RETROFIT OF TRANSOM FLAPS TO RN WARSHIPS

Transcription

1 393 RETROFIT OF TRANSOM FLAPS TO RN WARSHIPS BY MR MATTHEW WILLIAMSON UK MOD T45 DESTROYER IPT MRINA LIEUTENANTLt COMMANDERR PAUL CARROLL RN MCTA IMAREST MR BOB SCRACE QINETIQ PLC FRINA MISS CATHERINE INGRAM UK MOD GRADUATE ENGINEER ABSTRACT Following early experiments with Royal Navy ships and the experience gained by the US Navy and Coastguard, in 2004 a transom flap was fitted to the Type 23 Frigate, HMS ARGYLL. Given the 13% increase in fuel efficiency at speeds of over 16 knots reported for this ship, a similar flap was designed and installed in 2006 on the Type 42 Batch 3 Destroyer HMS MANCHESTER. This paper presents the development work done by QinetiQ Plc for these flaps including fullscale trials and the optimisation of controllable propeller pitch on the Type 42 Destroyer for the appended hull. Authors Biographies Mr Matthew Williamson is currently working as a Naval Architect for the UK MOD s MoD s T45 Destroyer Integrated Project Team (T45IPT). His previous post was as the Naval Architect in charge of stability and hydrodynamics for Major Warships IPT. Prior to this he worked for the Sea Technology Group developing and applying stability standards for the MoD. Lt Cdr Paul Carroll RN MTAU joined the Royal Navy as a University Cadet in After several seagoing appointments, he completed an MSc in Marine Engineering at UCL and subsequently worked on projects ranging from WR21 development at Rolls Royce to the design of the LSD(A). After serving as MEO in the destroyer HMS SOUTHAMPTON, he joined MCTA in 2005 where he was the lead marine engineering trials officer for Major Warships and the T45 Destroyer. He is now a student on the Advanced Command and Staff Course. Bob Scrace is a Principal Consultant in ship hydrodynamics at QinetiQ Haslar and has 30 years of experience in research and development focussed on current and future warships. He led much of the hydrodynamics research that culminated in trimaran warship demonstrator RV TRITON. He headed the team that developed and tested the outline design of the flaps that are the subject of this paper. He is a Fellow of the Royal Institution of Naval Architects.

2 394 Miss Catherine Ingram joined the UK MoD as a Graduate Trainee in Her training has included work periods with Major Warships IPT, Devonport Management Limited, QinetiQ Plc and the Royal National Lifeboat Institution. She is currently studying for an MSc in Naval Architecture at University College London. Introduction Although the Royal Navy had previously investigated the use of transom flaps (Type 21 Frigates) with a view to increasing top speed and reducing effects such as stern plumes, recent rises in costs have made them attractive for fuel economy. The first Royal Navy ship to be fitted with a fuel saving transom flap (HMS ARGYLL a Type 23 Frigate in 2004) showed a 13% increase in fuel efficiency as measured on full-scale trials. Following this successful implementation, a similar flap was designed and tested for a Type 42 Batch 3 Destroyer. Full-scale trials on HMS MANCHESTER showed a 10% increase in fuel efficiency. This paper presents the results from the model tests and trials that were conducted on both HMS ARGYLL and HMS MANCHESTER before and after their transom flap fits. TABLE 11 Ship Details HMS MANCHESTER HMS ARGYLL Commissioned 16-Dec May-91 Displacement (tonnes) Length (m) Beam (m) Propulsion COGOG. CODLAG Speed 2 Olympus gas turbines 2 Tyne gas turbines About 30 knots (56 km/h) 2 Spey Gas Turbines 4 x GEC-Alsthom Paxman Valenta About 30 knots (56 km/h) Transom Flap Description A transom flap is a small triangular appendage fitted to the lower edge of a vessel s transom. The following are the most common variations.

3 395 FIG.1 - TRANSFORM FLAP FIG.2 TRANSOM WEDGE FIG.3 TRANSOM "FLEDGE" Transom Flap Theory A transom flap affects a range of resistance and propulsion characteristics of both the vessel s hull and propeller. When correctly designed the beneficial changes to all these characteristics combine, resulting in a measurable improvement in performance. The following is a summary of some of the beneficial effects that transom flaps have. It should be noted that different factors dominate for different vessels and speeds. Wave making resistance is decreased both through an increase in water line length and by reducing wave effects at the stern of the vessel (behind the flap the local flow velocity is suddenly increased, resulting in a reduction of energy lost in the stern plume);

