EP1326773A1 - Transonic hydrofield & transonic hull - Google Patents
Transonic hydrofield & transonic hullInfo
- Publication number
- EP1326773A1 EP1326773A1 EP00922123A EP00922123A EP1326773A1 EP 1326773 A1 EP1326773 A1 EP 1326773A1 EP 00922123 A EP00922123 A EP 00922123A EP 00922123 A EP00922123 A EP 00922123A EP 1326773 A1 EP1326773 A1 EP 1326773A1
- Authority
- EP
- European Patent Office
- Prior art keywords
- hull
- waterplane
- stern
- transonic
- adjacent
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Withdrawn
Links
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B63—SHIPS OR OTHER WATERBORNE VESSELS; RELATED EQUIPMENT
- B63B—SHIPS OR OTHER WATERBORNE VESSELS; EQUIPMENT FOR SHIPPING
- B63B1/00—Hydrodynamic or hydrostatic features of hulls or of hydrofoils
- B63B1/02—Hydrodynamic or hydrostatic features of hulls or of hydrofoils deriving lift mainly from water displacement
- B63B1/04—Hydrodynamic or hydrostatic features of hulls or of hydrofoils deriving lift mainly from water displacement with single hull
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B63—SHIPS OR OTHER WATERBORNE VESSELS; RELATED EQUIPMENT
- B63B—SHIPS OR OTHER WATERBORNE VESSELS; EQUIPMENT FOR SHIPPING
- B63B1/00—Hydrodynamic or hydrostatic features of hulls or of hydrofoils
- B63B1/02—Hydrodynamic or hydrostatic features of hulls or of hydrofoils deriving lift mainly from water displacement
- B63B1/04—Hydrodynamic or hydrostatic features of hulls or of hydrofoils deriving lift mainly from water displacement with single hull
- B63B1/06—Shape of fore part
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B63—SHIPS OR OTHER WATERBORNE VESSELS; RELATED EQUIPMENT
- B63B—SHIPS OR OTHER WATERBORNE VESSELS; EQUIPMENT FOR SHIPPING
- B63B1/00—Hydrodynamic or hydrostatic features of hulls or of hydrofoils
- B63B1/02—Hydrodynamic or hydrostatic features of hulls or of hydrofoils deriving lift mainly from water displacement
- B63B1/04—Hydrodynamic or hydrostatic features of hulls or of hydrofoils deriving lift mainly from water displacement with single hull
- B63B1/08—Shape of aft part
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B63—SHIPS OR OTHER WATERBORNE VESSELS; RELATED EQUIPMENT
- B63B—SHIPS OR OTHER WATERBORNE VESSELS; EQUIPMENT FOR SHIPPING
- B63B1/00—Hydrodynamic or hydrostatic features of hulls or of hydrofoils
- B63B1/32—Other means for varying the inherent hydrodynamic characteristics of hulls
- B63B1/40—Other means for varying the inherent hydrodynamic characteristics of hulls by diminishing wave resistance
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T70/00—Maritime or waterways transport
- Y02T70/10—Measures concerning design or construction of watercraft hulls
Definitions
- This invention pertains to hulls for.ships, yachts, submersibles, seaplane
- TH Transonic Hull
- Displacement hulls sustain boat weight by buoyant lift. As designed in the
- Displacement hulls are very efficient well below hull speeds, having a weight-to-drag efficiency ratio of the
- the hull speed of a 50 foot hull is 9.5 knots, but for 300
- a bulbous bow may slightly decrease drag at
- buoyant lift distinct from buoyant lift, can overcome the speed limits of displacement hulls, and
- ratio of boat weight to resistance is only about 8. This is less than half that of a
- the semi-planing hull usually has a Vee
- semi-planing hulls can generate the appearance of a "flat" wake at high speeds, their lift is generated by a combination of buoyancy and dynamic forces, which is
- the semi-planing hull is an inefficient hybrid: at slow
- planing hull has a rougher ride than a displacement hull, with less
- rowing shell relies on oars for lateral stability. With a length-to-beam ratio of 30, its wave-making resistance is only 5% of the total at 10 knots, but its weight-to-drag
- sailplane with a wing span-to-chord ratio of 25. It can operate at weight-to-drag
- SWATHS are also multihulls which rely on totally submerged primary
- underflow prevent pitch up and eliminate stern wave, and favors the recovery of underbody energy as well as that from following seas.
- underflow prevent pitch up and eliminate stern wave, and favors the recovery of underbody energy as well as that from following seas.
- sailboats In the case of sailboats,
- TH also provides special hydrodynamic coverage of roots of appendages
- the transonic geometric shape TH is shown in Figures 12a, 12b, 14a and
- the transonic hydrofield TH is shown in Figures 13a, 13b, 13c, 15a, 15c, 14b,
- One critical objective of this writer's invention is to formulate the
- hulls or is it a statistical property of hulls as designed in the past and
- slender shape of Fig. 3 could be altered to a "waisted" canoe shape, if Dr.
- This sinkage includes a downward pressure force (suction) generated
- planing boats may not generate
- the added friction term is a function of the wetted length Lw (assume unit
- weight to drag is significantly less than 10, and close to 8 at speeds of the order
- the ratio 8 to 1 also compares adversely
- planing hull reduces their poor efficiency in a marked way.
- the corresponding hull transcends, in a pure buoyant lift regime
- TH Transonic Hydrofield
- TH transonic hull
- hulls also exhibit the capability of developing, in the displacement lift mode
- anti-planing at speed ranges normally considered suitable for
- TH pertains to a three-
- TH is a unique three-dimensional flow field
- the writer shapes the TH first, and then shapes the
- control system or, indeed, on how it can be a successful displacement hull at all.
- Objective 7a has
- AC1 a graphic wave summation of the Bernoulli, bow, and stern waves of a
- Figure 3a increases the stern wave, not desired "per se” in the TH. This leads to
- profile is a straight line defining, together with the waterplane, a triangular profile
- This drag effect can be made small by designing a sharp
- the - N sin ⁇ term is propulsive, and the second term which includes the
- archetype TH does not exist to please the human eye. Rather, it corresponds to
- Figure 14a shows as a starting case the TH hull having an undersurface
- Figure 14c and 14d illustrate the piercing and exiting flows corresponding to a
- curled flow vectors f non-scale
- hydrodynamic effect, denominated anti-wave subduction, is the displaced volume
- undersurface is hydrodynamically directed, by continuity, to exit as a high energy content wake. Because it is high energy, there is low vehicle drag.
- TH is a submerged displacement body with a bow and stern.
