DISABLED TANKERS Report of Studies on Ship Drift and Towage Oil Companies International Marine Forum Foreword Tanker owners and operators have become increasingly concerned about the problems associated with the drifting and towing of disabled tankers. A review of casualties involving freely drifting disabled tankers suggested that there is much to be learned about the behaviour of a disabled tanker under the influence of wind, waves and currents. In order to provide Masters of disabled tankers with information about the most probable drift behaviour of their vessels, research was initiated by a group of companies, members of the Oil Companies International Marine Forum (OCIMF), together with the American Institute of Merchant Shipping (AIMS) and the International Salvage Union (ISU). The British National Maritime Institute (NMI) was commissioned to carry out intensive model testing with free floating and tethered models. A steering group, known as the Disabled Ship Research Group (DSR), made up of marine representa tives of the sponsoring oil companies and of the ISU was appointed to participate in the tests. Their involvement was an essential ingredient of the experiments. To validate the results of the model tests, full-scale drift data were obtained from some vessels, pre dominantly VLCCs, with some smaller tankers. This report presents the results compiled from the experiments and investigations. The raw data on the experiments may be obtained from: The National Maritime Institute, Feltham, Middlesex, England. Introduction Page PART 1 This part describes an experimental and theoretical investigation into the drift of tankers in wind and waves 1 Results from free model tests, constrained model tests and calculations based upon the NMIWAVE computer program are presented and discussed Other data, from the free running and constrained model experiments relating to model tracks and sea-state data, are included as Appendix PART 2 To validate results of the model tests, full-scale drift data were obtained from a number of ships 73 This part presents the results PART 3 This part describes tank test studies to examine the effect of towing forces on a disabled tanker drifting in wind and waves. Following a description of the experimental techniques employed, results from all experiments are presented and discussed. Some conclusions are drawn relating to turning and towing a tanker, subject to the constraints imposed by the conditions of the exercise. 83 3 Contents Page 1 Introduction 5 1.1 Purpose of the Investigation 5 1.2 Details of Models 5 1.3 Synopsis of Experiments 5 2 Experiments with Free Models 6 2.1 Description of Experimental Method 6 2.2 Calculation of Wind Forces and Moments 6 2.3 Arrangement of Test Series 7 2.4 Results: Twin-Screw Model 5481 7 2.4.1 Free-drift experiments 7 2.4.2 Manoeuvring with limited power available 8 2.4.3 Manoeuvring with rudders flying free 8 2.5 Results: Bulbous-Bow Model 5480A 8 2.5.1 Free-drift experiments 9 2.5.2 Manoeuvring with limited power available 10 2.5.3 Manoeuvring with rudder flying free 10 2.5.4 Effect of heel with power available 10 2.5.5 Force to move model out of trough 10 2.5.6 Wind and waves from different directions 10 2.6 Results: Cylindrical-Bow Model 5480B 11 2.6.1 Free-drift experiments 11 2.6.2 Manoeuvring with power available 11 2.6.3 Manoeuvring with rudder flying free 12 2.6.4 Effect of heel with power available 12 2.6.5 Force to move model out of trough 12 2.6.6 Wind and waves from different directions 12 2.7 A Note on Coriolis Effects 12 2.8 Summary of Main Conclusions 14 2.8.1 Free-drift 14 2.8.2 Controlled drift 14 2.8.3 Towing forces to head into weather 14 3 Measurement of Drift Forces in Waves 14 3.1 Experiment Arrangement 14 3.1.1 Introduction 14 3.1.2 