4 396 A more benign environment is created for the propellers. Flow around the propellers is altered reducing flow velocity which allied to a reduction in cavitation effects, increases the efficiency of the propellers (lower shaft RPM are achieved for the same forward speeds); The angled flap generates lift thus stern trim is marginally reduced at high speed; The pressure field acting over the aftermost 5% of the immersed hull is beneficially altered; The reduced speed of the flow over the hull extends forward to the propeller disk resulting in a small increase in wake fraction. Because each of the above effects is highly complicated in its own right, when their combination is considered, no theoretical models exist to adequately describe the collective phenomena. Hence there is a need for model tests to evaluate potential designs, and select the optimum solution. 4.Transom Flap Benefits and Design Philosophy In essence there are 3 primary benefits that can be derived through the fitting of a transom flap: 1. An increase in speed for a given fuel consumption; An increase in range for a given speed; A decrease in fuel consumption for a given speed. Other ancillary benefits can include reduced maintenance costs through reduced stresses on the power train and a possibility to further optimise the prime movers for a more efficient hull and propulsor. Furthermore, predicted fuel savings are in line with present governmental targets on reductions of carbon emissions. The design of the flap is closely related to which of the desired benefits is considered key. It should be noted that a flap optimised for increasing top speed will not necessarily provide a reduction in fuel consumed throughout the speed range. Indeed experience from the design of flaps for RN Ships has shown that it is possible to refine the design in such a way that a detrimental effect on performance can be observed at one end of the speed power curve for a clear improvement at another end. It should be noted that even though a small amount of time is generally spent at high speed by RN warships, due to the exponential nature of the speed power curve, a disproportional amount of fuel can be burnt at these speeds requiring careful consideration of the operating profile when anticipating fuel savings. For RN ships the design philosophy was to provide a reduction in fuel consumption throughout the speed range, with specific attention paid to those speeds at which the ship spent the majority of time. An extensive knowledge of

5 397 the operating profile of these ships enabled the savings to be confidently optimised. 5.Type 23 Frigate Transom Flap Development Work to develop a flap suitable for retrofitting to the Type 23 frigate class began in 2000 with a full review of all existing model transom flap experiment data. Detailed examination of the data showed that the likelihood of deriving an algorithm to predict the absolute effect on resistance of fitting a flap of given length and angle was remote but an algorithm to estimate the relative effect of changing flap angle was possible. Data from Type 21 model experiments was used to derive a 6th order polynomial in terms of Froude number and flap angle. This was applied to some existing Type 23 model data to test whether it would accurately predict the effect of increasing the flap angle by 5 degrees. Excellent agreement between model data and prediction was found. A second algorithm was derived to estimate the relative effect of changing flap length. The derived algorithms were used to estimate the optimum flap length and angle for the Type 23 frigate. In this context, optimum was taken to mean the flap giving the maximum reduction in predicted annual fuel consumption. The optimum length was found to be marginally over 1 per cent of the ship s waterline length with an angle of between 0 and 5 degrees trailing edge down. The analysis predicted that the ship would use 7 to 8 per cent less fuel, based on a typical operating profile. 5.1.Model tests During 2001 model experiments were conducted to further optimise the flap using an existing 5 metre long GRP model held by QinetiQ at Haslar. In order to expand the limited database and confirm the algorithms, three flap designs were developed and tested. All had a length, at ship scale, of 1.28 metres and were given angles of 0, 5 and 10 degrees. In all cases the angle was relative to the tangent of the buttock line right across the width of the flap. For these tests the width of each flap was set to the waterline beam of the transom with the model ballasted to the Class average displacement and trim. Photographs of the 5 degree flap are shown in (FIG.4 and 5).

6 398 FIG.4 PROFILE VIEW OF A 5 DEGREE FLAP FITTED TO THE T23 MODEL FIG.5 TRANSOM VIEW OF A 5 DEGREE FLAP FITTED TO THE T23 FRIGATE MODEL

7 399 Resistance experiments with and without the flaps were undertaken in the 270m long towing tank at QinetiQ Haslar. They covered the speed range equivalent to 10 to 31 knots full scale and the resulting data was extrapolated to ship scale and faired in accordance with QinetiQ s standard methods. The percentage reduction in effective power due to the flaps is plotted in (FIG.6). 4 Percentage effective power reduction 0 degs 5 degs 10 degs 2 % reduction in Pe Speed (knots) FIG.6 PERCENTAGE REDUCTION IN EFFECTIVE POWER DUE TO THE FLAPS The results are in broad agreement with previous experiments. The 10 degree flap gives a significant penalty at low speed but is the best at top speed. However, both the 0 and the 5 degree flaps show benefits over the entire speed range, albeit that the gains are not quite as high as previous T23 experiments where a 4 per cent improvement was measured. The model was then re-fitted with all appendages and propellers and propulsion experiments were conducted at speeds equivalent to 12, 16 and 28 knots. These speeds were selected as representative of low, cruise and top speeds for the Type 23 class. The results of the propulsion experiments are summarised for convenience in (FIG.7).

