The Stepped Hull Hybrid Hydrofoil

Christopher D. Barry, Bryan Duffty

oil

ff co

e ap

earan

ilityepingrpose

ferri

el ex

nd

ation

solu

ncept

ry i

Planing @brid hydrofoils or partially hydrof

intentionally operate in what would be the takeo

compromise ofperformance and cost that might b

The stepped hybrid configuration has made app

1938. It is a solution to the problems of instabhybrids. It can be configured to have good seake

widely as would be justified by its merits. The pu

the marine community, particularly for small, fast

We have performed analytic studies, simple mod

them have determined some specljic problems a

concept. This paper presents background inform

stability, seakeeping, and propulsion and suggests

have limited the widespread acceptance of this co

Finally we propose a “strawman” design for a fer

BACKGROUND

A hybrid hydrofoil is a vehicle combining thedynamic lift of hydrofoils with a significant amount oflit? tiom some other source, generally either buoyancy

or planing lift. There are also concepts that useaerodynamic lift, such asvarious types of hydrofoilwindsurfers. There may even be concepts that use aircushions. The attraction of hybrid hydrofoils is thedesire to meld the advantages of two technologies in anattempt to gain a synthesis that is better than either onealone, at least for a specific mission.

Buoyant hybrid hydrofoils generally have one ormore torpedo-like submerged hulls or narrowcatamaran hulls and derive reduced resistance throughreduction of wavemaking drag and skin fkiction.

Meyer (1992) presents a number of concepts mergingbuoyancy and hydrofoil lift, and an experimentalHydrofoil Small Waterplane Area Ship “HYSWAS”implementing this proposal was in operation on theChesapeake in 1996. Smith (1963) described a highspeed sailing craft that combines buoyancy and foillift.

299

supported planing boat are hydrofoils that

ndition for a norma[ hydrofoil. They ofler a

propriate for ferq missions.

ces in the high speed boat scene as early as

and inefficiency that has limited other type ofas well, but the concept has not been used asof this paper is to reintroduce this concept to

es.

periments and manned experiments, andfiom

issues for the practical implementation of this

, discusses key concepts including resistance,

tions to what we believe are the problems that

.

n a particular service using this technology.

Partially hydrofoil supported planing hulls mixhydrofoil support and planing lift. The most obviousversion of this concept is a planing hull with ahydrofoil more or less under the center of gravity.Karafiath (1974) studied this concept and ran model

tests with a conventional patrol boat model and ahydrofoil, both of which were literally “off-the-shelf’.His experiments showed drag reductions of up to 50%.His studies also revealed one of the most importantproblems of hybrid hydrofoils: many of hisconfigurations were unstable in pitch. The subject ofthis paper is a particular configuration of partiallyhydrofoil supported planing hull that addresses thepitch instability issue.

The attraction of a partially supported planinghull is obvious: Hydrofoil lift is at least twice asetllcient in terms of lift to drag as planing lift, but ahydrofoil needs a surface reference to maintain acontrolled depth below water. By combining the hvo,the vehicle is much more efficient than an unsupportedplaning hull.

For example, arrange the vehicle so that abouthalf the weight is supported by the foils, the other half

— .—

by the hull. The foils carriy half the load but produce aquarter of the initial (planing) value. The reducedweight load on the remainder of the planing surface

produces lower drag per unit lift because drag due toplaning lift is proportional to load squared. Thus, thehalf load supported by planing is also carried moreefficiently. The total comes to less than three quartersof the original value.

Hybrid hydrofoils have been a surprisingly fertilefield of invention (though with perhaps an equallysurprising lack of practical implementation). There isa specific subclass of patents just for hybrid hydrofoilswith patents dating back to the early part of the century(Hayward, 1965). To our surprise (and disappointmentwhen we got back a rejected patent application),Supermarin obtained a patent specifically on steppedhull hybrid hydrofoils in Sweden in 1951 following a1943 application. The authors have also found hints inthe literature that stepped hybrid hydrofoils actuallysaw service during World War II, though all of thecraft we have been able to positively identi~ wereprobably full hydrofoils or hybrids with surfacepiercing foils forward and planing surfaces aft.Despite this heritage, hybrid hydrofoils haven’t had asignificant presence in the field of high speed boats.Interestingly enough, even the pure hydrofoil itselfseems to be disappearing atler a very promising start.Is this due to the traditional reluctance of the marineindustry to embrace new concepts, some technicalproblem, or is the hybrid hydrofoil a solution lookingfor a problem?

The authors initially became involved with thehybrid concept when working on FMC’SHighWaterspeed Test Bed (HWSTB). The Marine Corps

had determined that future amphibious assaults wouldhave to be launched from over the horizon. Thisrequired armored amphibian vehicles capable of atleast twenty five knots. It is the nature of armoredvehicles to be very heavy for their size, so heavy that aplaning hull is massively overloaded for the availablesurface of the bottom plate. This requires eitherextending the planing surfaces or somehow reducingthe load. DufTtysaw Karat5ath’s paper and convincedFMC that this was a solution. A hydrofoil provided anobvious means to reduce the planing load on thesystem so that the bottom plate could lift theremainder.

The progress of the High Waterspeed Test Bedproject is beyond the scope of this paper, but suffice itto say that the concept worked and a half scaledemonstrator representing a 66,000 pound armoredvehicle made 35 knots true (not scale) speed. The

300

authors decided to investigate application of some oftheir concepts to purely maritime boats, and FMC hasallowed them to use some of their work and conceptsdeveloped on HWSTB on their own.

This project has been limited by the time andfimding constraints familiar to any garage inventor,(especially with one of us on each coast) but someprogress has been made. This paper thereforerepresents a work in progress and is intended toreintroduce the stepped hybrid hydrofoil concept to themarine industry and to provide some inspiration toothers.

We feel it also shows that some usetid researchcan be done with finding limited to pocket change andscrounged equipment. Though we don’t have a lot ofgood quantitative data, we have learned a lot.

PITCH INSTABILITY

Pitch instability is the chief issue in a hybridhydrofoil. Planing hybrid hydrofoils can exhibit adynamic pitch instability similar to porpoising, thoughthe term describes this phenomenon even better than itdoes in a conventional planing hull.