Programme of tests 14 3.1.3 Sea-state spectra 15 3.1.4 Mooring arrangements 16 3.1.5 Force measurements 16 3.1.6 Instrumentation 16 3.1.7 Conduct of the tests 16 3.2 Results Obtained 17 3.2.1 Method of presentation 17 3.2.2 Accuracy and repeatability 18 3.2.3 Effect of wave encounter angle ? w 19 3.2.4 Effect of sea-state 19 3.2.5 Effect of rudder 19 3.2.6 Effect of hull shape 20 3.2.7 Effect of trim 20 3.3 Conclusions 20 4 Calculations of Wave Drift Forces 20 4.1 Introduction 20 4.2 Method of Calculation 20 4 Disabled Tankers Page 4.3 Ship Models 21 4.3.1 Model 5480B 21 4.3.2 Modified model with L/B = 4.5 21 4.3.3 Natural periods 22 4.4 Spectral model 22 4.5 Numerical Models 22 4.6 Results 22 4.6.1 Regular waves 22 4.6.2 Irregular waves 23 4.7 Comments on Results 25 4.7.1 Regular waves 25 4.7.2 Irregular waves 25 4.8 Comparisons Between Calculated and Measured Results 26 5 Conclusions 26 6 References 27 92 Disabled Tankers The start of each run is indicated on the computer plots by a ship diagram with a star-shaped symbol amidships. Other positions of the model midships position are denoted by square-shaped symbols which are separated by a time interval of approximately 36 s full scale. These points are illustrated diagrammatically on Fig. 7. 5 Discussion of Results 5.1 Ship Model at Normal Load Draught 5.1.1 Effect of initial headway/sternway Experiments described in Reference 1 showed that with the tow force directed at right angles to the ship centreline plane very large forces would be required to turn the vessel. The forces in question were in fact much larger than the static bollard pull of the largest available ocean-going salvage tug. It was clear therefore that ‘brute force’ towing techniques were unlikely to turn the vessel successfully. It was also known that turning a disabled vessel was made easier if some headway could be obtained which allowed the natural hydrodynamic lift and drag characteristics of the underwater hull to come into play if a small drift angle could be induced by the towing vessel. Tow force at bow. If the vessel is towed ahead and headed slightly into the direction of the turn a sway force toward the centre of the turn is generated by the ship hull together with an initial turning moment which in most ships tends to cause the bow to head into the turn. This was exploited in runs 1 and 2 where it is seen that initial towing angles of about 15° were used, increasing to 50° or 60° once the model was turning. When the model was nearly head-to-weather it was necessary to reduce both towing force and towing angle if the model was to be towed into the weather and not pass through a heading of 000°. However, there is a case where any headway induced on the disabled ship may be a disadvantage and this occurs if the rudder is jammed in the hard-over position and it is necessary to attempt to turn the ship against the direction of the rudder. This is illustrated in run 7 for example, when the rudder was set hard a-port for the duration of the run and an attempt was made to turn the model to starboard. It may be noted that initial tow angles of 60° were used to counter the tendency of the ship to turn to port. Such large towing angles in conjunction with any headway on the ship may provide practical problems for the towing vessel in that if too much headway is induced on the vessel it may over-run the towing vessel which may then have problems of manoeuvring to hold station relative to the disabled ship. Tow force at stern. It is notable from Table 5 that when using a towing force at the stern to turn the model to present its stern to the weather, the values of amax were in general less than those used with a single tow force at the bow. Once again some measure of sternway was found to be useful and