8 400 Percentage propulsive coefficient increase 0 degs 5 degs 10 degs 6 4 % increase in PC Speed (knots) FIG.7 PERCENTAGE CHANGES IN PROPULSIVE COEFFICIENT DUE TO THE FLAPS Experimental results The results at 12 knots are somewhat questionable since it might be expected that the 0 degree flap would have little effect on increasing wake, as measured at 16 and 28 knots. However, it is clear that the 5 degree flap gives the greatest improvement in propulsive efficiency at all speeds. (FIG.8) shows a photograph of the flow around the 5 degree T23 flap at 28 knots.

9 401 FIG.8 FLOW BEHIND THE TYPE 23 FRIGATE MODEL WITH THE 5 DEGREE FLAP AT 28 KNOTS The 5 degree flap was thus found to give the best results in terms of effective power reduction and propulsive efficiency increase. These effects were combined to give a prediction of shaft power reduction that is shown in (FIG.9). 7 Percentage shaft power reduction due to the 5 degree flap 6 % reduction in Ps Speed (knots) FIG.9 PERCENTAGE REDUCTION IN SHAFT POWER DUE TO THE 5 DEGREE FLAP

10 402 Cost benefit analysis conducted by Frigates IPT supported the case for retrofitting the flap to the Type 23 class. QinetiQ worked with BMT Defence Services Ltd to develop the detailed design drawings and obtain design approval by Lloyd s Register. For structural reasons the flap was extended further around the section of the transom but the key hydrodynamic features were unaltered. 6.Full Scale Trials Pre Fit The first opportunity to retrofit the flap arose when HMS ARGYLL was programmed for dry docking and routine maintenance at Devonport in summer In order to measure changes in ship performance due to the flap it was necessary to plan and conduct a pre-fit speed and powering trial ahead of the docking. This was conducted by QinetiQ in May 2003 while the ship was returning to Devonport from a visit to Copenhagen. The underwater part of the hull and the propellers were cleaned by Royal Navy divers approximately 20 days before the trial. QinetiQ fitted instrumentation and a data acquisition system to measure and record speed, shaft torques and revolutions. 6.1.Trial details Trials began early on 7 th May in an area between the northern tip of Denmark and the southern tip of Norway where the water depth exceeded 150 metres. The visual sea state was 3 with a slight swell while the wind speed was 10 knots, both from the SW. The ship s stabilisers were shut down and locked at nominal zero incidence. The trial aimed to measure the relationship between speed though the water and total shaft power at 11 different fixed shaft rates of revolution nominally covering the speed range 5 to 28 knots. Each of these 11 runs comprised of 4 legs; 2 into the tide and 2 with the tide. At the start of each run the shaft revolutions were set to that required and then not altered until all 4 legs had completed. Once the revolutions were set the speed over the ground, as measured by GPS, was monitored until it appeared steady. Speed, shaft torsion and other data were then acquired for 5 minutes during which the Quartermaster was asked to maintain heading with minimal use of the rudders. A limit of 3 degrees rudder angle was set for all but the 5 knot runs, where up to 10 degrees was required to counteract wind yawing effects. The 5 minute data acquisition period was increased if the speed or torsion data appeared unsteady. At the end of the data acquisition period the ship continued on the same heading for another minute before a Williamson turn was conducted in order to bring the ship around onto the reciprocal heading and track. The rudder angle used for the turn varied from 20 degrees for the low speeds runs to 5 degrees for the higher speed runs. Once steady on the reciprocal heading the GPS speed was again monitored until steady, from when the next 5 minute data acquisition period began. The trial continued throughout the 7 th May. The wind and sea conditions remained reasonably steady throughout the trial with the absolute wind speed close to 14 knots and the visual sea state 2-3. The underlying swell appeared to reduce. The acquired data were analysed to yield mean speed over the ground, mean shaft torsion, mean running trim angle, mean relative wind speed and direction and rms