This phenomenon can be best understood for anominal configuration with a single hydrofoil beneaththe center of gravity of a planing hull. If such aconfiguration is slightly disturbed bow up ffom anequilibrium position, the lift on both the foil and thehull will increase. The hull will begin to accelerateupwards and the intersection of the water surface andthe keel will move aft. This develops a bow downmoment, but at a relatively slow rate. By the time thebow drops enough to reduce eliminate the excess lift,

the vessel is well above the equilibrium position, andthe keel/waterline intersection is well aft. It falls backdown toward the equilibrium position bow down, as ifit had tripped on its stem, As a result of kinetic

energy, it carries through the equilibrium position,takes a deep dive and springs up again. This cyclerepeats, each time growing more severe. The motionresembles a porpoise even more than the similarmotion in a planing boat, because the vehicle seems tojump through a series of invisible hoops, diving intothe water after each one. The only way that this

motion can be damped is if the hull provides enoughdarnpingto prevent the increasing overshoot. Notethat this is a smooth water instability and occurs withonly a nominal initial disturbance.

Even if the parameters of a vehicle are such as toprovide sufficient damping to limit the growth ofporpoising, the pitch’heave mode is still very weakly

——

damped and therefore motions in head seas will beamplified in waves of the appropriate fi-equency.

Karafiath found that this instability in hisexperiments was correlated to the ratio of foil lift tototal displacement and the ratio of foil lift moment toweight moment (referred to the transom) and was afimction of speed. Values of foil liftlhull lifi exceeding40% and foil moment to hull to hull moment of 50%was an approximate limit for stability. Thiscon-esponds to a foil under the center of gravitycarrying half the vessel weight.

This limit is most unfortunate. It limits theeffectiveness of the concept since the more weightcarried on the foils, the more the drag reduction.

Other solutions include making the foil sense thesurface and lose lift at a relatively low draft excursionfrom equilibrium. Some Soviet river hydrofoils wereactually hybrids: They were carried by forward foilsrunning close to the surface so that their lift wasreduced by “biplane” effect. The stem was held up byplaning so they were “tail draggers”. A Japanesesystem (Kunitake, 1991) uses a forward surfacepiercing foil, and Rodriguez has run a foil assistedpassenger ferry with an aft surface piercing foil.

A pair of foils forward and afl of the center ofgravity can also be used. If the forward foil has ashallower rate of increase of lift with angle of attackthan the atl one, the total foil center of gravity willmove aft with pitch up and the distance of the foilsfrom the CG will produce damping. This differencecan be produced by reduced aspect ratio, so that theforward foil has a relatively small span and is longalong the length of the boat. Unfortunately this meanssome of the lift is being produced by an inefficient foil.Such a cratl is currently in service with the Thai navy.

Examining the static component of instability isilluminating: If the foil is forward of the CG, pitch upwill result in increased foil angle of attack, more foillifl and more pitch up moment, producing still morepitch and more foil lif?. If pitch up moment due to thefoil exceeds pitch down moment due to the aftwardsmovement of the waterline/keel intersection (resultingin afhvards movement of the planing center ofpressure) the vessel will pitch back until the foil sensesthe surface (or emerges) and loses lift.

The obvious solution is to move the foil aft sothat increasing foil lift due to increased angle of attackfrom pitch up produces a bow down moment.Unfortunately, this results in reduced efficiency as wellsince the proportion of lift carried by each componentis going to be distributed according to the relativedistance from the center of foil and hull pressure andthe center of gravity.

3

The hull will carry the lion’s share of litl at itsrelatively low eftlciency unless it is designed to riderelatively deep so that the planing lifi center is wellforward. The HWSTB designed by the authors in factoperated in just this condition and was stable.However, as noted above, the overriding feature ofarmored vehicles is that they are very heavy for theirlength, (the displacementilength ration of the HWSTBwas over 2,500) and no other marine vehicle would beso heavy.

STEPPED HULL

The stepped hull concept is obvious ti-omthisdiscussion. The foil is at the extreme stern of thevehicle and a step is provided forward of the CG. Thestep confines the planing lift to the forward part of thehull so that the relative position of the center ofgravity, the step and the foil control the proportioningof lifl between hull and foil. Bow up pitch of thevehicle produces a strong bow down moment, directlyproportional to pitch, that reduces the pitch much morerapidly than the movement of the center of planing lift.

h addition, the pitch damping of a lifting surfaceis proportional to the square of distance fi-omthe centerof pitch. The rotation in pitch in the bow up directionproduces a downward motion of the foil if it is aft thatadds to the forward motion vectorially. Thisdownward motion is seen by the foil as a rotation ofthe relative flow downward so that the effective angleof attack of the foil increases even more than the bowup rotation. This increases the foil lift proportional tothe rotation rate. The location of the foil well aftmeans that the moment is relatively large for a given

lift due to the long moment arm. Over all, therefore,pitch damping increases with the square of distance ofthe foil aft of the CG.

The step also means that the running attitude ofthe planing hull can be set at a trim producingoptimum lift. (This is one of the major advantages of astepped planing hull also.)

ANALYTIC STUDIES

Calculation of Resistance

A first order computer model of a planing hybridwas made by combining the Savitsky (1964) equationsand first order wing (lifting line) theory. For brevity,the details are not repeated here, but code for a similarprogram implementing this approach is presented byKarafiath.

01

-——

—.

A! +Y (UP)

ture an

as always zero by aligning it with the keel,because it is defined by foil position anddekalage as well, but geometry optimizationis easier by letting this change.

T Trim of the coordinate system from thedynamic waterline, positive bow up.

_ BX12 _

lFigure 1- Nomencla

The authors also developed some nomenclatureand conventions that we feel are usefid and are shownin Figure 1. The coordinate system is fixed to thevehicle with positive X forward and positive Y up withthe origin at the transom/keel or stepkeel intersection.The terms used in the figure are:

BX Planing beam, the effective beam of the eachhull, generally the beam at transom.

P The deadrise at the station chosen foreffective beam.

Stagger Location of the foil fore and aft, negative ifaft of the step or transom.

Gap Location of the foil below the transom,negative if below.

Dekalage The angle of the nominal foilmidline to the coordinate system. (Thesethree term derives from terms used withbiplanes.)

LCG The longitudinal center of gravity, positive ifforward of the step.

Lc The length of the wetted chine, includingwave rise.