- Figure 14f shows two submerged control volumes interconnected at a
- Figure 14f shows a shallow forward submerged control volume
- control volume must be V sin , perpendicular to the sides a, c, e and
- volume is of the form
- V sin e ac x ce V ⁇ cb x ce (1)
- control volume idealized for mathematical convenience extending between vertical
- V w gf x gh V ⁇ cb x ce (2)
- V w is an average wake velocity perpendicular
- V sin ac x ce V w gf x gh (3)
- the SUPERCRITICAL is a preferred hydrodynamic
- the surface flow field is flat and equipotential in the
- region 32 is flat, it includes
- region 32 continues to expand in a unique way, because of its highly
- the minimal humps are
- surface flow fields of TH are flat and equipotential in regions 41 , 42 and 43.
- Figure 14k shows the wetted planform 54 of planing hull at a positive planing
- Vee-transom a hollow wake 56 around which there is high turbulent wake which
- TH provides buoyancy lift and eliminates wave ⁇ making resistance in the
- sides 60 should be steeply inclined or vertical
- underbody slope discontinuity can also be used.
- TH and TH Another characteristic of TH and TH is the absence of suction forces from
- Figure 141 addresses the need to reduce wetted area for a given
- a TH having vertical side surfaces such as 61 in Figure 141, and TH's
- undersurface 62 may have sectional shapes as follows:
- submerged arc being circular or quasi-circular.
- portion is elliptical, or quasi-elliptical.
- a catenary section is gravitationally derived and is well suited.
- TH Another shape for section of TH is the hyperbolic spiral with the
- longitudinal axis should be considered with a vee-relation between axis meeting
- body of TH in Figure 141 can have a lower surface element 66
- Figure 16a shows a diamond waterplane archetype, which could be
- Figure 16b shows my DISCOVERY that the TH's triangular planform
- centroid CA of the waterplane shape as shown in Figure 15a, with a large "arm"
- TH as in Fig. 14h and the planing hull as in Fig. 14k can be
- TH triangular planform 34 is at a negative angle to water surface
Landscapes
- Physics & Mathematics (AREA)
- Fluid Mechanics (AREA)
- Chemical & Material Sciences (AREA)
- Engineering & Computer Science (AREA)
- Combustion & Propulsion (AREA)
- Mechanical Engineering (AREA)
- Ocean & Marine Engineering (AREA)
- Other Liquid Machine Or Engine Such As Wave Power Use (AREA)
- Hydraulic Turbines (AREA)
Abstract
A transonic hydrofield has been established which defines the shape of a displacement transonic hull (TH). The hydrofield is characterized is not having curvatures such as shoulders, midbody, or rear quarter, with a deep draft forward and a shallow draft aft. In consequence of the hydrofield shape, surface wave-making sources typical of conventional displacement hulls are eliminated. The transonic hull (34) corresponding to the hydrofield has substantially triangular waterplane in dynamic displacement conditions, maximum draft forward (a), and minimum draft aft (b, c), and planing lateral surfaces inclined at large angle, near vertical, to the waterplane.
Description
TRANSONIC HYDROFIELD & TRANSONIC HULL
BACKGROUND OF THE INVENTION
1. FIELD OF INVENTION
This invention pertains to hulls for.ships, yachts, submersibles, seaplane
hulls, and other water-borne vessels. More particularly, it pertains to new
hydrodynamic structures, the Transonic Hydrofield (TH), which intrinsically shapes
a new type of displacement form, the Transonic Hull (TH). Together, TH and TH
are characterized in substantially eliminating wave-making drag of displacement
hulls, without incurring the drag penalties of planing hulls.
2. DESCRIPTION OF RELATED ART
The potential of this writer's hull invention can be better appreciated by first
reviewing the advantage and limitations of conventional hulls.
Displacement Hulls
Displacement hulls sustain boat weight by buoyant lift. As designed in the
past and present, they have an upper speed limit called "hull speed," above which
hydrodynamic resistance (drag) grows at a high exponential rate. "Hull speed"
varies inversely with on the square rootof waterline length. Displacement hulls are
very efficient well below hull speeds, having a weight-to-drag efficiency ratio of the
order of 80. At very low speeds (creep), the efficiency ratio increases to much
higher values, because drag approaches zero but weight remains constant.
However, near or above hull speed, their weight-to-drag ratio becomes physically
and economically unacceptable. Therefore, greater speed of displacement hulls
is attainable principally by increasing hull length. However, the advantage of length
is not large. For example, the hull speed of a 50 foot hull is 9.5 knots, but for 300
foot hull speed, it is only 23 knots.
The hull speed limit in intrinsic of displacement hulls, because of their wave
generation properties as they translate in the water, i.e., "wave making," which
becomes critical at hull speed, as will be reviewed later on. This is a very serious
problem in the economics of maritime transportation. For that reason, considerable
research has been done in various ways to overcome it, unfortunately with only
minor improvements. For example, a bulbous bow may slightly decrease drag at
certain speed. Also, long slender hulls are less sensitive than beamy hulls, but
carry less cargo, and have other problems, as will be reviewed later on.
Accordingly, there remains an urgent need for improving the speed range
and high speed efficiency of displacement hulls. A practical solution of this problem
is needed, especially if it is able to eliminate wave-making drag without recourse
to hydrodynamic planing.
Planing Hull
There is a widely held view that the planing hulls, in which weight is
supported principally by a hydrodynamic lift force from momentum change (as
distinct from buoyant lift), can overcome the speed limits of displacement hulls, and
furthermore that they are efficient at high speed. Actually, while planing permits
high boat speed, it does so only for boats with an approximately flat underbody
having relatively light weight and equipped with large propulsive thrust. But the fact
remains that planing is a grossly inefficient hydrodynamic regime, since the best
ratio of boat weight to resistance is only about 8. This is less than half that of a
modern jet transport flying about 10 times faster, and only 1/5th that of a
displacement hull of "reasonable" length near, but below, hull speed. The
limitations of planing are inherent in their generation of lift by angle of attack,
described mathematically with equations which are analogous to those of
supersonic flight, as will be reviewed later on for the limiting case of inviscid
planing.