11 403 pitch and roll motion for each leg. Mean shaft revolutions were obtained by Fourier analysis of the torsion meter data and confirmed by comparison with values recorded manually from the ship s instrumentation during the trial. 7.Flap Fitting and Full Scale Trials The flap was constructed and fitted to HMS ARGYLL by Devonport Management Ltd in July and August Fitting was overseen by QinetiQ in order to ensure that the key hydrodynamic aspects of the design were maintained. The angle between the ship s after buttock lines and the lower surface of the flap was measured at seven positions and found to average 4.9 degrees. (FIG.10) gives a view of the flap during fitting to the ship and shows clearly the 5 degree angle. (FIG.11) shows the completed flap before the ship was re-floated while (FIG.12) shows how it sits relative to the ship s waterline. FIG.10 PROFILE VIEW OF THE HULL AND FLAP CENTRELINE SHOWING THE 5 DEGREE FLAP ANGLE

12 404 FIG.11 THE COMPLETED FLAP AS FITTED TO HMS ARGYLL IN DRY-DOCK FIG 12 THE COMPLETED FLAP AS FITTED TO HMS ARGYLL Post Fit Trial The post-retrofit trial was conducted in Falmouth Bay on 17 th June 2004 in water depths of approximately 90 metres. The hull and propellers were cleaned by divers 6 days beforehand. The trial was conducted by the same team from QinetiQ

13 405 using the same instrumentation and following exactly the same trials procedure as the pre-fit trial. The ship s MEO made every attempt to attain the same draughts as for the pre-fit trial but was unable to match them due to fuel requirements for follow-on deployment. As such, the ship s displacement was 118 tonnes (2.7%) greater than during the pre-fit trial. The prevailing wind and sea conditions were very similar to those during the prefit trial and remained steady throughout the trials period. The mean absolute wind speed was only 3 knots higher, at 17 knots, while the visual Sea State was 3 but without any underlying swell. (FIGS.13-15) compare the measured performance of HMS ARGYLL during the pre- and post-retrofit trials. T23 Flap Full Scale trial results Total shaft power No flap Flap 13% reduction of No flap line Total shaft power (kw) Speed (knots) FIG.13 COMPARISON OF PRE- AND POST-FIT TRIAL SHAFT POWERS

14 406 Percentage reduction in shaft power % reduction in Ps Speed (knots) FIG.14 PERCENTAGE REDUCTION IN SHAFT POWER DUE TO THE FLAP Running trim angle No flap Flap Trim angle (degs - positive bow up) Speed (knots)

15 407 FIG.15 COMPARISON OF PRE- AND POST-FIT RUNNING TRIM ANGLE Without correcting for any differences between the pre- and post-trial ship displacements, (FIG.13) shows an approximate 13% reduction in shaft power to attain any speed above 10 knots due to the flap. A further 2-3% reduction could reasonably be added to this to account for the displacement difference. It is estimated that the saving is equivalent to about 70,000 per annum, based on typical annual Type 23 fuel usage. It is evident that the shaft power reductions measured on HMS ARGYLL are far larger than those deduced from the model experiments. This is in line with US Navy experience with flap retrofits to the Spruance Class where an averaged shaft power reduction of nearly 12% was measured on ship trials; some 4-7% more than predicted by model experiments. The reasons for this apparent scale effect are unclear. To date, the flap has subsequently been retrofitted to Type 23 frigates HMS KENT and HMS NORTHUMBERLAND. T42 Batch 3 Destroyer Transom Flap Development The success of the Type 23 flap prompted Major Warships Integrated Project Team (MWIPT) to task QinetiQ to develop a flap for retrofitting to the Batch 3 Type 42 destroyer class. The transom of the Type 42 is flatter in section than that of the Type 23 and has a much tighter bilge radius. The Type 42 also has significantly greater transom immersion at the normal operating displacement. While this required development of a different geometry, it also gave QinetiQ an opportunity to address the only reported negative aspect of the Type 23 flap. The Commanding Officer of HMS ARGYLL reported informally that the flap had affected control of the ship when going astern. Intuitively, the flap would be expected to affect the quality of flow over the upper portion of the rudders and it is also possible that its upper, scooped shape introduces turbulence that affects control. The C.O. did not believe that this change was significant operationally but it did result in higher angles being required to achieve astern steering control. Model Tests The agreed experiment programme allowed for the testing of four flap designs. Based on the Type 23 flap, the first of these was given a chord of 1.4m and an angle of 5 degrees. Its width was set to 5.74m in order to land its edges on one of the ship s vertical stiffeners and maximise its lower surface area. The other three flaps each featured an alteration to one parameter of this design in order to explore the influence of length, width and angle. The depth of each flap at its attachment point to the transom was set to half the chord, based on outline structural evaluation of previous flap designs for other vessels. Thus, each flap was triangular in profile. The ends were tapered in order to encourage the water to flow reasonably cleanly off the flap and around the transom when going astern. The four model flaps and their attachment arrangement to the model are shown in (FIGS.16 & 17).

