the tail shaft bearing slope. We note from the Figure 5 that aft S/T bearing misalignment angle increases relatively little as the intermediate bearing is beinglowered. By increasing the offset at the bearing, the misalignment slope improves at S/T bearing as long as the

shaft maintains contact with the forward S/T bearing – afterwards the installation behaves as without forward S/T bearing.





Figure 6System sensitivity to intermediate bearing offset change – No forward S/T bearing

System With No Forward S/T Bearing (Figure 6): by adjusting the intermediate shaft bearing offset, a significant

influence on M/E Brgs. 1 and 2 is achieved, as well as much higher sensitivity (relative to the previous case) of thetail shaft bearing slope. Reason for this higher sensitivity is a different load distribution among the bearings. Themisalignment angle change at aft S/T bearing follows linearly the change in intermediate bearing offset. The

misalignment angle is reduced as the intermediate bearing is lowered down, and increases with offset increase atintermediate bearing.

The solution with the forward S/T bearing (Figure 5)

may be preferred from the point of view of sensitivityof the aft stern tube bearing misalignment angle changedue to the adjustment of intermediate bearing offset.

Otherwise, larger flexibility (e.g. no forward S/T bearing) is desired as the hull deflections have lessereffect on the propulsion system.

In the above statement, we shall distinguish between thesingle bearing offset adjustment and the effect of the

hull flexibility. The hull deflections affect all bearings

in the system simultaneously and is therefore preferableto have more flexible system (no forward stern tube

bearing is beneficial). However, we see that the exactopposite is the case with single bearing adjustment(intermediate bearing in this case), where stern tube

bearing misalignment, in systems with no forward sterntube bearing, is quite sensitive to offset change.

In this section we investigated only intermediate shaft bearing offset change by simulation of the bearingadjustment as it may be conducted by the yard. As the

intermediate bearing offset varies, the remaining bearings are held in their original position.The results are as expected, in that varying the

intermediate shaft bearing offset has more effect on the

after stern tube bearing misalignment when the forward

S/T bearing is removed.

In Figure 5 and Figure 6 we discuss the effects of

intermediate lineshaft bearing offset on themisalignment-slope curve. In Figure 6 the slope-curveis almost a straight line, connecting the solutions

between plus 1[mm] and minus 1[mm] offset change.This is so because forward stern tube bearing isremoved; therefore, no sudden change in slope occurs.

The forward stern tube bearing is unloaded with larger

offset increase on intermediate bearing (Figure 5).

There is some controversy in the above statements, as both of the alignment designs, with and without forwardstern tube bearing, have their advantages and their

drawbacks. Eventually, the decision has to be made bythe yard and the owner. If shipyard is confident inobtaining good alignment without significantlyamending the intermediate shaft bearing offset, then thesolution without forward stern tube bearing should be preferred; and result will be a shafting design less

sensitive to hull deflections. However, if shipyard mayexpect difficulties, such as unloaded M/E bearings, itmay be safer to opt for solution with forward stern tube

bearing.





5. CRANKSHAFT MODELING

It is often seen that shaft alignment analysis contains

diesel engine crankshaft model which does notrepresent whole crankshaft but only few aftmost

bearings. The reduced crankshaft model may beinappropriate as it may result in errors related to Sagand Gap procedure, and incorrect bearing reactions.Also, when reduced crankshaft models are used it is

obvious that designer’s concern was only shafting andnot M/E at all.

The Sag and Gap is a procedure conducted prior to theshafting assembly. It consists of measurements of sagsand gaps between mating flanges, and verification of

the system’s compliance with calculated data. However,if Sag and Gap is erroneously defined, the yard will not be able to comply with requirements withoutreadjusting offset of one or more bearings. By doing so,

the system’s alignment will be wrongly conducted fromthe very beginning with very small possibility of

rectifying it.

The partial crankshaft modeling may therefore result in

incorrect prediction of the crankshaft aft-flangesagging. The calculation accuracy may be particularlyaffected when the model includes less then three

bearings, and when load acting on the flange andaftmost crankpins are not taken into account.

Reduced crankshaftmodelSag and gap @ int-M/Eflange:

SAG = -1.2616Gap1 = -1.2228Gap2 = 0.0166

Gap1+Gap2 = 1.2394(below flange)

Extended modelcrankshaft

Sag and gap @ int-M/Eflange:SAG = -1.2301

Gap1 = -1.2228

Gap2 = 0.0832Gap1+Gap2 = 1.3060

(below flange)

Differences in estimated sags (0.03 [mm]), and gaps (0.067 [mm]) are appreciable.

Figure 7

Crankshaft modeling affecting alignment results

Actual Sag and Gap values are expected to be similar tothe results of the second model with extended

crankshaft. However, if the crew conducting thealignment is given the data as per the first model (whichis reduced to two bearings only), the alignment, which

may actually be correct, will be readjusted to theincorrect Sag and Gap values. Consequently the bearing

reactions would be impossible to verify against analysisas well.