Lk The length of the wetted keel,

302

.-

d Axes System I

Drag The angle of the keel with respect to thecoordinate system. This angle could be taken

Draft Draft of the origin below the dynamicwaterline.

The trim and draft, and hence the wetted chine,keel lengths and foil angle and position will changesuch that the vehicle is in equilibrium, but must beinitially assumed. The lift and drag forces andmoments are then calculated and compared to weightand thrust. If the sum of forces and moments is non-zero, the trim and draft must be changed and the forcesand moments are recalculated.

—- -. .—-—

——.

The program uses the Savitsky method asmodified by Blount/Fox (1976) including theiraddition for hump drag. In this case, though we basethe hump drag factor on the lift the planing surface isproducing rather than the total vehicle weight. Thisaddition is probably incorrect, but we felt it was thebest option available. The hump condition is not welladdressed by any simple approach. The wave makingof the bow section will interfere with the foils and theaft section will be planing in the wake of the forwardsection until the foils lift it clear. Good model testswill be required to examine this condition.

Foil lift and drag can be calculated provided thelimits of lifting line theory are met: The foil must havea large aspect ratio, be nearly elliptically loaded, bemoving reasonably fast, reasonably well below a freesurface and be of small dihedral and sweep. Sincethese limits also produce best performance any way,they are not important constraints for the model.DuCane (1972) gives the equations that we used tocalculate foil lift and drag under these assumptions andsummarizes the derivation of these equations.

The effect of struts is to change the effectiveaspect ratio of the foil by partial blockage of the tipvortex much as the horns of newer airliners do. Insome cases this effect is straight forward, in other casesit is very complex, and this represents the mainchallenge for calculation of performance for someconfigurations. DuCane gives equations for estimatingsome common cases. Tank tests or numeric flowmethods must be used for other cases. An “ELL” foilwith a strut is such a case. A foil with a strut of the

same chord aligned with the foil with a typical motorpod was found by the authors experimentally to havean effective aspect of about 1600/0the geometric value.Since a foil against an infinite wall has an effective

aspect ratio of twice the geometric one, this suggeststhat an aligned, fill chord strut is 60°Aof an infinitewall, which seems plausible.

Though DuCane gives methods for calculatingfoil section drag, this is also found in the standardliterature for specific foil sections and we use thesevalues.

The effective angle of attack of the foil is theangle between the zero lift angle of the foil and theincoming flow. This angle is generally negative; thefoil lifts with its nose down slightly.

DuCane also gives an approximation for spraydrag and Hoemer ( 1958) gives values for interferenceeffects.

303

Since many configurations of foils will requireunderwater housings for the propulsory or the controleffecters, there will generally be one or more pods.The drag of the pods is estimated based on the standardliterature with interference effects as required.

Minor terms such as aerodynamic effects, themoment due to thruster location and similar terms areadded as required.

The result is a computer program that gives atleast a fwst order approximation to the performance ofa planing hybrid hydrofoil. The program was designedto allow the user to either select arbitrary draft and trimor to automatically fmd equilibrium. The formeroption allows investigating quasi-static stability terms,particularly of unstable configurations.

The program also allows any vessel parameterto be modified at each assumed step. The mostimportant use of this feature is to modify dekalage,thereby simulating a rotatable foil. In order to allowfinding equilibrium, the program has a factor thatmultiplies each iteration correction by a user setfactors. This factor is a crude analogy of dampingand inertial terms, and to a very limited extent,suggests the character of the vehicle’s dynamicbehavior: If each succeeding draft and trim derivedby the misbalance of forces is multiplied by unity andthe model still finds equilibrium quickly, the vehicleis more likely to be dynamically stable. If the factoris small and the model repeatedly goes off intonumeric left field, the resulting vehicle may havedynamic stability problems. Though this is strictlyan intuitive issue, in the FMC HWSTB, the authorsfound just such a correlation between computer

simulation and tank tests.

Resistance ResultsResistance results are given in Figure 2 for a

simple comparison case:Length O.A. 20 Ft.Weight 2,000 LbsLCG 7 Ft.

(forward of the extreme aft end of the boat)Planing Beam 6 Ft.Deadrise 15 Degrees

The hybrid version has its step eight feetforward the extreme aft end of the boat and isequipped with a pair of two foot span by half footchord foils at the extreme rear end one half footbelow the baseline. The foil section is the GeneralAviation (Whitcomb) 1 section. The foil dekalage isset for several values.

.—.

____

I IDrag,LbsI 1’600 --- ----.—– Planinghull ,

,

400*

——k

200 –~–\– -‘-k -1Dekalage -- ‘-–” -

I

ODekalage

L20 30 40 —

Speed,Knots

lFigure 2- Resistance Comparison I

Figure 2 shows that the reduction in dragbetween the planing hull and the hybrid can be verylarge. It also shows that the resistance is verydependent on the dekalage angle, suggests thatvariable dekalage may be required and that anincorrect dekalage actually increases drag.

PT

Figure 3 shows the proportion of lift tlom thefoil for each dekalage and speed. The resistancecurves are plotted in the background to show thecorrelation between the percentage of foil lift andminima of drag. It is interesting to note that in thisparticular case, the minimum drag is not associatedwith maximum foil lift. Figure 4 suggests thoughthat it seems to be more related to the condition whenthe foil percentage of drag is a maximum.

In fact this is a subtle clue. Though theconfiguration in question has reduced resistancecompared to a planing craft, it is actually non-optimum.

The initial selection of parameters was such thatit drove a local optimization away from the bestcondition, which would have more load supported bythe foil, but this is not possible with the basicassumed parameters.

..

ercentof Foil Lift in ~otal Lift Forces

ODekalage ~

6~oA------ _.– . . ‘

T

—---- +-1 Dekalage—- ~~~ ------

1- ,~k:

t

-2 Dekalage,,/. .,—

1,1/’

1 ‘40% – ...-_... .-

~ -“+- 1-1

/ /’..— /\,,

~Y””-”-”-”””

>?/z ‘- ;-3 Dekalage

.=-.—.

20% –—- -- –— :-— -/—----+ -- ----- ‘-–--””

L 20 30 40 —Speed,Knots

Figure 3- Drag Minima Vs Foil Lift II 1

Percentof Foil Dragin Total DragForces i’ 1

60% ---—~--1

40% –—----—

-——----- r-/’ “-”.–+––.– -

/’/,

/’

1i

/ :’/,

–-+—-–-, ——---+-—- --—/ \, /1 ,/’ i

-3 Dekalage

\-/

Y_-

*()% - --- — .