16 408 FIG.16 THE FOUR T42 BATCH 3 FLAPS FIG.17 THE CHOSEN T42 BATCH 3 FLAP ATTACHED TO MODEL TRANSOM Resistance and propulsion experiments with and without the flaps fitted were conducted in the towing tank at QinetiQ Haslar during June The effects of the various flaps on model resistance are shown in (FIG.18).

17 409 Experimental Results 8 Percentage effective power reduction 5 deg 0 deg Long Narrow 6 % reduction in Pe Speed (knots) FIG.18 PERCENTAGE REDUCTION IN RESISTANCE DUE TO EACH FLAP In terms of resistance reduction it is clear that the 0 degree flap (i.e. a 1.4m long hull extension) produces the most consistent reductions across the speed range. The longer, 5 degree flap is marginally better at high speeds but much worse at lower speeds. When combined with the propulsion experiment and propeller performance data, the estimated reductions in shaft power due to each flap are as shown in (FIG.19).

18 410 Percentage reduction in shaft power 5deg 0deg Narrow Long % reduction in Ps Speed (kts) FIG.19 PREDICTED PERCENTAGE REDUCTIONS IN SHAFT POWER DUE TO EACH FLAP It is evident that the 1.4m long, 5 degree flap is the best over most of the speed range, thus QinetiQ recommended this design for retrofitting to the Batch 3 Type 42 class. MWIPT tasked Fleet Support Ltd at Portsmouth to develop the detailed structural design. Full Scale Trials pre Fit At opportunity to retrofit the flap soon arose as HMS MANCHESTER was due to be dry-docked at Devonport for routine maintenance during A limited prefit trial took place in Bute Sound off the NE coast of Arran on 22 nd and 23 rd February 2006 in water depths in excess of 100m. Approximately two weeks beforehand the underwater hull, appendages and propellers had been cleaned by divers to remove all fouling. In order to minimise costs, no specific trials instrumentation was fitted. Instead, data from the ship s display panels were recorded manually at a rate of one reading every 30 seconds. Speed trials at 13 different power lever setting runs were completed in very good weather conditions following the same manoeuvring and general trials procedure as for the HMS ARGYLL. The sea state was estimated as 1-2 and the absolute wind speed varied between 7 and 23 knots with a mean of 15 knots.

19 411 Flap Fitting and Full Scale Trials The flap was fitted during April/May 2006 while the ship was in dry dock at Devonport. MWIPT and QinetiQ Sea representatives oversaw an accurate survey of the flap and requested minor adjustments. The fitted flap was thus set accurately at an angle of 5 degrees relative to the local hull buttock line angles across its width. A picture of the flap taken just before the ship was re-floated is shown as (FIG.22). FIG.20 PRODUCTION DRAWINGS FIG.21 HMS MANCHESTER TRANSOM FLAP AT BUILD

20 412 FIG.22 HMS MANCHESTER TRANSOM FLAP AS FITTED The ship was re-floated in mid June and sat alongside until late September. Due to programme and budgetary constraints and the fact that a new paint scheme had just been applied, the hull was not specifically cleaned prior to the post-fit trial, although the ship did sit alongside for over 3 months in summer. Instead of a hull and propeller clean, an underwater survey using an ROV with a fitted camera was completed by the MoD s Salvage and Marine Operations Integrated Project Team (S&MO IPT). The survey took place ten days before the post-fit trial in relatively poor visibility. Nonetheless, the camera captured enough video of the forward, port side of the hull between the gun and the bridge with which to make a relative judgement about the fouling state. (FIG.23) shows a portion of the hull at approximately half draught adjacent to the gun. It is evident that the hull was not clean, but was largely covered in a layer of green slime. (FIG.24) shows the hull adjacent to the bridge where it appeared cleaner although still dotted with growth.