the towing forces used were in general somewhat smaller than those used when turning by the bow. Towing forces at bow and stern. The advantage of inducing some headway on the vessel was again demonstrated when towing forces at bow and stern were used. This is illustrated in runs 18, 19 and 20 when for the first two runs the bow and stern forces tended to act against each other to produce small XT and YT values with an appreciable turning moment NT . This, it is known, is ineffective in turning a vessel and it was not until the stern towing force was adjusted to pull ahead as well as to one side (as in run 20) that a turn was successfully accomplished. 5.1.2 Effect of rudder angle Towing force at bow. Turning the model to starboard was made somewhat easier when the rudder was jammed hard-a-starboard, but was made more difficult when the rudder was jammed hard-a-port. This is illustrated in Fig. 8 from which it is clear that a better rate of turn r and a smaller FT max are obtained with rudder hard-a-starboard. This feature is perhaps not surprising, as when the rudder is hard-a-starboard in a turn to starboard it is offering little resistance to the flow passing it at the stern. If the rudder is opposing the turn of the vessel then flow around it is likely to separate and cause a locally high drag. Towing force at stern. When turning the model stern to weather, this time to port, the benefit gained from the rudder being jammed hard-a-starboard was lost as shown in Fig. 8. In this case there were benefits to be gained by having the rudder hard-a-port although the benefits were in general small. This reduction in benefit is probably due to the fact that the drift angle of the ‘leading edge’ of a turning ship is generally smaller than that at the ‘trailing edge’. When the ship is being towed from the stern and turning, the rudder, being at the ‘leading edge’, 93 PART 3 - This Part Describes Tank Test Studies to Examine the Effect Run Heading ? 0 ? f V 0 (knots) T t (min) r (°/min) F H (°) ? H (°) Bow F T max a max (°) Stern F T max a max (°) Power used Turn X max Y max (°) Tow forces DraftTrim d (°) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 171 18 19 20 21 22 23 231 232 233 24 25 252 26 261 27 28 29 30 31 32 321 33 34 351 36 37 38 39 391 40 41 42 421 431 44 451 46 47 481 482 483 49 50 231 260 236 270 254 264 272 99 96 105 92 83 – 95 90 91 95 97 284 287 280 293 284 266 263 266 278 265 268 242 277 268 257 79 97 99 100 94 98 99 268 268 263 251 269 277 267 257 262 90 122 102 125 100 103 108 262 269 260 277 280 0.0 357 0.0 0.0 0.0 0.0 330 180 110 180 180 180 – 180 180 177 124 160 314 322 358 359 358 292 349 351 296 0.0 300 0.0 325 0.0 337 180 152 158 156 146 180 171 352 0.0 357 337 348 312 0.0 345 356 153 180 180 180 180 180 180 320 320 359 0.0 0.0 2.1 2.2 2.1 2.2 2.1 2.3 2.4 2.1 2.1 1.9 2.3 2.3 – 2.0 2.2 2.0 1.9 2.0 1.9 2.2 1-9 2.4 2.2 2.8 2.9 2.9 2.9 2.9 2.8 2.9 2.8 2.8? 2.8 2.5 2.8 2.8 2.6 2.7 2.5 2.5 2.5 2.6 2.4 2.6 2.7 3.2 2.6 2.8 2.9 2.4 2.5 2.8 2.8 2.8 3.0 3.0 2.8 3.2 3.0 3.0 3.0 29 30 24 32 28 14 24+ 31 14 23 25 21 – 29 28 28 27 28 25 25 22 26 25 19 29 31 22 31 18 29 28 30 24 35 29 23 23 23 22 23 23 20 24 32 28 19 25 22 22 17 17 28 21 17 16 19 18 18 29 18 19 4.45 3.23 5.17 2.81 3.79 6.86 2.37 2.61 1.0 3.26 3.52 4.62 – 2.93 3.21 3.07 1.07 2.25 1.20 1.40 3.55 2.54 2.96 1.37 2.97 2.74 0.82 3.06 1.78 4.07 1.71 3.07 3.33 2.89 1.90 2.57 2.43 2.26 3.73 3.13 3.65 4.60 3.92 2.69 2.82 1.84 3.72 4.0 4.27 3.71 3.41 2.79 2.62 4.71 4.81 3.79 3.22 2.83 3.41 4.61 4.21 220 190 222 200 178 215 ? 220 220 225 215 215 215 217 215 190/235 210 212 151 – 215 205? 225 215 212 210 225 215 211 205 240 235 220? – 210? 200/240 230 220 215 200/220 230 230 235 215? 205 215 – 215 215? 200 213 211 212 215 – 212 200 205? 240 240 – 250? 