I ODeka

/’ )% ‘/

=,--”

_/.

J “---;” ;

“““-2Dekalagf-1 Dekalage -Ige \

I 20 30 40

Speed,Knots

Figure 4- Drag Minima Vs Lift I

These figures serve as a warning. Exploringvarious designs with the model shows that it is quiteeasy to design a stepped hybrid hydrofoil with foilsthat are the wrong size, combined with a center ofgravity in the wrong place so that the resistance iswell above the comparable planing hull.

304

so they are also very large.

—

In fact, changing the foil from two “en” foilstotaling four feet of span to a single six foot span boxfoil produces a hybrid that has several times the dragof the equivalent planing hull for such a wide rangeof dekalages that no acceptable configuration wasfound.

There are many possible combinations ofparameters with this concept, few of which work, sothat use of a computer model is vital. Perhaps this iswhy the hybrid concept did not seem to catch onmuch earlier.

FoilsThe use of the GA(W)-1 (McGhee & Beasley,

1973) in the example is worth discussing further.This section is one of a class of supercritical sections(also known as “barn roof’ sections) that aredesigned by computer to achieve maximum liftcoefficient with minimum possibility of stall. Thephrase “barn roof’ refers to the fact that the sectionachieves uniform chordwise load distribution at thedesign lift coefficient, resulting in a lower peakpressure and a higher percentage of pressure forwardand aft of the peak. This foil was used in theHWSTB for the same reason.

The pressure distribution curve is flatter thannormal, and more filled in. As a result these sectionsare highly resistant to stalling. This is critical forsome stepped hybrids as the lifted tail means that thevehicle may require extremely high Iitl coefficientsin the takeoff mode. Cavitation is also reduced. Thereduction of the peak pressure delays cavitationcompared to NACA 65 series sections, at least at therelatively low speeds that hybrids operate. Suchsections were not available at the time the stepped

hull hybrid hydrofoil was initially developed and theproblems of low speed lift may have contributed tothe lack of popularity of the concept.

Dynamic StabilityThough it seems unlikely that a stepped hybrid

will develop pitch dynamic instability, there has to bea definitive criteria.

Martin (1978a) and Payne (1974) havedeveloped theoretical methods of determiningstability for high speed planing boats. The methodsare similar, though the authors happen to have usedMartin’s method as a basis. Extending theseequations to the case of a hybrid hydrofoil merelyrequires adding the effect of the hydrofoil to thevarious terms of the equations. Again, in theinterests of brevity, the only an overview of themethod is presented

30

The pitch equation of motion has six terms.Three are the direct terms comprising pitch, pitchvelocity and pitch acceleration times the spring-liketerms due to planing lift changes due to trim; thedamping terms; and the mass and mass-like termsrespectively. The other three are the cross productsproducing moments from vertical position, velocityand acceleration.

The heave equation has a similar set of threeterms producing vertical force due to verticalkinematics and three producing vertical force due topitch kinematics.

The terms due to the hull determined stripwiseintegration of the sectional properties based ondeadrise, chine beam and whether the particularsection is fully immersed at the chines, wetted to thechines by spray or has dry chines.

The additional terms due to the foil are derivedfrom ship control methods (Crane, et al, 1989).Since the rate of change of foil lift due to change inangle of attack (“lift slope”) and the change of dragwith respect to angle of attack are known fi-omtheresistance calculation, the important terms can beeasily calculated.

The damping moment coefficient of the fin dueto pitch velocity is the most important term and isdirectly proportional to the lift slope and the arm ofthe foil to the center of gravity squared. For anypractical hybrid both of these will be large, thusproducing very large pitch damping.

The other important terms in pitch are themoment due to pitch acceleration, and the momentdue to pitch. Both increase with the arm squared andthe moment due to increases with lift slope as well,

The darnping moments and forces in the heavedirection are also substantial, because again, they aredominated by the efficiency of the foil, which we aretrying to maximize for the sake of reduced resistance.

Other terms are quite small: Damping heaveforce due to heave is the change in lit? due to depthwhich is negligible for a foil more than one chordbelow surface. The pitch forces and moments due tothe foil rotating around its own center are likewisesmall. The heave forces on the foil due to heaveacceleration are very small compared to the similarterms on the hull.

The actual form of each term will vary becauseit is customary to non-dimensionalize the terms intostability derivatives. The selection of thecharacteristic values used to non-dimensionalize willtherefore change the expressions for the derivatives.

5

———

—

Once the various derivatives are known, theheave and pitch equations are combined and solvedassuming the solution is a sum of exponential. Theresult is a fourth order polynomial. Martin hasassembled the various derivatives into thecoefficients of the polynomial, so adding the foilsmerely requires adding the foil term to theappropriate hull terms and following Martin’sprocedure.

The resulting fourth order equation has fourroots which can be real or complex. A complex rootcorresponds to an oscillatory motion, and if the realpart is positive, the resulting motion growsexponentially, indicating dynamic instability. Sinceeach root corresponds to a different mode of motion,all four have to be found and examined to ensure thatthere is no mode of motion that is unstable. Such aset of equations has to be numerically solved, so noinsight can be gained directly by examining ananalytic solution, Instead, numerous systematicvariations have to be examined. This task has beenplaced in our inbox.

Though we are still trying to get such a codeworking reliably, for any practical stepped hybridconfiguration the method clearly results in very largevalues for those coefficients that characteristicallyproduce stability; i.e. the damping terms.

SeakeepingMany high speed craft are limited by motion in

waves rather than power. Methods to analyzemotions will be required to determine limitingconditions for crew and passengers, and structuralloads.

Martin (1978b) has demonstrated how this

proceeds for pure planing cratl by extension of thestability method. This can be extended in a similarfashion by adding the foil terms for forces andmoments from waves, but is worth noting that thefoil excitation due to waves is relatively smallbecause the foil is effected only by the orbitalvelocity of the waves and very slightly by theelevation of the foil beneath the waves. Thevelocities are small compared to the vehicle speedand the effect of elevation is minimal if the foil is insubmerged below a chord length. The particlevelocity effects and wave height effect also areopposed, so the net force is even smaller.