21 413 FIG.23 HMS MANCHESTER HULL FOULING PRIOR TO THE POST-FIT TRIAL (HALF DRAUGHT ADJACENT TO THE 4.5 INCH GUN) FIG.24 HMS MANCHESTER HULL FOULING PRIOR TO THE POST-FIT TRIAL ADJACENT TO THE BRIDGE Post Fit Trial The post-fit trial took place south of Start Point to the east of Plymouth on 24 th and 25 th October 2006 in water depth of 70m. The same procedures for running the trial and manually recording data as those used during the pre-fit trial were adopted. Care was taken to ensure that the controllable pitch propellers were initially set to the same pitch angle as during the pre-fit trial. The weather conditions were far from ideal and very different from those prevailing during the pre-fit trial. At the start of the first trial run the wind was blowing at 30 knots and the combined swell and wind-sea conditions were

22 414 estimated by the officer of the watch at state 4-5. Fortunately, the wind moderated during the first day so the first 2 trials runs, made in the worst conditions, were repeated. As such, the wind speed varied between 10 and 24 knots with a mean of 15 knots (i.e. nearly identical to that during the pre-fit trial) although the swell remained at an estimated 2-2.5m (i.e. sea state 4-5). Subsequent to the trial QinetiQ obtained hindcast sea state data from the Met. Office which showed the significant wave height reducing from 2.6m to 1.6m (sea state low 5 to mid 4) during the period of the trial. The pre- and post-fit shaft power curves are shown in (FIG.25). It should be noted that for this figure, the post-fit trial data has not been corrected for the effects of sea state 4-5, nor for the effects of 4 months of fouling growth. Type 42 Batch 3 Destroyer Full Scale Trial Results Total shaft power - uncorrected Pre-fit Post-fit Ps (KW) Speed (kts) FIG.25 - COMPARISON OF PRE-FIT AND UNCORRECTED FOR FOULING OR SEA STATE POST-FIT TRIAL POWER CURVES Despite the influence of the Sea State and fouling, the flap has still reduced the shaft power required to attain speeds above 25 knots. The differences between the curves are small around the maximum cruise speed of 19 knots. The post-fit data was corrected to remove the influence of sea state 4-5 and 4 months of fouling, using empirical methods derived from QinetiQ s extensive full scale trials databases. The sea state correction was speed and relative wave

23 415 heading dependent, ranging, in head seas, from 6% at 29 knots to 16% at 10 knots. The ship was approximately 125 (summer) days out of dock when the post-fit trial commenced, thus an increase in frictional resistance of at least 13.7% was deemed appropriate. The resulting corrected power curve is plotted in (FIG.26), where it is again compared to the pre-fit curve. It is evident that the post-fit is now below the pre-fit curve at all speeds. Total shaft power - corrected Pre-fit Post-fit Ps (KW) Speed (kts) FIG.26 COMPARISON OF PRE-FIT AND WAVE PLUS FOULING CORRECTED POST-FIT TRIAL POWER CURVES The data in (FIG.26) has been used to deduce percentage power savings due to the flap over the speed range 10 to 29 knots and this is shown in (FIG.27).

24 416 Percentage reduction in shaft power % reduction in Ps Speed (kts) FIG.27 ESTIMATED SHAFT POWER REDUCTIONS DUE TO THE FLAP It was reported separately by the Commanding Officer and Navigating Officer of HMS MANCHESTER that maximum speed when powered by the Tyne cruise gas turbines had increased by 1-2 knots but the maximum speed when powered by the Olympus sprint gas turbines had not increased noticeably. It was further reported that the height of the stern plume had reduced noticeably, indicating that less power was being lost in the wake. The data plotted in (FIG.27) indicates that the speed increase at Tyne max. would be closer to 0.7 knot and at Olympus max. Finally, it was reported that at 12 knots astern the upper surface of the flap simply guided water onto the open quarterdeck rather than encourage it to flow around the sides of the transom. Implications for Propulsion Dynamics The fitting of the transom flap to a COGOG ship with a controllable pitch propeller necessitated a check of both steady state and acceleration propulsion dynamics. This check was essential as it was anticipated that propeller loading and advance conditions would almost certainly change post flap fitting. The cumulative effect of the changes (outlined in section 3) meant that propeller thrust and torque distribution were also likely to alter. This, in turn, would change propeller operating conditions such that the ideal full ahead pitch, pitch schedule and shaft speed schedule settings would require adjustment to deliver best fuel economy against the operating profile. However, in making such changes it was