250 300 310 245 300 280 140 100 130 ? 110 – 125 100 110 100 103 275 – 300 320? 320 286 278 275 290 290 290 290 295 300 280? – 112 115 120 113 116 118 320 300 308 295 288 298 – 280 290 144 152 130 140 120 – 140 295 292 330 330 – 67/110 126 110 126 152 101 180 – – – – – – – – – – – 110 97 97 84 93 118 122 118 84 95 122 122 139 126 110 – – – – – – – 55 97 80 126 126 126 135 126 112 – – – – – – – 93 84 105 116 116 60 50 62 60 63 60 60 – – – – – – – – – – – 45 62 25 25 25 45 60 60 55 62 61 65 65 62 45 – – – – – – – 45 40 52 47 47 63 63 47 45 – – – – – – – 45 45 50 46 45 – – – – – – – 110 116 no 105 118 – 76 80 84 89 93 51 63 72 74 72 – – – – – – – – – – 126 110 105 67 67 89 93 84 63 59 – – – – – – 135 120 131 118 118 114 105 59 61 55 63 59 – – – – – – – 45 50 45 48 47 – 25 26 45 45 45 45 45 133 133 133 – – – – – – – – – – 46 47 45 47 47 47 47 135 133 132 – – – – – – 60 45 45 47 47 45 48 113 112 112 110 111 None “ “ “ “ “ “ “ “ “ “ DSR “ “ “ “ “ “ None “ “ “ “ “ “ “ “ None “ “ DSA “ “ None “ “ DSR “ “ “ None “ “ “ “ “ “ DSA “ None “ “ “ DSR “ “ None “ “ “ “ 0.72 0.65 0.68 0.90 0.80 0.39 0.62 0.69 0.52 0.63 0.67 0.38 – 0.70 0.60 0.76 0.91 0.77 0.85 0.88 0.68 0.87 0.78 0.82 0.99 0.84 0.93 0.80 0.78 0.58 0.93 0.81 0.59 0.70 0.84 0.75 0.86 0.88 0.59 0.69 0.81 0.72 0.83 0.90 0.81 0.88 0.69 0.70 0.63 0.81 0.63 0.91 0.83 0.49 0.49 0.71 0.83 0.84 0.85 0.67 0.59 0.43 0.09 0.27 0.08 0.08 0.21 0.17 0.22 0.32 0.41 0.48 0.39 – 0.53 0.55 0.51 0.06 0.57 0.30 0.15 0.19 0.15 0.21 0.35 0.64 0.82 0.63 0.72 0.38 0.65 0.82 0.86 0.82 0.60 0.59 0.38 0.41 0.45 0.41 0.42 0.53 0.53 0.62 0.69 0.70 0.30 0.19 0.44 0.42 0.13 0.41 0.78 0.43 0.40 0.42 0.49 0.33 0.33 0.68 0.42 0.24 1B 1B 1B 1B 1B 1B 1B 1S 1S 1S 1S 1S None 1S 1S 1S 1S 1S 1B, 1S 1B, 1S 1B, 1S 1B, 1S 1B, 1S 1B 1B 1B 1B 1B 1B 1B 1B 1B 1B 1S 1S 1S 1S 1S 1S 1S 1B, 1S 1B, 1S 1B, 1S 1B 1B 1B 1B 1B 1B 1S 1S 1S 1S 1S 1S 1S 1B, 1S 1B, 1S 1B, 1S 1B, 1S 1B, 1S L L L L L L L L L L L L L L L L L L L L L L L B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – Bow “ “ “ “ “ “ “ “ “ “ “ “ “ “ “ “ “ 0° 0° 35S 35P 35P 35S 35P 0° 0° 35 S 35P 0° 0° 0° 0° 35S 35P 35P 0° 0° 0° 35P 35S 0° 0° 0° 0° 35S 35P 35P 0° 0° 35S 0° 35S 35P 0° 35S 35S 35P 0° 35S 35P 0° 35S 35 P 35P 0° 35S 0° 0° 35S 35P 0° 35S 35P 0° 0° 0° 35S 35P Table 5 94 Disabled Tankers is subject to flow at a small drift angle. Therefore a rudder positioned at an unfavourable angle to the flow is possibly less prone to separation and locally increased drag than the unfavourably angled rudder when the ship is turned by bow. Towing forces at bow and stern. When towing forces were deployed at bow and stern and the model turned by the bow, to starboard, it is clear from Table 5 that a faster rate of turn was achieved with the rudder jammed in the favourable position of hard-a-starboard. 5.1.3 Use of limited VLCC power While at load draught, limited VLCC power was used in the model tests only when turning by the stern. The results are shown in runs 12 to 171 and in Fig. 8. It is clear that use of dead-slow-astern in this situation helped to reduce the maximum towing force required during the turn by some 20%. It is of interest to note that in run 13 no towing force was applied to the model which was allowed to drift in wind and waves with dead-slow-astern set on the motor. The purpose of this was to see if, with the rudder amidships, there was any tendency of the model to turn under such limited power until it was stern to weather. The run showed that while the model turned a few degrees into the weather, there was insufficient power available to turn it further and it drifted across the tank, ultimately reverting to a heading close to that at the start of the run. 5.1.4 Drift directions while turning A feature observed while turning the model by the bow or by the stern