Semi-Planing Hulls
Unlike displacement hulls which have upwardly curved sterns and
curvatures at the bow, causing their CG to sink with forward speed (increasing their
apparent weight), and unlike planing hulls having flat undersurfaces and a CG
which tends to rise with forward speed, the semi-planing hull usually has a Vee
bottom and, for practical reasons, is heavier than a pure planing hull. Although the
semi-planing hulls can generate the appearance of a "flat" wake at high speeds,
their lift is generated by a combination of buoyancy and dynamic forces, which is
very inefficient. The borders of their "flat" wakes, as seen from an aerial view, join
together at some distance behind the stern, generating a trailing "hollow" on the
water's surface, which can be interpreted, from the viewpoint of a fish trained in
hydrodynamics, as an virtual displacement hull of larger length than the waterplane
of the semi-planing hull. The semi-planing hull is an inefficient hybrid: at slow
speeds, it has excessive drag compared to a displacement hull. It requires very
large power to reach semi-planing speed, at which regime it is less efficient than
a pure planing hull. On the other hand, a semi-planing hull provides smoother ride
for a greater payload in a rough sea, and is probably more seaworthy than a
planing hull. However, it has a rougher ride than a displacement hull, with less
favorable sea keeping characteristics, and is commercially not viable for most
maritime applications.
Semi-Displacement Hulls
As length-to-beam ratio is increased in slender hulls, wave-making drag
decreases. According to Saunders, slender displacement power boats were
common in the 1910s. Later on, the German Schnell boote (fast boat), having a
round-bottom hull, was successfully developed as an S-boat for WWII, performing
well at high speeds in the rough North seas. However, as the length-beam
slenderness ratio of semi-displacement boats is further increased, the lateral
stability and payload capacity is further decreased. In the extreme, an 8-man
rowing shell relies on oars for lateral stability. With a length-to-beam ratio of 30,
its wave-making resistance is only 5% of the total at 10 knots, but its weight-to-drag
ratio is only 20, approximately. An appropriate comparison in aircraft is the modem
sailplane with a wing span-to-chord ratio of 25. It can operate at weight-to-drag
ratio of 40, at 6 times the speed.
In the limit as beam of slender hull approaches zero, wave-drag tends
towards zero, but viscous drag subsists and payload capacity vanishes.
Accordingly, recent development of high speed semi-displacement boats have
proposed a mixed lift mode, using complex additions to the hull to generate
hydrodynamic lift at higher speeds, in order to decrease buoyant lift component,
and to compensate other shortcomings of the slender hull at high speeds, for
example, lateral instability and a tendency for nose high attitude. As is the case for
semi-planing hulls, their ratio of weight-to-drag is not very satisfactory, and in
consequence, payload is not large. Although they appear to have performance
advantages over semi-planing near hull speeds, and are less sensitive in pitch and
their complex shapes appear to have an inherent size limit. It may be added that
the proper name for this kind of vessel should be displacement-dynamic hull or
quasi-displacement hull, rather than semi-displacement.
Multi-Hulls
The wave-making and other drag problems of the various hull types
reviewed above are so serious that considerable recent efforts have ben applied
for the development of new multihulls. Although this field is outside the scope of
this review, a few remarks are in order. A pair of very narrow slender displacement
hulls of a Catamaran, widely spaced laterally for stability, have been successfully
developed and are being used at high speed for various applications, especially in
Asia. SWATHS are also multihulls which rely on totally submerged primary
displacement for performance and smooth riding. These developments and other
high speed hull developments (see, for example, Jane's high speed marine craft)
have so far been restricted to special applications, highlighting the need for ship
manufacturers to increase the speed and improve the riding qualities of
displacement monohulls.
BRIEF SUMMARY OF THE INVENTION
The TH is characterized in having engaged triangular waterplane shape in
dynamic conditions, a triangular profile, or modified triangular profile in side view
with maximum draft forward and minimum draft aft, planar lateral surfaces at large
inclination or vertical to the water, a double-wedge volume distribution with a
narrow entry angle in planview and a fine exit angle aft profile.
The hydrofield TH is characterized in absence of surface wave-making
sources such as shoulder, midbody, or quarter curvatures; narrow entry forward
which minimizes the water volume displaced per unit of time, and induces special
underbody flow, favoring adherence to the vehicle's sides of new types of bow ray
effects of very reduced size; absence of trough; favorable contracting streamline
on the sides; favorable pressure gradients on the hull's lower surface; broad stern
underflow prevent pitch up and eliminate stern wave, and favors the recovery of
underbody energy as well as that from following seas. In the case of sailboats,
special TH shapes permit long stern overhangs if required, or favored, by racing
rules. TH also provides special hydrodynamic coverage of roots of appendages
by the elimination of wave-making troughs.
BRIEF DESCRIPTION OF THE SEVERAL VIEWS OF THE DRAWINGS
The transonic geometric shape TH is shown in Figures 12a, 12b, 14a and
141. The transonic hydrofield TH is shown in Figures 13a, 13b, 13c, 15a, 15c, 14b,
14c, 14d, 14e, 14f, 14g, 14h, 14i, 141, 14m and 16b. They are applicable to any
type of waterborne vessel, or submersible vessel operating near or at surface, or
hulls of seaplane.
DETAILED DESCRIPTION OF THE INVENTION
CONCEPTUAL INQUIRY LEADING TO THIS INVENTION
One critical objective of this writer's invention is to formulate the
hydrodynamic design for a physical hull which is not susceptible to wave-making
drag limits of conventional hull, or of planing hulls. His conceptual inquiry
addresses the following fundamental questions:
• Is "hull speed" a physical "law" applicable to all possible displacement
hulls, or is it a statistical property of hulls as designed in the past and
in current "state of the art"?
• What are the ultimate hydrodynamic causes of wave making of
displacement hulls?
• What is the potential of very slender hulls to reduce wave-making drag
with acceptable volume for payload?
• What are the ultimate drag limits of planing hulls, for example in
inviscid planing?
• If one disregards all past hydrodynamic shapes and "laws", is there a
practical displacement hull shape that does not generate wave-making
phenomena of the type that currently limits speed and economic
performance of conventional displacement ships, without incurring into
the drag penalties of planing hulls?
ON THE PHYSICS OF WAVE-MAKING OF CONVENTIONAL
HULLS & CERTAIN DISCOVERIES
Wave phenomenon in liquids and wave-making around ship forms are
discussed at some length, and with adequate lucidity in Saunder's "Hydrodynamics
in Ship Design," written under the auspices of the U.S. Navy. Less extensive
reviews are also available in Gutelle's "The Design of Sailing Yachts," Marchaj's
"Aero-Hydrodynamics of Sailing" and "Sailing Theory and Practice"; inTodd's "The
Fundamentals of Ship Form. " The theoretical formulation of wave-making is also
reviewed in some depth by Kelvin, Stokes, Lamb, Durand, Innui, Wigley, Cummins,
Reid, Calderon, Lunde and others.
Important descriptions of principal physical phenomena which cause
wave-making by ships, taken from some of the references cited, are reviewed
below.
(a) The Bernoulli mathematical wave contour system around a simple ship
form is shown in Figure 1 , after Durand and Saunders.