25 417 important to ensure that material limits on loadings, such as the spindle torque (FIG.28), were not exceeded. In other words because the ship had Controllable Pitch Propellers fitted, there was an opportunity to further enhance economy so long as the changes did not impact on availability or reliability. FIG.28 ILLUSTRATION OF FORCES ACTING ON CPP BLADE Rationale for Adjustment of Propulsion Control Variables The Type 42 Destroyer propulsion control is based around a closed loop system built upon pitch, shaft speed and associated engine fuel schedules. This system aims to deliver a set shaft speed and optimal pitch for a given power control lever demand setting. Variations for hull fouling and displacement are overcome through the use of a low gain, limited output integrator which adjusts the engine fuel schedule to compensate for errors between desired and achieved shaft rpm. Studies undertaken by Hawdon et al [1] identified that controllable pitch propeller thrust, torque and spindle torque are highly dependent upon variations in the wake field. Given that the transom flap would influence the wake field, especially when operating at high powers, it was considered essential to quantify these effects and assess how full ahead pitch, pitch schedule and related shaft speed schedule could be best matched to the new hydrodynamic performance.

26 418 FIG.29 CURRENT TYPE 42 DESTROYER PITCH DEMAND SCHEDULE CFD Model Initial estimates as to the best fit propeller pitch post transom flap fitting were undertaken using by using the Paramarine software package to develop distinct powering data for comparison with the pre-fitting trial measured results. FIG.30 T42 PARAMARINE MODEL SCHEMATIC (WITHOUT FLAP) To deliver best economy this data was matched against the operating profile for the ship, focusing upon the knot transition zone between the Tyne cruise and Olympus boost engines. Calculations indicated that best economy could be delivered by increasing the speed range delivered by the Tyne engines with the consequent considerable cost savings when compared with use of the Olympus in

27 419 this range. Software results indicated that an increase of full ahead pitch from 29 to 33, would deliver power efficiency savings of up to 5% at full power. However, further refining to a specific full ahead pitch setting would require test result validation. The relationship between T42 CPP full ahead pitch angles and non-dimensional pitch ratio is summarised at (TABLE 2) below: TABLE 2 Full Ahead Pitch Angles and Non-Dimensional Pitch Ratio for a T42 CPP Blade Angle Pitch Diameter Ratio Sea Trial Results - Pitch Variation 29 o o o o o o o 1.30 During the post fit full-scale trial, runs at Tyne max and Olympus max were conducted at different propeller pitch angle settings. The maximum pitch that the Class is currently authorised to use is 30.5 (although this limit was lifted for the conduct of the trial). The comparison of propulsive power to pitch ratio is independent of engine performance being an assessment only of hull and propeller characteristics. Work by Rubis [2] identified that fuel usage rates are typically insensitive to pitch ratio changes above 1.06 at low speeds. Consequently, the examinations of performance changes due to pitch ratio variation focused upon Tyne and Olympus max speeds.

28 420 Effect of pitch variation on power Power (KW) Without Flap With Flap, 31deg Pitch With Flap, Original Pitch With Flap, 30.5deg Pitch Speed (kts) FIG.31 HMS MANCHESTER UNADJUSTED POWER VS SPEED DATA FOR COMPARISON OF FULL AHEAD PITCH SETTINGS Effect of pitch variation on RPM Pre Flap Fit results With Flap, 31deg Pitch With Flap, Original Pitch With Flap, 30.5deg Pitch RPM Speed (kts) FIG.32 HMS MANCHESTER UNADJUSTED SRPM VS SPEED DATA FOR COMPARISON OF FULL AHEAD PITCH SETTINGS

29 421 FULL POWER APPROACH CORRIDOR LIGHT CLEAN 250 Comparison of stbd shaft full power approach corridor pre and post transom flap fit. Green is pre-fit, blue is post fit. DEEP DIRTY SHAFT RPM 240 PITCH TOO FINE 230 SATS - CLEAN 50% FUEL 220 PITCH TOO COARSE TORQUE FIG.33 FULL POWER APPROACH CORRIDOR (STBD SHAFT) PRE-FLAP (GREEN) 29.5 PITCH SETTING AND POST FLAP FIT (BLUE) WITH FINAL 30.5 PITCH SETTING Summary of Analysis of Pitch Change Results QinetiQ estimated that maximum shaft power reductions would be realised with pitch set at 31.2 degrees at full power, although the effects of changing the pitch were expected to be small. Analysis of runs at different pitch angles proved inconclusive and the changes in performance were less than 2%. Nonetheless, the full power approach corridor results (FIG.33) did identify that the combination of flap fitting and propeller pitch increase had improved CP propeller performance based upon the experimental cavitation tunnel results which originally defined the corridor. Spindle torque settings, assessed by monitoring CPP ahead and astern line hydraulic pressure fluctuations) also remained well within tolerance throughout the trials. Consequently, given that the best fuel savings within the operational profile were delivered by increasing Tyne max to reduce the time spent using Olympus engines, the 30.5 degree pitch setting was maintained on completion of the trial. Given the limited variation measured in the post fitting trial, no adjustment to the pitch or shaft speed schedules was undertaken at this stage although a further review will be made following trials in HMS GLOUCESTER scheduled for Autumn 2007.