It is difficult to make general predictions aboutthe seakeeping of stepped hybrid hydrofoils becausethis is even more profoundly affected by optimizationbut there are two important points that suggest goodseakeeping is possible:

30

.-

First, let’s examine what has become the normfor planing and semi-planing craft designed for goodmotions in waves. A useful dry land analogy is the“chopper” motorcycle. The extended forks of thebike place the front wheel, which “senses” the roadsurface, well forward of the center of gravity and actas a soft spring connecting them to the rest of thebike. Thus, the angle of pitch induced by the frontwheel striking a bump is reduced because it movesthe same distance up, but acts on a longer lever arm.The rate of acceleration of pitch is also reducedbecause the instantaneous force of the bump isreduced by the springingness of the forks. It actsthrough a longer time, though and thus achieves thenecessary rotation for the bottom of the bike to avoidthe bump. Motions in head seas dominate theproblem of seakeeping for fast craft, because at highspeeds, all seas are head seas, so the analogy to ridingon a bumpy road is very apt. Offshore racing craft,and “wave-piercing” catamarans both approach theproblem of reducing motions in head seas in the sameway, by moving the sensitive load as far aft aspossible and by reducing the rate of lifl force withrespect to immersion of the forward sections, usuallyby making them narrow, with high deadrise.

However, if a planing hull strikes a wave, theforce induced on the hull by the wave is primarily atthe intersection of the hull and the instantaneouswater surface. As the hull travels, this intersectionmoves aft, and the force becomes larger as the hullgets wider and deeper. The craft rotates more andmore, and the rotation induced by the wave alsotends to move the aft end down so that thedisturbance increases. In extreme cases, racing craftare sometimes thrown completely upside down, with

the stem passing under the bow. This hasn’t beenseen in fast ferries, but some of the tendency to overrotate probably contributes to increased accelerations.In addition, once the wave passes under the hull, the

pitch rotation changes and the craft can over rotatesdown into the next wave, increasing the followingpitch up.

In contrast, a stepped hybrid hull will initiallyrotate, but the rotation will increase the angle ofattack of the aft foils, which lifts the vehicle bodilyupwards from the rear and reduces pitch acceleration.The hull is therefore “anticipating” the oncoming

wave and goes over it like a horse clearing a hedge.This motion has to be carefully tuned to theanticipated wave environment for optimumperformance, but it is clear that a properly designedstepped hybrid hydrofoil could have excellentmotions.

6

.—

—.

Second, with the wide range of parametersavailable to the designer, it is clear that there isconsiderable latitude to optimize for motions. A hullform with very high deadrise, low freeboard planinghulls forward and foil support aft could be developedwith very good motions because the foil would bearthe majority of the load and the hulls could berelatively inefficient, hence relatively soft riding. Ina pure planing hull, the designer has to loseefficiency by accepting a high deadrise, soft ridinghull. The cost of non-optimum lift production for thesake of seakeeping would be much less for a hybridhydrofoil.

PropulsionA problem of hydrofoils that hybrids share to a

significant measure is that of propulsion: Getting theforce into the water often requires passing it throughthe struts which is costly in terms of money,appendage drag, complexity and efficiency.Hydrofoils use mechanical, electric and hydraulicdrives to props on foil pods, jets taking suctionthrough the foil, and shafts from the hull.

Each of these methods has problems. Jetstaking suction through the strut add strut drag andcause loss of velocity head. Hydraulic and electricdrives add cost and efficiency losses. Mechanicaldrives have lower efficiency losses, but are complexand costly and have large, highly loaded bevel gearsin large, drag-producing pods. All types of pod

mounted proptdsors produce drag due to the frontalarea of the submerged drive components.

The possibility that variable dekalage mayberequired on a hybrid adds another problem toconfigurations with props on pods and mechanicaltransmissions. The most obvious way to producevariable dekalage, especially if the foils need to beretracted, is to rotate the foil and strut assemblyaround its connection to the hull. Transmittingpower through such a variable angle joint is possiblebut adds additional complexity not present in ahydrofoil strut that is mechanically connected in onlyone position.

There is some consolation that the struts of ahybrid are somewhat shorter, but this is onlyimportant for through-strut jet drives, and jet drivesrequire higher flow rates for efficiency at the lowerspeeds of a hybrid.

307

.— -.

However, unlike a pure hydrofoil, a hybrid canbe propelled by hull mounted components. A jetdrive could be mounted in the forward planing hulland discharge at the step. A prop shaft couldpenetrate through the step as well or surface piercingprops could be mounted on or below the raised tailand dip down to the water. This gives some addedversatility to the hybrid concept that a pure hydrofoildoesn’t have.

The choice of propulsion method is economicand operational and will be determined by themission. The hybrid offers wider latitude for lesscostly methods than a pure hydrofoil, but requires aninnovative approach to the issue.

EXPERIMENTS

Thomas Edison performed over a thousandexperiments in the course of inventing the electriclight. Most were failures in the sense that they didnot produce a working light, but Edison regardedthem as successes in that he learned from each ofthem. We have taken solace in this, because ourmost ambitious experiments also failed to produceworking craft, but we learned a great deal and feel wenow understand much more about the practicalproblems of stepped hybrid hydrofoils.

Small ModelsMr. Kenneth Foster, formerly CEO of Munson

Manufacturing, met the authors and becameinterested in this concept, and as a sanity checkpurchased two identical radio remote controlled

model boat kits. He modified one to a stepped hybridhydrofoil configuration and ran the two side by side.He found the hybrid configuration was oftensubstantially faster than the unmodified one thoughthe placement of weight, size and angle of the foiland other parameters made substantial differences.Some configurations were in fact much slower. Henever noted any pitch or roll instability anddiscovered that getting the hybrid to turn was asubstantial problem, as his model had no ailerons,only rudders.

This experiment was not well controlled and issubject to many objections, however, it showed thatthere is something to the claim that a stepped hybridcan be faster than a comparable planing hull forsimilar power, can be stable in pitch and roll, and isstrongly dependent on the details of configuration,particularly foil angle.

Though not intended as an experiment per se,

Since then the authors have built a number ofsmall self-propelled models and found much thesame thing. There were numerous cases of hybridsbeing twice as fast as similar pure planing craft, butthe hybrid definitely could be slower if improperlyoptimized.