(b) A description of wave-making sources in terms of Bernoulli, bow,
shoulder, quarter and stern wave components of a simple double wedge form are
shown in Figure 2, after Saunders and others, including resulting wave summation.
The analytic advantage of this shape is that it has specified clearly the location at
which wave components start. Also, it is similar to some theoretical airfoil shapes
for supersonic analysis.
(c) Bernoulli, bow, forward curvature, aft curvature and stern wave
components for a simple, slender, smooth lenticular airfoil shaped body are shown
in Figure 3, after Wigley's calculations in Todd, including resulting wave
summation.
(d) The relation of a hull's wave-making on a boat's angle of attack
(attitude) and therefore of part of its resistance are shown in Figure 4 as function
of speed-to-length ratio, after Marchaj and Gutelle, including typical wave-making
profiles at the hull's sides.
(e) The theoretical wave-making drag (resistance) of a hull is shown in
Figure 5 as function of Froude number, after Gutelle and others, simplified as
speed-to-length ratio, so its drag effects comparing theory to physical reality such
as sketched in Fig. 4 can be better understood.
(f) Even an extremely narrow ship form such as may be used in the light
displacement amas (pontoons) of a Catamaran or semi-displacement hulls,
continue to generate wave-making drag, the theoretical aspects of which are
described in Figure 6, after Saunders. As the beam of these slenders is
decreased, payload also decreases, so their practical utility is very restricted.
But hulls with a high payload, even with higher slenderness ratio of Figure 8a,
with a waterline-to-beam ratio of 8.2 after Saunders' ABC ship., continues to have
a high percentage of wave-making (residual) drag, reaching 50% at speed-to-
length ratio of only 1.0, as shown in Fig. 8b also after Sanders.
The latest, most advanced displacement monohull racing designs such as
used in International America's Cup Class (IACC) boats have a very smooth profile
and a length-to-beam ratio of about 4.5 to 1. Nevertheless, they continue to have
an adverse "hull speed" limit due to wave-making, the origin of which is related to
all of the wave-making components described in Figure 3.
If a hull (canoe, no appendages) of this general type is optimized for less
weight, such as is shown in Figure 7(a) with a length-to-beam ratio of 4.5, and is
then elongated in the computer, for example, to a length-to-beam ratio of 6.2 to 1
as in Figure 7(b), a slender boat results which even a reduced weight continues to
have wave-making components described in Figure 3, but of less magnitude.
Given sufficient thrust, such a hull could enter into a semi-planing regime, an
unpromising situation, as will be reviewed later on.
Alternative attempts to reduce wave-making of displacement hulls with "area
rule" principles borrowed from Dr. Whitcomb's aerodynamic transonic theory for
fuselage design have had uncertain results in hulls, for example, in P. de Saix's
searching efforts at Hoboken, described by Marchaj. It is conceivable that the
slender shape of Fig. 3 could be altered to a "waisted" canoe shape, if Dr.
Whitcomb's ideas applied to the water-air interface. One difficulty is that wave-
making drag of ships of Figure 3 is predominantly of gravitational origin, appearing
at the interface of two fluids - water and air -- having very different densities (ratio
of 840:1), but constant atmospheric pressure. This is a boundary condition totally
different from the three-dimensional flows of a transonic "area ruled" fuselages in
which thermodynamic and pressure variations predominates. The results of de
Saix's valuable efforts and similar work has not been successful.
Towards an explanation of the ultimate causes of wave-making resistance,
this writer has reviewed above how the shape of hulls generate wave-making
phenomena, and what effect this has on resistance of ships. As part of his studies,
however, some time ago this writer has made a first discovery that the smooth
curves used in ship design such as waterplanes in Fig.7a and 7b have inherently
misleading characteristics in real physical flow. These types of smooth hull shapes
have been used in displacement hulls from antiquity to present (Phoenician boats,
Viking ships, commercial vessels, and navy vessels all had very smooth hull
shapes). However, their waterplanes, for example, as in Figure 8a, while
geometrically real, are hydrodynamically divorced from reality. They exist only in
hydrostatic conditions without forward motion. When forward motion occurs, the
hydrostatic "fairlines" and "fairbodies" become significantly changed. When motion
approaches hull speed, new "waterplanes" parallel to the undulated water surface
must be considered: they become "unfair", resulting in a hydrodynamically
distorted inflected body shape, compared to the hydrostatic waterplane.
For example, the hydrostatic (flat water) lenticular waterplane of Figure 9a,
which is fair and smooth, must be reconsidered, according to this writer's
discovery, since at a speed-to-length ratio of about 1.3, it changes towards a
"dumbbell" type of real dynamic "waterplane" shape parallel to the undulating water
surface shown in Figure 9b, which is evidently inflected and "unfair", obviously
different in kind from Figure 9a.
Also, the cross-sectional area curve distribution for Figure 9a in hydrostatic
(flat water) conditions, generates a classic "bell shaped" smooth area distribution
of Figure 10a with a single maximum point. This is changed by forward motion with
a tendency for a double hump distribution of the type shown in Figure 10b, towards
two maximum points "ma" and a mid-body minimum point "mi": a "camel area curve
shape" corresponding to the "dumbbell waterplane shape" of Figure 9b, which may
be now understood as a dumb shape as well, often governing the hydrodynamics
of real ships, if not of swimming camels (because of their legs).
Due to the serious discrepancies between smooth hydrostatic waterplanes
and those parallel to the water surface in hydrodynamic conditions, this writer does
not grant priority nor, indeed, validity to current practice of ship design with
hydrostatically faired curves such as Figure 7 and 8, and as a second discovery
regards observed speed "laws" such as hull speed limit, only as statistically valid
test resulting from certain traditional design tendencies.
It is evident from the prior review, i.e., Figures 2 to 10b, that displacement
hulls in motion generate an undulated water surface which modified the "faired"
waterplanes which are used to calculate hydrostatic "displacement." The waves
travel attached to the hull as shown in Figure 4, and produce an energy-consuming
wave train symptomatic of wave-making resistance of Figures 5 and 8b. Toward
an explanation of the ultimate causes of wave-making drag of ships as design in
the past, this writer now reviews the physics of the flow of such displacement hulls
as they approach and attempt to overcome hull speed barrier.
Without forward motion, the boat weight is equal to buoyant lift "displaced" by
the hull position submerged relative to flat water in hydrostatic conditions; for
example, Figures 7a and 8b.
At moderate speeds, say speed-to-length ratio of 0.9 shown in Figure 4, the
wave-making of current displacement ship shapes produces an undulated surface
having three crests traveling attached to the hull, with the crests above the remote
waterplane. Evidently, the crest at midbody is particularly effective in generating
added buoyant lift from the broad midbody, compensating the deficiency of buoyant
lift caused by the troughs of the wave system at the shoulders and quarters of the
hull.