30 422 Transient Shaft Torque Analysis of the mutual influence of the flap and the propulsion system during fast accelerations and crash stops was not fully examined during the post fitting trial due to the poor Sea State. However, a comparison of transient torque on maximum acceleration identified a notable increase post flap fitting (TABLE 3). TABLE 3 - Transient Torque on Maximum Acceleration Post Flap Fitting Flap On acceleration from rest to max speed Peak transient torque Average transient torque (KNm) (KNm) Fitted Not fitted Although this raised concerns for the spindle material design, given that spindle torque typically reaches its maximum at close to minimum pitch at high SRPM [3], measurements indicated that this remained in tolerance throughout the acceleration. Consequently, attention focussed on possible gearbox implications although transient torque limits were not exceeded by a considerable margin. It is intended to further examine these results during future trials in HMS GLOUCESTER. Business Case With modern warships frequently having an operational life measured in decades, a transom flap can provide savings over many years, and even comparatively late in a vessels life it can still prove financial beneficial to install a transom flap. HMS MANCHESTER Specifically considering HMS MANCHESTER, in justifying the development work, the following fuel savings were initially used. These enabled the flap to pay for itself in 12 months: 2.5% from knots; 6% from knots; 8% from knots. Other assumptions included non-competed labour, fuel at 250/tonne, an in service operating profile and a single ship fit to re-coup investment costs. Clearly if the cost of fuel rises, or the ship operates predominantly in the speed ranges where the biggest fuel savings are seen, the time to payback can be further reduced.

31 423 The above figures were used to justify the HMS MANCHESTER fit prior to fullscale trials. As can be seen from the results presented in this paper 12 months is a pessimistic period for the return on the investment to be realised. It should also be noted that over and above the clearly defined fuel savings, additional economies can be gained from reduced wear and tear on the whole propulsion train. Also the change over to the boost turbines can be delayed, putting back a step change in fuel consumption, and further reducing wear on the prime movers. Conclusions Transom flaps are a small and low cost addition to a ship s hull, which improve the fuel efficiency of a vessel and rapidly pay for themselves, also contributing to a reduction in greenhouse gas emissions. The following conclusions have been drawn from the work to develop transom flaps for Royal Navy frigates and destroyers: Optimum transom flap length is marginally over 1 per cent of the ship s waterline length; The optimum angle is between 0 and 5 degrees trailing edge down; Full scale results show a 10% fuel saving for T42 Batch 3 Destroyers with 13% for Type 23 Frigates; Model tests tend to under predict power reductions; Propeller pitch increase improves propeller performance based upon the experimental cavitation tunnel results which originally defined the full power approach corridor; Transient torques on the propeller system on maximum acceleration increase post fit; Depending on cost assumptions payback can be achieved in around 12 months. The results presented here confirm the validity of retrofitting transom flaps to Royal Navy Ships. Indeed such is their perceived benefit that tank tests have been undertaken to investigate their retrofitting to the INVINCIBLE class of aircraft carriers. They are also being considered for the Royal Navy s future aircraft carriers (CVF), whilst the new T45 Destroyers have a fledge fitted. Disclaimer The views expressed here are those of the authors, and do not necessarily reflect those of the UK MoD or QinetiQ Plc.

32 424 Acknowledgements HMS MANCHESTER Commanding Officer and ship s company HMS ARGYLL Commanding Officer and ship s company Frigates Integrated Project Team (FIPT) Major Warships Integrated Project Team (MWIPT) The Graphics Research Corporation Mr Doug Ira NAVSEA 05D US Navy Technical Exchange Officer to the UK MoD The Salvage and Marine Operations Integrated Project Team (S&MO IPT) References 1. Hawdon et al, The analysis of controllable pitch propeller characteristics at off design conditions, Trans IMarE Vol C J Rubis, Acceleration and steady state propulsion dynamics of a gas turbine ship with controllable pitch propeller Trans SNAME Vol 80 p K Brownlie, Controllable pitch propellers for future naval ships, p232 Paper 20 INEC 94 British Crown Copyright 2007/MoD. Published with the permission of the Controller of Her Britannic Majesty s Stationery Office. 20.