One phenomena we ignored until later alsoproved to be an important hint. The small modelspropelled with model airplane motors announced thatthey had come onto plane by a sudden change in tonefrom a growl to a high whine.

Manned ModelOne of the authors (Duffty) built an eight foot

plywood stepped hull, modified from plans for askiff. The hull was fitted with a box foil made fromaluminum by taking plane cuts with a millingmachine and hand finishing.

The model was built prior to completion of theresistance program and thus not optimized. The foilsection is flat bottomed similar to a “Clark Y“, butactually not a specific tabulated section, due to limitsof the manufacturing process. It has provision forchanging the foil angle and vertical and horizontallocation. It was intended to be powered by a smalloutboard motor or to be towed and is provided withbuoyancy and other safety features to be manned.

The initial test was run in San Francisco Bay offBerkeley. The weight of the motor and driver

resulted in an aft static freeboard of less than twoinches and the vehicle swamped during takeoff.Additional foam flotation was added but the vehiclefailed to get over hump.

The outboard was not able to produce fullpower, partly as a result of being swamped and partlydue to being in excess of thirty years old. However,there was a key propulsion problem which might alsooccur in full size craft: The propeller was suited forhigh speed and had so much pitch that it would notallow the engine to achieve full RPM, and hence itdid not have the power required for coming uponhump, even if the engine had the necessary power atfill RPM. This was the warning of the smallexperiments: The hump condition is critical forhybrid hydrofoils and some radical provisions maybe required to allow engine matching in both thehump condition and the running condition.

30

To alleviate the problem of motor performancethe boat was towed. Finally, while coming up tohump, the vessel rose, began to fly, but took a suddenroll to starboard, dipping the bow. Additionaloutboard foils were added with dihedral to gain rollstability. Again it failed to go over hump, butseemed to be trying to bury the bow.

After the last experiment, the resistanceprogram was complete and the model performance isbeing studied with the program. Further experimentsare being planned.

The problem of burying the bow is of someconcern, especially considering the sudden roll tostarboard: It is known that weight too far forward ona planing surface can cause the rounded portions ofthe forward buttocks to enter the water and cause rollinstability (Codega and Lewis, 1987, Cohen andBlount, 1986). This is because the flow around thecurve causes suction. The skiff used was, of course,designed to be eight feet long and run by an operatorin the stem. The forward buttock lines are thus quiterounded and could create a condition which might bethe source of the roll instability and the failure tocome over hump. This phenomena may presentspecial issues in the design of the planing portions ofstepped hybrid hulls,

Human Powered Vehicle

when both authors were at FMC, they made aproposal that FMC explore the commercial viabilityof the hybrid as an alternative product line.

Since FMC had marketed commercialhydrofoils in the early 60’s this was received withsome interest. A demonstration was conceived thatwould definitively show the concept, its niche as anintermediate speed vehicle and incidentally provide acorporate recruiting publicity video. FMC thereforegranted limited funding (materials and use ofcompany facilities) for a volunteer attempt at theDuPont prize for the first human powered vehicle toachieve twenty knots. At that time, FMC employed amember of the US Olympic sprint cycling team as amechanical engineer, and had a substantial advancedtechnology base in ultra high strength low weightcomposites, so this project seemed ideal.

8

.-

—-

The hybrid concept is suited to speed rangessomewhat lower than pure hydrofoils. Brooks,Abbott and Wilson (1986) note that an ideal humanpowered hydrofoil has too high a takeoff power attoo high a speed to be practical. The hybrid conceptis better suited to the speed ranges required for theDuPont prize. The vehicle was never completed butthe design process and limited experiments alsotaught lessons.

First, the optimum vehicle obviously had verylow hull loading. This, in turn, resulted in a verynarrow planing surface on the forward hull. Inretrospect, this would probably have resulted in a rollstability problem at speed, because the stabilizingmoment of foil dihedral is related to the cosine of thedihedral angle and is thus very small for acceptableranges of dihedral. This in turn suggests thatmonohull stepped hybrid hydrofoils may not beoptimum for resistance.

Second, the pods were a major source of drag,as much as the induced drag of the foils. Bevel gearsto transmit the power levels required had a diameterin excess of an inch, even at a lifetime of a fewminutes, resulting in a pod on more than three inchesin diameter. This reemphasized the problem ofpropulsion.

In the middle of the project, the Soviet Unionbegan to collapse and the need for new armoredvehicles, and hence corporate recruiting, reduced.Worse, the critical component, the “engine”, found abetter job with another firm, so the project endedwithout hitting the water.

The participants are still in contact and have

many components left over from the project and hopeone day to reassemble and try again.

Future TestingClearly, model tests capable of measuring

resistance, speed and motions with some accuracy arerequired.. The resistance program methods must be

checked and validated.

. A deliberately unstable model should be runto validate the dynamic stability analysis.

. The issue of bow curvature could beexplored.

. There is a possibility that the spray from theplaning hull will impinge on the raised aftportion and cause drag not accounted for inthe simple theory.

30

—-

● Use of the Blount/Fox hump drag factorcould be checked and the hump regionperformance can be examined.

. Measuring trim angle and varying foil andhull parameters would allow exploring theeffect of the hull flow on the foil.

● The effect of alternative planing hull formsnot covered by the Savitsky equations can beexplored.

. Wakes produced by hybrids can be comparedto comparable planing hulls.

LESSONS LEARNED

Our experiences suggest the following pointsfor the design of fiture stepped hull hybridhydrofoils:. A monohull hull form is probably

inappropriate for a combination of optimumefficiency and adequate stability.

. High lift foil sections are probably required.

. Propulsion matching is an important problem.

. Optimization is a considerably more difficultproblem than it fust appears and is critical.

. Hull form design has subtleties we don’t fillyunderstand.

However, we should also tabulate the practicalproblems and advantages of the stepped hullhydrofoil● The struts are a potential problem for

collisions with debris, though they aresomewhat shielded by the forward hull andshorter than those of a hydrofoil.

. The strut connections are heavily loaded,though not as much as the forward strut of ahydrofoil.

● Severe waves will load the hull more than ahydrofoil (though this reduces the load on thestruts).

. Propulsion will always be more complex thana planing hull because of the hump powerproblem and the need to transmit power to thewater below the hull.

. When waterborne at slow speed, the hull willeither be severely trimmed bow up or requireballast or movement of weight forward.