Figure 9Gear driven propulsion - even gear-shaft bearing reactions – 0.21 [mrad] gear misalignment angle

9. HULL DEFLECTIONS AND ALIGNMENT

OPTIMIZATION

Figure 10Containership - Hull deflection - Still Water Condition

The container vessel example points out that even in ahighly robust shafting system that is almost insensitive

to the hull deflections, alignment problems may still be present on the diesel engine side. With a detailedcrankshaft model we can see that the diesel enginealignment may not be satisfactory unless action is takento rectify engine bedplate hogging. Failure to do so mayresult in unloaded M/E bearings.

Since the diesel engine bedplate hogging is a result ofhull deflections, we can alleviate its effect by

introducing sag in the bedplate during the constructionstage prior to chock casting.

It is important to select appropriate set of displacementsto ensure satisfactory bearing loading condition. Byensuring robust static alignment, we can expect trouble

free dynamic operation of the shafting; i.e. lateralvibration (whirling) may be expected to result inacceptable response, and the operating condition of the

bearings (in particular tail shaft bearing) will result in

prolonged life due to larger contact areas and faster oilfilm development.

The ABS optimization program is based on a genetic-algorithm method where a solution is sought by parallel

search throughout solution space bounded by two“extreme” deflection curves (e.g. estimated hulldeflections). Within the defined solution space, we

extract a number of acceptable solutions that complywith basic alignment requirements; e.g. as proposed byClassification Societies. It is then up to the designer to

select the solution that provides the most robust design.

The subject alignment condition is evaluated forcalculated hull deflections (Figure 10). The bearingoffset is optimized within given ballast and laden hull

deflection curves (Figure row No. 1). Two solutionsare shown on Figure 1:

• hull deflections and no engine prescribed sag,

• hull deflections with engine prescribed sag.

Without accounting for engine sag (Figure 1 row 3), the

second aftmost engine bearing, in loaded vesselcondition, is unloaded and solution is not acceptable.When engine bedplate sag is considered the solution isacceptable (Figure row 4).

The optimization program provides a number ofsolutions that fulfill requirements for bearing reactions.

For the particular case above the single selected“optimal” solution results in relatively low load at

aftmost M/E bearing in ballast condition, with tendencyof load to increase as the vessel is loaded (Figure 1rows 3 and 4).

Detailed explanation of the ABS optimization program

is forthcoming in the paper to be presented at theSNAME Propeller/Shafting 2003 conference. Below weshow the tabular output that the optimization routine

provides:

Optimization results: selected solutions from pool of 10 solutions





Optimization with Genetic Algorithm

Generation: 368 String: 15 FITNESS: 1.100000

| SUPPORT REACTIONS | Total Total GA Max Hull Min Hull Thermal Engine

| Ry[0] delRy Ry Ry Ry | Offset Offset defined Deflect. Deflect. Offset Sag.

Sup. Node | (Max.Offs) (Min.Offs) (dy) | Max. Min. dy

No No | [kN] [kN] [kN] [kN] [kN] | [mm] [mm] [mm] [mm] [mm] [mm] [mm]

-------------------------------------------------------------------------------------------------------------------------------

1 < 8> | 1551.999 -12.141 1361.512 1508.647 1539.858| 0.000 | 0.000 | 0.000 | 0.000 | 0.000 | 0.000 | 0.000

2 < 16> | 25.312 66.294 333.733 168.626 91.606| 15.792 | -0.284 | -1.208 | 17.000 | 0.924 | 0.000 | 0.000

3 < 32> | 468.801 -121.674 327.546 267.662 347.127| 32.980 | -0.122 | -2.068 | 35.048 | 1.946 | 0.000 | 0.0004 < 46> | 400.431 52.500 383.705 493.101 452.931| 47.233 | 12.739 | 5.156 | 42.077 | 7.583 | 0.000 | 0.000

5 < 59> | 415.428 93.890 523.455 491.690 509.318| 53.391 | 18.341 | 6.483 | 46.908 | 11.858 | 0.000 | 0.000

6 < 75> | 524.187 -151.565 424.382 405.641 372.622| 46.238 | 13.776 | -0.586 | 46.824 | 14.362 | 0.000 | 0.000

7 < 83> | 466.488 106.232 313.877 455.114 572.720| 41.568 | 12.959 | -0.721 | 42.289 | 13.680 | 0.000 | 0.000

8 < 93> | 108.899 -633.691 818.386 74.578 -524.792| 37.272 | 12.277 | -0.064 | 37.086 | 12.091 | 0.250 | 0.000

9 < 97> | 351.846 897.741 129.934 745.827 1249.586| 36.191 | 12.036 | 0.090 | 35.901 | 11.746 | 0.250 | -0.050

10 < 99> | 448.377 -360.808 103.794 97.216 87.569| 33.829 | 11.277 | 0.095 | 33.604 | 11.052 | 0.250 | -0.120

11 <101> | 440.149 76.614 521.842 523.506 516.763| 31.305 | 10.448 | 0.099 | 31.136 | 10.279 | 0.250 | -0.180