However, at a speed-to-length ratio of 1.25, as can be seen on Fig. 4, there
is only one trough at mid-body of the hull, well below the remote waterplane. This
sets certain physical buoyancy deficiencies reviewed below:
• The midbody is unable to generate its large intended buoyant lift
contribution, because the water surface adjacent to it is at a low level
relative to the remote waterplane.
• This deficiency of buoyant lift necessarily results in a significant
sinkage of the hull, and of the boat's center of gravity (for example,
0.75% of water-line length for a smooth displacement hull).
• This sinkage includes a downward pressure force (suction) generated
by the high speed flow under the curved undersurface of a
conventional displacement hull, increasing the "apparent" weight of the
boat.
• As the mid-hull sinks, the bow and stern must also sink to generate
added buoyant lift from the bow and stern volumes engaging the bow
and stern waves, so that buoyant lift equal to weight.
• The apparent density of the water at the bow and stern waves,
however, must be modified on account of their orbital velocities, in
order to determine the buoyant lift contribution at the bow and stern.
• The additional engagement of displacement volume of bow and wave
portions of the hull increases the height of bow and stern wave, which:
necessarily causes (by energy conservation) a corresponding
increase in the depth of the midbody trough leading to:
a further sinking of the hull, and yet additional engagement and
growth of bow and stern waves, and of midbody trough.
This vicious circle results in the "hull speed" limits of displacement hulls as
designed in the past and present, and produces the very rapid increase of
hydrodynamic wave-making drag (residual resistance) with speed, as was shown
in Figures 5 and 8b.
The resistance problem is further aggravated if a displacement hull, as
designed in past or present state of the art, attempts to exceed hull speed, for
example by use of a large amount of engine power, or gravitational energy when
a sailboat begins "surfing" in the inclined plane of an ocean wave: as the speed-to-
length ratio increases above 1.3 (speed in knots, divided by the square root of
length in feet), the stern of the boat sinks in the trough of the wave-making system,
changing the angle of attack of the hull to a nose-up attitude shown in Figure 4 at
speed-to-length ratio of 1.4, which can be recognized as a condition leading
towards "planing", provided the relation between thrust, weight and shape is
adequate. If planing were an efficient hydrodynamic regime, this might be
considered a desirable solution for "hull speed" limit. Whether this is the case or
not will be reviewed in the next section.
ON THE PHYSICS OF PLANING HULLS & EFFICIENCY LIMITS OF
IN VISCID PLANING
A design approach intended to eliminate wave-making of displacement boats
is to shape the hull and select the weight, speed and power to produce a so-called
"planing" regime, which incidentally is seldom defined with precision.
Although all designers recognize that at slow speeds "planing hulls" do not
plane and generate very large drag, some believe that once planing speed is
reached, the planing regime is efficient, and may even believe that drag decreases
with speed, once planing begins. These beliefs probably originate in the physical
sensation that the rate of growth of drag with speed decreases once planing begins
(second derivative of drag with speed negative being true), which is incorrectly
interpreted as a decrease of drag force with speed (first derivative of drag with
speed negative being incorrect). The fact is that tests of planing boats in their
planing regime show drag increases with speed.
Although the planing of a boat hull is a very complex phenomenon, the
fundamental physics of drag generation while planing can be understood in an
analogy to supersonic flight. In the latter, the lift-to-drag ratio (equal to weight-to-
drag ratio) is an efficiency index which for a thin flat airfoil is a function of angle of
attack, in a totally different manner from the lift-to-drag ratio index of an airfoil in
subsonic flight. The difference is due to the existence of a propulsive (suction)
pressure force at the leading edge of a subsonic airfoil, which is not present in the
supersonic case.
According to this writer's investigations, planing boats may not generate
conventional wave-making drag inherent of displacement hulls. Instead, they
necessarily generate, specifically in the planing mode, another type of high drag
force, which is a function of angle of attack of planing, is inherent in planing, and
is not present in a "pure" displacement hull. The function has the same
mathematical form as the pressure drag of thin wings in supersonic flight.
To elucidate the disastrous inevitability of this type of drag for planing hulls,
consider in Figure 11 the most efficient of planing hulls, namely, a flat plate having
a weight W, planing in inviscid (best) flow with mass density of water, at an angle
of attack alpha (« ). From Newton's second law, an angle of attack greater than
zero is essential to generate dynamic (i.e., non-buoyant) lift to oppose and balance
weight W. Moreover, by fundamental consideration of physics, the only force
acting on the plate is a resultant pressure force N due to change of momentum of
the inviscid flow under Newton's second law, such that:
• The resultant force N is normal to the flat plate, since there is no
viscosity, and
• In uniform motion, the weight W of the flat plate must be W = Ncos «
• In consequence, there is a drag due to angle of attack D = N sin «
• Accordingly, the efficiency ratio of weight-to-drag while planing in
inviscid flow is
W_ = N cos < = 1/tan ∞ (inviscid)
D« N sin «
The above description properly defines the planing condition. It follows that
D∞ and W/D« are therefore trigonometric functions inherent in planing under
Newton's second law, and are not dependent on local geometry in inviscid flow
conditions.
Thus, inevitably, there is a maximum possible planing efficiency measured
in weight-to-drag terms for inviscid planing. It is evaluated below, independently
of fluid density, for two typical real efficient planing angles, 3° and 6°:
W/D« @ 3° = 1/tan 3° = 19.08 (inviscid)
W/D- @ 6° = 1/tan 6° = 9.51 (inviscid)
It is evident that the efficiency of inviscid planing tends to increase as the
angle of attack decreases. Unfortunately, the presence of viscosity adds adverse
complexity to the physics of planing, impeding that trend. As can be seen in Figure
11 , the added friction term is a function of the wetted length Lw (assume unit
beam) of the flat plate, which necessarily increases if the angle of attack is
decreased.
Thus, two opposing functions of the planing angle combine such that best
efficiency for flat plates in real water occurs at angle of attack in between 3° and
6°. In perfectly flat "water," the optimum weight-to-drag efficiency ratio is under 10.
The addition of sea waves or wind driven "chop" lowers this ratio significantly.
When a real semi-planing boat is considered for ocean conditions, a Vee-
type shape is necessary for structural reasons and comfort. Their maximum ratio
of weight to drag is significantly less than 10, and close to 8 at speeds of the order
of 40 knots in flat water. As mentioned earlier, this is approximately half the
efficiency of a subsonic jet transport aircraft having a weight-to-drag ratio of 16
flying at 400 knots, i.e., ten times faster. The ratio 8 to 1 also compares adversely
to a displacement hull which can typically have a weight-to-drag ratio of 40 at half
the speed, i.e., at about 20 knots. The addition of ocean waves or chop to semi-
planing hull reduces their poor efficiency in a marked way.