9

and fluid torque converters or electric couplingsduring the pre-takeoff run, then use the stored powerfor the few moments of high demand.

——

, Motions are reduced compared to a planingboat, but they will never be as good as ahydrofoil with automatic controls.

● The running draft at speed is more than aplaning boat.

. The concept is inherently a single speed one.Performance at other than top speeds will bepoor.

. The foil is not a common component and willbe costly unless methods for low cost foilmanufacture can be developed.

STEPPED HYBRID HYDROFOIL FERRIES

A high speed ferry is an obvious mission for astepped hybrid hydrofoil

A ferry is a two speed vehicle, and the hybridconcept is well suited for this.

Most practical new ferry routes in the USrequire a speed on the order of thirty to forty knots tocompete with automobiles provided the speed can beachieved at an acceptable level of cost and reliability.This appears to be the optimum range for this

concept, so a stepped hybrid may be able to achieve alower cost and better reliability at these speeds thanhydrofoils, SES’S,or planing boats.

The stepped hybrid concept is much lessdependent on size for speed and seakeeping than aconventional planing hull, so smaller, less expensive

ferries are feasible. This allows either more ferrieson a given run or use of ferries for runs with muchless traffic.

A number of high speed ferries are limited bywake damage to the shore. As a result, they can onlyrun at speed for a small portion of the route. Ahybrid should produce substantially less wake than aplaning monohull or even an SES because the foilsgenerate substantially smaller waves. Whether thiswould be enough reduction is a second question, butthe comparison of wakes should be a goal of futuremodel test programs.

Proposed DesignThe authors envision a 80 passenger ferry that

would have a pair of narrow hulls well forwardterminating near midship with a central hull aft,raised above the catamaran hulls. (Figure 5)

310

The configuration would be reminiscent of a“picklefork” three point hydroplane, but the aft pointwould be a “U” foil running under the aft end of thecenter hull. The foil would be pivoted at the top endfor dekalage adjustment and could be provided withan upper ladder foil that would be dry at full speed,but help in takeoff. The aft end of the cross deckwould be immersed at low speeds to minimize bowup trim and provide additional lift to get over humpand would be fitted with a wedge or flap to increaseplaning lift in takeoff mode.

The craft would be propelled by surfacepiercing drives or waterjets mounted in the transomsof the forward hulls. Since waterjets do not tend tooverload the engine at low boat speed and the jetsuction would always be immersed in this position.Waterjets are also well suited to the 30-40 knotspeed range. Surface piercing props are also anattractive alternative. The use of surface piercingpropellers would require some means to address theproblem of overloading the engines at hump, but ZFhas recently placed a range of two speed gearsets onthe market to address similar problems for planingmonohulls. If the power required to take off isexcessive, an aft engine could be provided on thecenter hull.

An other alternative is to use a high speedcomposite flywheel for storing power. The engineswould spin up the flywheel through high ratio gears

Foils for such a craft would be less expensivethan for a conventional hydrofoil. High strengthstainless steel is the normal material for fully flyinghydrofoils, to resist cavitation, to provide adequatestrength and to resist corrosion, but it is expensive,both for materials and to fabricate.

For example, the HWSTB had a pair ofaluminum foils only a few feet long, but eachrequired over twenty-four hours on a very large CNCmill to produce.

The lower speed of a hybrid means thatcavitation is not such a problem and that the foils willbe larger and hence have greater section thicknessthan a normal hydrofoil. The high lift sections alsohave relatively thick sections as well.

—.

tsb

aswsrcce

—

all

Francisco Bay Area to serve a multipoint commute

/&/

#p’

Figure 5-

Our studies indicate that adequate foils could bemade by casting plastic materials over a welded steelcore. Polyurethane with a Shore hardness in the 80-90 range cast over stainless steel cores has been usedfor rudders of planing craft. This material might befamiliar to some of us as it used for roller bladewheels. It also has also been used in tape and paintedon form as a protective barrier to resist cavitation and

Sm

abrasion in slurry service. It is relatively inexpensiveand can be cast in simple molds at room temperature,so that heat treated steels can be used if required forstrength.

Banking control would be achieved by blowingair over the top of the foil through tubes cast into it,hereby eliminating the need for expensive controlurface actuators and the drag on their housingselow the surface.

Such a crafi would be about 75,000 lb. full loadnd require only about 675 EHP to achieve 35 knots,o a pair of diesel engines in the 700 BHP rangeould be sufficient. Such comparison are always

uspect, but about 2000 BHP (total) would beequired propel a conventional monohull planingraft of the same weight to the same speed. Aatamaran of the same weight with two 700 BHPngines would only be able to achieve 26-27 knots.

311

,, ‘~/“ L“ /’

--”z’/</’,..-”

Ferry

The main merit of a hybrid solution for this typeof craft is in the relatively small size (and cost) forsuch high speeds with acceptable seakeeping asdiscussed above. This makes it uniquely suited forurban service in competition with overcrowdedfreeways.

We intend this craft specifically for the San

route, with loops connecting San Francisco with thenorthern East Bay (Berkeley), the mid Peninsula(Redwood City) and the mid-southern East Bay(Hayward) and other loops connecting Marin County,Vallejo, San Francisco, and Richmond.

The small size of the ferries would allow themto enter recreational marinas and travel intoshallower water at high speed. The low cost wouldallow many ferries on a route so that the delaybetween ferries would be small thus reducing theprobable trip time (which includes some probabilityof missing a ferry).

The small passenger load would also interfacewell to city buses or shuttles. There is no point ingetting off a 600 passenger ferry and waiting in atraffic jam often buses and twenty shuttles in theparking lot.

...—

,

Short runs in crowed areas like San Franciscoalso involve a relatively high propotion of time atlow speed, Our studies indicate that the overall triptime saved for typical voyages is only a few percentfor speed increases above about 35 knots, thoughtrips at speeds below 30 knots are substantiallyincreased by top speed reduction.