12 <103> | 442.363 -16.038 382.937 400.952 426.326| 28.685 | 9.580 | 0.104 | 28.561 | 9.456 | 0.250 | -0.230

13 <105> | 439.699 2.270 469.346 455.382 441.970| 26.031 | 8.709 | 0.109 | 25.942 | 8.620 | 0.250 | -0.270

14 <107> | 450.588 2.888 421.072 439.039 453.476| 23.323 | 7.824 | 0.114 | 23.249 | 7.750 | 0.250 | -0.290

15 <109> | 286.737 -5.655 310.031 299.005 281.082| 20.567 | 6.925 | 0.118 | 20.499 | 6.857 | 0.250 | -0.300

16 <115> | 286.768 4.260 302.505 287.139 291.028| 18.134 | 6.131 | 0.123 | 18.061 | 6.058 | 0.250 | -0.300

17 <117> | 450.581 -1.352 425.887 445.622 449.229| 15.297 | 5.208 | 0.127 | 15.200 | 5.111 | 0.250 | -0.280

18 <119> | 439.793 0.268 443.315 438.535 440.061| 12.430 | 4.280 | 0.132 | 12.298 | 4.148 | 0.250 | -0.250

19 <121> | 441.910 0.033 427.477 433.437 441.943| 9.530 | 3.348 | 0.137 | 9.353 | 3.171 | 0.250 | -0.210

20 <123> | 442.283 -0.432 439.987 445.976 441.851| 6.585 | 2.405 | 0.142 | 6.343 | 2.163 | 0.250 | -0.150

21 <125> | 438.334 0.667 501.878 462.098 439.001| 3.551 | 1.426 | 0.147 | 3.234 | 1.109 | 0.250 | -0.080

22 <127> | 378.189 -0.302 332.560 360.368 377.887| 0.401 | 0.401 | 0.151 | 0.000 | 0.000 | 0.250 | 0.000





Ballast Laden

1 Whole system (y axis scale 0-20 [mm]) Whole system (y axis scale 0-60 [mm])

2 Engine only (y axis scale 0-1 [mm]) Engine only (y axis scale 0-2 [mm])

3

Hull deflections Hull deflections

4

Hull deflections and engine bedplate sag Hull deflections and engine bedplate sag

Figure 11Containership - Diesel engine bearing reactions as a function of hull deflections and bedplate sag





10. ACCEPTABILITY CRITERIA

The alignment procedure is not always uniformly appliedas it depends on shipbuilder’s practices and experiences.

Low confidence that shipyards seem to have in alignmentanalysis may explain why some shipyards insist on high

acceptability margins. These margins are sometimes ashigh as 50%, essentially rendering the analysis useless.The large difference between allowed and calculated

values may not result in acceptable loading of the bearings. This is not acceptable.

With such high tolerances the shipyard is depending moreon intuition than on analysis. Caution should be exercisedin defining these tolerances; there is a difference betweeneducated guesses (where higher tolerance may be

explained by previous experiences, and by controlling theoutcome of the measurements), and dependence on puretrial and error to obtain good alignment. This is where

regulatory bodies may come into the picture by definingclear and definitive guidelines and criteria for alignment

acceptance.

The regulatory requirements are a trade off between the

necessity to regulate the industry and ensure the safety ofthe vessels on the one hand, and a desire not to burden theindustry with unnecessary requirements on the other.

11. CONCLUSION

Good static alignment is a prerequisite for acceptable

dynamic behavior of the propulsion system. In order toobtain good alignment, we should be able to controloutcome of the alignment analysis, and provide reliable

data to the crew that is conducting the alignment.

The design related shaft alignment problems that are

particularly considered in this article are: the tail shaft bearing problems, the consequences of intermediate shaft bearing offset adjustment, crankshaft modeling, bearing

clearance, gear meshes, hull deflections, alignmentoptimization, and the alignment acceptability criteria. Ifthese issues are not taken into account during the

analytical stage, serious consequences may be expected inthe final alignment condition.

We have also concluded, that under ideal conditions, the

solution to the problem of alignment accuracy would bein completing the alignment procedure while the vessel is

in the dry dock when hull deflections are known and the bearing offsets optimized,.

However, hull deflections are rarely known with suchaccuracy. Therefore, a more practical solution may be totake advantage of the dry dock condition by completing

the alignment but making provisions to correct bearing positions when the reactions are verified afloat. For thatto be possible, chocking the main engine, gearbox, and

line shaft bearings should be postponed until bearingreactions are accepted.

The conclusions in this paper are primarily applicable toVLCC’s and large bulk carrier vessels, where all aspectsof alignment (i.e. stern tube bearing, line shaft bearings,

gearbox and diesel engine) are very sensitive to smalldisturbances in bearing offset. In contrast, most largecontainer vessels have more flexible systems; though,

they may still experience problems on the diesel engineside.



Documents

Design Concerns Propulsion Shafting Alignment