GENESIS OF THE TRANSONIC HULL AND ANTI-PLANING
This writer has considered the problems reviewed in sections 3
(displacement) and 4 (planing) for some time, firstly in theoretical studies of
planing and displacement hulls while working as a Research Associate at Stanford
University, and later on during his work for various America's Cup yacht programs
(as Chief Aero-Hydrodynamist for the St. Francis Challenge, and then as Principal
Designer for Team Dennis Conner). However, this writer, as an aerospace
engineer, advanced his studies independently of yacht design rules, applying
aerospace principles and methodology to the hydrodynamics of displacement hulls,
based not on hull speed "laws" (which omit density and gravity terms), but on
original Froude number which has gravitational and density terms.
Thus, within the broad objective of a breakthrough in drag-reduction to
significantly improve the weight-to-drag efficiency index of ships, this writer
sometime ago formulated new hydrodynamic fields for displacement hulls which,
for theoretical reasons, he predicted to have no wave-making drag growth above
hull speed. His preliminary experiments have indeed established that with this new
type of field, the corresponding hull transcends, in a pure buoyant lift regime, the
"hull speed" limit of displacement hulls with excellent efficiency below hull speed
as well. This new development is considered as important in hydrodynamics as
breaking the "speed barrier" was in aerodynamics. Hence, in a loose but
interesting analogy (but also for disinformation) , the new type of hydrodynamic field
is denominated Transonic Hydrofield (TH). and the corresponding hull, or more
properly, a family of hulls, has been denominated "transonic hull" (TH). The TH
hulls also exhibit the capability of developing, in the displacement lift mode,
important and unique propulsive hull pressures from the hydrofield, in a regime
referred to as anti-planing, at speed ranges normally considered suitable for
planing hulls, as will be described later on.
THE TRANSONIC HYDROFIELD & TRANSONIC HULL
INVENTIVE SUBSTANCES
The reader is referred to the last statement of his conceptual inquiry in
Section 2: "If one disregards all past hydrodynamic shapes and laws, is there a
practical displacement hull shape that does not generate wave-making phenomena
of the type that currently limits speed and economic performance of conventional
displacement and planing hulls?" The answer is "yes": the Transonic Hydrofield
(TH) and its corresponding Transonic Hull (TH).
It is important to clarify their inventive substance: TH pertains to a three-
dimensional flow field capable of supporting a displacement translating hull without
water surface alterations of the physical shape and energy consumption level of
conventional displacement hulls. Thus TH is a unique three-dimensional flow field,
which intrinsically imposes a consequent and corresponding submerged body form,
the TH . This writer's invention process reverses the naval architect design process
which is to first shape the body for minimal drag and other requirements, and then
verify it by tests. The writer shapes the TH first, and then shapes the
corresponding TH. Hence, both TH and TH are matter for claims of their invention.
It is didactically preferable, however, to define first TH, and then review TH with the
aid of TH.
OBJECTS OF INVENTION & ARCHETYPE CHARACTERISTICS
One Object & Archetype Characteristics: eliminate from the engaged
dynamic waterplane all curves on the sides of the hull in the longitudinal direction.
Without offering proof yet that such a shape can provide a displacement hull of
useful purpose, it can be stated that as an immediate consequence, the real flow
of the hydrofield cannot generate "shoulder wave," nor wave resulting from
"forward curvature" as there is neither , nor a "midbody" wave, as there is no
midbody inflection; and there will be no "quarter" wave, or wave resulting from
"curvature aft" as there is no corresponding aft curvature. The physical solution to
this historical impossibility is the archetype's engaged waterplane having the simple
and elegant geometry sketched in Figure 12a: a triangle.
It is noted that this engaged waterplane archetype in Figure 12a is, as yet,
uncommitted as to underbody shape, stern hydrodynamics, propulsive power,
control system, or, indeed, on how it can be a successful displacement hull at all.
But it is noted that the triangular waterplane, although successful in not
having "shoulders", no "forward curvature" or no "midbody curvature", no "middle
length system", no "quarter" and, no "curvature aft", retains a bow and inevitably
some type of stern. Also, it needs displacement, while keeping in mind planing is
not an objective. Obviously, further design is needed.
Another objective and archetype characteristic: a key point is that the
summation of as yet undefined bow and stern waves has to be significantly
different from the summation of waves of conventional hull shapes which have
quarter and shoulder waves, and mid-body curvature effects, because it is the
adverse sum of the various wave components which determine the magnitude of
the critical trough near hull speed limit of a conventional displacement hull. This is
the trough which causes hull sinkage near hull speed already discussed, and buries
the forward and rear end of the boat into the typical bow and stern wave,
aggravating the trough, etc., and setting the hull speed physics. Objective 7a has
clearly avoided the adverse summation of classical wave components, by
eliminating some components. Is that sufficient?
This writer asks the reader to consider the effect of suppressing in the
hydrodynamic situation of Figure 3, by supreme edict of mathematical analysis if
the reader allows, the waves associated with forward curvature (shoulder),
midbody parallel sides, and aft (quarter) curvature of the body. This is at this time
a virtual but illustrative graphic exercise, which, using Figure 3 as a baseline,
shows the interesting result of Figure 3a. The estimated result is labeled "Est.
AC1", a graphic wave summation of the Bernoulli, bow, and stern waves of a
theoretical body without shoulders, midbody, or quarter. The result, though not
necessarily a physical reality for TH (as will be reviewed later on), does alter the
situation of Figure 3. Is it sufficient?
The cross-hatched areas in Figure 3a show a substantial reduction of
mid-body trough, which should decrease (but not eliminate) wave-making drag rise
with increasing Froude number for ships in general. However, the exercise of
Figure 3a increases the stern wave, not desired "per se" in the TH. This leads to
a second virtual exercise: the elimination, by supreme edict, of the stern wave, as
estimated by the wave summation of Figure 13b. The result, unfortunately, does
not eliminate the trough. It only reduces it. Thus, 7(b) is qualified as a desirable
but insufficient characteristic. More thinking is needed.
Another objective and TH characteristic: Unfortunately, nature does not
bend to the wishes of inventors. This writer has to contend to the underwater
three-dimensional shape corresponding to the triangular archetype hydrofield
waterplane of Figure 12a, for it to achieve displacement and eliminate stern wave.