Another advantage of a relatively small craft forareas like the Bay Area arises from the current trendtoward dispersion of jobs and housing. The growthof “second generation” businesses in the suburbs(founded by employees of first generation businesseswho got tired of commuting) has made the commutepattern completely disorganized. The classic “NewYork” model of workers commuting from all overinto one point is almost dead in many areas, becauseas people change jobs to work for these firms, theydon’t move, they just commute in differentdirections. This is occuring in many other areas. Inthe Chesapeake, firms are beginning to locate on theEastern Shore, so a counter commute eastward hasbegun. One of the authors had just such a commutehistory, living in San Francisco and working for awhile in Marin County, then in San Jose, commutingagainst the classic flow. Small ferries can addflexibility to the commute pattern, because transfersno longer involve long waits, and one or two peculiarflows can be accommodated by one or two runs eachway per day.

Perhaps the most important advantage of smallsize, however, is the possibility of “growing” aservice. A conventional planing hull or catamaran

must generally carry several hundred passengers tobe large enough to achieve acceptable speed andseakeeping. This means that the operator has to havesome notion that there are a thousand or morepassengers per day to be able to run a systemeconomically, and those passengers have to appear intime to reverse the operator’s cash low before he runsout of money. A small, fast, inexpensive ferry allowsthe service to be built up, changing commuter’s habitgradually. This also means that small enterprises canstart these ferry lines to serve a specific route, such asone out of a new real estate development. Theseservices could also be community owned enterprises.

A stepped hull hybrid might be competitive inother services, but as the optimum craft gets larger,the speed and seakeeping advantages of the hybridbecomes less significant, and the difficulty offabricating large foils begins to be a concern.

31

We believe that the small “mosquito boat” typeservice we envision might be an optimum role forthis concept, and the stepped hull hybrid, in turn, isthe optimum concept for such a service.

FURTHER DEVELOPMENT

This paper is definitely a report of work inprogress. Much further development is required todetermine if this technology has merit and in whatcases.. Resistance calculation methods have to be

verified and extended to the hump speedregime.

● General guidance for optimization has to bedeveloped.

● Stability and seakeeping analysis techniqueshave to be developed and verified.

● Structural criteria have to be developed,particularly for foil fatigue loading. This inturn requires seakeeping analysis techniques,and probably cooperation of one of theclassification authorities.

● Strawman ferry designs for various serviceshave to be done and evaluated.

● Methods to produce foils at low cost have tobe proven.

CONCLUSIONS

The stepped hybrid hydrofoil is presently a little

known historic curiosity. It has merits in reducedresistance compared to planing hulls at lowercomplexity than pure hydrofoils. It may have meritsin seakeeping and other operational areas. Its currentstatus may be due to being eclipsed by the purehydrofoil, but it should not be viewed as a partial stepto the hydrofoil. It is a valid concept on its own withits own special characteristics and capabilities.

The authors have presented some of the issues,proposed methods to analyze critical areas ofperformance, and shown concepts to address keyareas of concern in design.

Stepped hull hybrid hydrofoils especially meritconsideration for high speed ferry service forpartially sheltered runs were seakeeping is aconsideration but not an overriding one, there arefactors limiting size on a given run, such as trafficdispersion and moderately high speeds are required.

2

We suggest that San Francisco Bay presents justsuch an opportunity, and there are probably otherservices with similar characteristics worldwide.

ACKNOWLEDGMENTS

The authors would like to acknowledge FMCCorporation for their support and generosity, themembers of the FMC Human Powered Vehicleproject, especially George Thomas and Bruce Wadefor their expertise in composite structures.

They would like to thank the members of theCal Sailing Club for support in the manned prototypetesting, especially Paul Kamen, who has contributedmuch technical insight as well. They would also liketo thank Kenneth Foster for his interest and forbuilding and testing the radio remote controlledmodels. Michelle Barry deserves credit for preparingsome of the figures.

The views and opinions expressed herein are those of

the authors and are not to be construed as oficialpolic>~or re$ecting the views of the U. S. Coast Guard

or the Department of Transportation.

REFERENCES

Blount, D.L. and Fox, D.L, “Small Craft PowerPredictions”, Marine Technology, January, 1976,Society of Naval Architects and Marine Engineers

Blount, D. and Codega, L. “Dynamic Stability ofPlaning Boats”, Marine Technolo~, January, 1992,Society of Naval Architects and Marine Engineers

Brooks, A. N., Abbott, A. V., and Wilson, D. G.“Human Powered Watercratl”, Scientl& American,

December, 1986

Cohen, S. and Blount, D., “Research Plan for theInvestigation of Dynamic Instability of Small High-Speed Craft”, Transactions of the Socie~ of Naval

Architects and Marine Engineers, 1986

3

Crane, C. L., Eda, H., and Landsburg, A.,“Controllability”, Chapter IX, Principles of Naval

Architecture, 1989, Society of Naval Architects andMarine Engineers

DuCane, CDR P. M., High Speed Crajl, 1974, Davidand Charles Ltd., Devon

Hayward, L., “The History of Hydrofoils”, Hovering

Crap and Hydrofoils, July 1965

Hoerner, S. F., Fluid @amic Drag, 1958,

(Published by the Author)

Karafiath, G., An Investigation Into The Performance

of NSRDC Model 5184 Configured as a Partial

Hydrofoil Supported Planing Craft and a Comparison

with a Powering Prediction Technique ReportSPD-585-01, December, 1974, NSRDC

Kunitake, Y., Nojiri, T., Kurihara, K. “PlaningBoat”, U.S. Patent 5,002,004, March 1991

McGhee, R. J., and Beasley, W. D,, Low-Speed

Aerodynamic Characteristics ofa 17-Percent-Thick

A irfoii Section Designed for General AviationApplications, TN D-7428, NASA, December, 1973

Martin, M., “Theoretical Determination of PorpoisingInstability of High-Speed Planing Boats”, Journal of

Ship Research, March, 1978, Society of NavalArchitects and Marine Engineers

Martin, M., “Theoretical Prediction of Motions ofHigh-Speed Planing Boats in Waves”, Journal of Sh@Research, September, 1978, Society of NavalArchitects and Marine Engineers

Meyer, J. R, “Hybrid Hydrofoil TechnologyApplications”, HPMV ’92, June, 1992, American

Society of Nmal Engineers

Payne, P., “Coupled Pitch and Heave PorpoisingInstability in Hydrodynamic Planing”, Journal of

Hydronautics, April, 1974

Savitsky, D., “Hydrodynamic Design of PlaningBoats”, Marine Technology, October, 1964, Society ofNaval Architects and Marine Engineers

Smith, B., The 40-Knot Sailboat, 1963, Grosset &Dunlap, NY

13

. -—