was that while it was comparatively easy to turn the model through the first 15° to 20° from its original heading, there then appeared to be a barrier to further turning. During this time the model drifted at a nominally steady track heading fH while at a fairly constant heading angle ?H. Values of fH and ?H estimated from the computer listings are given in Table 5 and shown diagram-matically in Fig. 7 and it is clear that for all runs the value of fH is approximately constant in the range 200° to 230°. In most cases there was a difference of about 90° to 100° between the direction of tow and fH (see run 8 for example). The reason for this apparent barrier to turning at ?H is probably associated with the turning moment maxima due to waves and wind which act on the vessel. For wave forces these maxima occur at wave incidence angles of 60° and 130° (in the ship body axes) as shown in Reference 1 while maxima due to wind alone occur at wind incidence angles of about 130° in full load draught. It is apparent from Table 5 that values of ?H are grouped around values of about 300° (when turning by the bow) and 110°–120° (when turning by the stern). Therefore the wave incidence angles when turning by the bow are about 360o–300° = 60° and those for turning by the stern about 110°–120° which approximately coincide with the measured wave- and wind-induced turning moment maxima mentioned above. These ‘turning barriers’ demanded particular towing techniques if excessive forces or excessive xmax values were to be avoided. It was found to be especially important to maintain the angular momentum generated during the early part of the turn. This was done by increasing the towing force as soon as the critical heading angle ?H was reached so that the model continued to turn. It was then of course necessary to reduce this towing force and adjust it to a more fore-and-aft direction in order to avoid the model swinging too rapidly once past the critical heading angle, and possibly turning through the weather. Indeed, once through the weather, the model would continue turning until a further turning moment maximum was reached. 5.1.5 Towing after turning Some experiments were continued after the model had been successfully turned either bow-to-weather or stem-to-weather with a view to assessing the ease with which the model could be towed into the prevailing wind and waves. Two general results arose from this study. A much reduced towing force was required to move the model directly into the wind and waves. This is not surprising because the model was presenting its minimum above- and below-water cross-sectional areas to wind and waves. When towing by the stern it proved very difficult to tow the model directly into the waves due to its tendency to sheer violently to port or to starboard. Generally such a sheer would end when the model reached a critical heading angle ?H as discussed above in which position any velocity into wind and waves would generally be lost and the model would begin to drift down-weather. Such a sheer is apparent in run 12 in which some astern power was used. In a real case such sheers could provide considerable difficulties for any towing vessel as additional towing force and a changed towing angle are required to recover the situation. This could provide the towing vessel with manoeuvring problems. 5.2 Ship Model at Ballast Draught A feature of the runs made at ballast draught was that the model exhibited a higher free-drift speed than when in load draught. This is apparent from the V0 results shown in Table 5 and is due to the fact that the wind forces on the vessel are greater because of increased windage while the resistance to motion is less due to the reduced underwater hull form. In general the free-drift velocity