If the underbody near the stern were curved as in conventional yachts, this writer's
hull would exhibit a tendency towards "wave-making," and "sinkage." To prevent
these disorders and for fundamental reasons of mathematical hydrofield
formulation, another TH archetype characteristic is formulated: the underbody in
profile is a straight line defining, together with the waterplane, a triangular profile
with its base at the bow, as shown in Figure 12b. Note that the base of the
waterplane is at the stern, forming a unique orthogonal imbrication of
complementary triangular shapes and corresponding volumes, the geometric name
of which is not recognizable and is denominated for convenience "double wedge."
It is a unique property of the archetype underbody shape of Figure 12a and
12b that the pressure integral of its underbody surface, shown in Figure 13a,
responds in hydrofield conditions to a negative angle β, relative to the horizon, with
no undersurface area at positive β. Thus, the resultant pressure force N
necessarily provides a propulsive thrust equal to -N sin β. This is referred
hereafter as the anti-planing propulsive condition of a displacement TH. This is in
direct opposition to the pressure drag term N sin « discussed already for planing
in Figure 11.
In Figure 13b there is shown a drag force from the integral of the lateral
forces on the hull's side area from element da, integrated symbolically in Figure
13b as drag Dy , where y is half the angle at the apex of the hydrofield's
waterplane triangle. This drag effect can be made small by designing a sharp
entry, and by other means, to be reviewed later on.
What is of extreme interest is the form of the total pressure drag D equation
implicit in Figures 13a and 13b, as follows:
D = - N sin β + K (VG) sinγ J Ps (x, y, z) da
Where K is function of velocity V and gravitational force G
The - N sin β term is propulsive, and the second term which includes the
side pressure term Ps in the integral can be made small by choosing a small value
of angle y , and by other means to be described later on, pertaining to a trailing
flap at the stern of TH , and/or unique combination of propulsive flows with the hull's
hydrodynamics.
Another objective and TH characteristic: A critically important requirement
of the triangular archetype of Figure 12b is the very sharp entry angle of the body
in the hydrofield. This is achieved by setting maximum beam substantially at the
hull's stern, instead of somewhat ahead of the stern, nearer the middle, as is usual.
This results in reducing the entry angle by about half. As mentioned, this sharp
entry reduces the pressure drag on the sides, as described in Figure 13b. Equally
important for the hydrofield, it reduces the volume of near-surface water displaced
in the bow region per unit time at a given speed, thus reducing the tendency for
bow wave (see Figures 3, 3a, 3b), although the classical concept of bow wave
does not apply in the traditional sense to TH, as will be explained later on.
Additionally, the placement of maximum beam at stern has its own large
beam buoyant lift contribution compensating its small local draft and, in addition,
a hydrofield lift and propulsion capability, each of which prevents the formation of
a stern wave and opposes a pitch up, otherwise causing drag due to angle of
attack, i.e., retains the TH in TH out of the planing mode. Although wide stern aids
in the extraction of energy from a following sea overtaking the hull, more important,
it is an energy recovery feature of JJH's three-dimensional hydrofield, acting on
TH's stern design.
Another objective and TH characteristic: Although elegant, the geometric
archetype TH does not exist to please the human eye. Rather, it corresponds to
the real inventive substance of TH, which is a three-dimensional hydrofield
archetype that does not generate wave-making drag at its design speed envelope.
TH's inventive substance is better explained by steps with the aid of isometric
figures.
Figure 14a shows as a starting case the TH hull having an undersurface
triangular plane a, b, c, with the corresponding nearby waterplanes indicated.
Figure 14b adds on the undersurface a complementary imaginary pyramidal
flow control volume having corners a, b, c, d and e. This control volume permits
to raise the very interesting question: how can the flow rate leaving the control
volume perpendicular to base rectangle b, c, d, e exist?
Mathematically, such flow rate must pierce lateral triangles a, c, e and a, b,
d. Figure 14c and 14d illustrate the piercing and exiting flows corresponding to a
continuity equations to the previously defined control volume. Referring first to the
flow indicated on the left flower side of undersurface of Figure 14c, it is seen that
continuity adjacent the left corner of the hull requires an inward local flow adjacent
to the hull's submerged corner, exemplified by curled flow vectors f (non-scale),
rather than the usual outward flow (the inward curl has an alternate solution
described later). The unique benefit is that such local flow tends to decrease the
height of the water on the sides of the translating double-wedge displaced volume
of TH, as is evident on the left side of Figure 14e (non-scale). By this special
hydrodynamic effect, denominated anti-wave subduction, is the displaced volume
per unit time caused by motion of TH is induced to flow, to a large extent, below
TH's bottom, rather than appearing as surface waves. The inventive
complementary consequence is exemplified at a rear lower level of the control
volume: the remote flow pierces the lower sides of the control volume undisturbed,
as exemplified on the right (upper) side of Figure 14c, by straight vector g, also
shown on the right of Figure 14e.
Under anti-wave subduction, the inward lateral flow below TH's
undersurface is hydrodynamically directed, by continuity, to exit as a high energy
content wake. Because it is high energy, there is low vehicle drag. The flow
combination thus described under a mathematical study by means of the
submerged control volume of Figure 14b according to the assumptions of Figures
14c, d, and e, promises that there should be no stern wave to TH of any kind, and
moreover that at certain speed range the drag growth with speed growth should be
a function of the square of speed gain only, i.e., no wave-making drag growth even
though TH is a submerged displacement body with a bow and stern. The
mathematical situation, however, requires the inclusion of the wake behind the
stern, as reviewed below.
Figure 14f shows two submerged control volumes interconnected at a
vertical rectangle below the transom of a TH hull, which is equipped with a small
trailing flap so that the flow exits from the front volume to rear volume through the
interface rectangle, vertical to the remote waterplane.
Specifically, Figure 14f shows a shallow forward submerged control volume
pyramid a, b, c, d, e below the undersurface of TH having lateral sides a, c, e and
a, b, d, with a rectangular vertical base below the stern b, c, d, e. With the flow
streams parallel to approximately horizontal planes a, b c and a, d, e, the following
equation of continuity can be written with the definitions of Figure 14g.
• Let Y equal half the hydrofield's waterplane apex angle.
• Let V equal boat speed.
• The average flow vector component entering each side of the forward
control volume must be V sin , perpendicular to the sides a, c, e and
a, b, d.
• The average flow vector component exiting perpendicular to the
rectangular base b, c, d, e below transom is Vτ.
Consequently, the continuity equation for the submerged forward control
volume is of the form
V sin e ac x ce = Vτ cb x ce (1)
The above equation implies certain momentum distribution of the submerged
hydrofield, which enters at the vertical face of TH's rearward submerged wake
control volume, idealized for mathematical convenience extending between vertical
rectangle b, c, d, e below the transom as a plane of momentum entry, and exits at
vertical rectangle f, g, h, i, at considerable length downstream.
The writer now assigns continuity condition within the rear control volume,
noting that the sides b, d, f, h and c, e. g, i are coplanarwith the sides of TH's hull.
This imposes a continuously expanding flat surfaced submerged TH wake 32
shown in Figure 14h, rather than the usually contracting wake seen to the rear of
semi-planing boats, such as 57 in Figure 14k, and without the "traverse" surface
wave, such as 49 and 50, at the stern of displacement boats such as in Figure 14j.
With the previous transonic hydrofield requirement, as in Figure 14f and 14g,
we have imposed that the flow thus enveloped is parallel to top and bottom planes
of the rear control volume. Accordingly, we can write a continuity equation for the
rearward control volume as follows:
Vw gf x gh = Vτ cb x ce (2)
where Vw is an average wake velocity perpendicular
to the downstream vertical rectangle.
From equations (1) and (2) we can establish, for the idealized three-
dimensional hydrofield, a relation between remote wake velocity and boat velocity
as follows:
V sin ac x ce = Vw gf x gh (3)
Speed Envelope of TH and TH. It is possible to solve several local flow
velocities from equations (1), (2) and (3). The estimated corresponding hydrofield
for TH have two types of quasi-steady solutions, subcritical and supercritical.
Supercritical Regime. The SUPERCRITICAL is a preferred hydrodynamic
design condition for the submerged transonic hydrofield. It permits the general
outlook of Figures 14f and 14g and has the surface appearance of Figure 14h,
noting that the upper surface of the rearward control volume is at the air-water
constant pressure interface. TH has exception performance.
Specifically in Figure 14h, the surface flow field is flat and equipotential in the
gravitational sense in zones 30, 31 and 32. While region 32 is flat, it includes
molecular agitation, because of friction below undersurface 34 of TH.
Nevertheless, region 32 continues to expand in a unique way, because of its highly
directional steady momentum, indicative of successful anti-wave subduction for
optimum performance of TH.
The volume displaced by translating TH emerges principally in region 32,
with the minimal surface alteration appearing as left and right three-dimensional
rays 35 and 36, having the minimal elevation shown by hump 37 at downstream
wake cut 38, and minimal hump 39 at midbody station 40. The minimal humps are
better understood from the teachings of Figure 14e regarding anti-wave
subduction. This type of surface flow field applies to the TH's design speed range
(above classical hull speed) in which the unique properties of the 3-D TH flow field
has the unique feature for displacement hull: increase of drag with speed to only
the second power of velocity, i.e., no added wave-making with added speed: This
is the super-critical TH regime for TH, and contrasts with drag growth of
conventional displacements hulls which can vary with 3rd power and up to 7th
power of speed.
Certain supercritical hydrofield solutions of continuity is exemplified in Figure
14m. The corresponding TH archetype geometric sections are shown as seen
from the rear. The inward flow induced by the double-wedge three-dimensional
geometry is shown as dashed streamlines S1 , S2, S3, each with a constant-
elevation inward flow before they initiate their sloped rearward accent along the
underbody of TH, entering the left half of rearward control volume at left half of
rectangle c, b, d, e.
Subcritical Regime. In the SUBCRITICAL speed regime, shown in Figure
14i, surface flow fields of TH are flat and equipotential in regions 41 , 42 and 43.
But undersurface viscosity forces, relative to momentum content of flow at
subcritical speeds limits the shape and area of the wake to a gothic arch type of
planform such as 43. Rays 44 and 45 have larger humps. Downstream of flat
wake 43, there is some eddy and hump formations 46 and 47 and a central hump
48. In this sub-critical regime, there may be in some cases drag growth with speed
slightly higher than second power of speed, because of the eddies and elevations,
even though there are no transverse stern wave nor traverse bow wave of the type
of conventional displacement hulls, such as crests 49 and 50 of Figure 14j, nor is
the re a divergent wave 51 , all of which correspond to smooth displacement hull
51 at its hull speed.
Comparison of Wakes of Various Hulls and TH. When a displacement hull
such as 51 is lightly loaded and moving at speeds considerably higher than hull
speed, its wake forms long longitudinal humps and troughs such as 52 and 53
respectively, causing a much higher energy consumption than TH in its supercritical
TH condition of Figure 14h.
As the design speed envelope of TH is considerably above that of
conventional displacement hulls, it is interesting to compare the surface conditions
of supercritical TH flow and planing flow.
Figure 14k shows the wetted planform 54 of planing hull at a positive planing
angle reviewed in Figure 11 throwing lateral water sheets 55 and generating at the
Vee-transom a hollow wake 56 around which there is high turbulent wake which
closes at 58.
A comparison of the non-planing TH subcritical surface flow of Figure 14i
and supercritical flow in Figure 14h indicate how TH has much less drag than
planing or displacement.
The comparison of Figures 14h to 14k have been make at equal length,
maximum beam, and assumed weight. The efficiency index weight-to-drag of TH
is vastly superior at speeds of supercritical flow, and significantly superior at
subcritical speeds.
Additional Characteristics of TH. The double-wedge submerged shape of
TH provides buoyancy lift and eliminates wave÷making resistance in the
displacement mode. In Figure 14e, sides 60 should be steeply inclined or vertical
in its submerged area and somewhat above the hydrofield's waterplane, to interact
with water displaced under motion with TH's fine bow, which adds water volume
in the proximity of the hull's midbody and aft body, with rays 35 in Figure 14h,
rather than forming a midbody or aft body trough of conventional hulls, even if
mildly curved at high speeds, as shown in Fig. 15d.
It should be understood, in connection to the discussion of Figure 15c, that
the pressure integral over its wetted underbody prevents, with maximum beam at
stern, the pitch up of a displacement hull. This, together with the side effects of the
double-wedge, uniquely captures any energy otherwise leading to a stern wave,
and what is critical, by choosing a large angle of exit β, for example 10° or more,
there is achieved a gain of propulsive pressure force and of displacement
adequacy. For such cases, a trailing flap on the stern at the waterplane's base is
provided, because with its maximum beam, it provides maximum recovery of
underbody flow energy towards positive heave and nose-down pitch, which can aid
propulsive pressures in special combination, as shown by flap 70 in Figures 15a
and 15c.
It should be further understood, in respect to the triangular hydrofield
waterplane of Figure 14b with near vertical sides that by generating a linearly
increasing water beam from bow to stern, there results contracting and stable
streamlines without eddies in the entire length of the sides of the hull, as shown in
Figure 15a, without the usual inward curvature of streamlines near a conventional
stern as in Figure 15b, which has the corresponding eddy sources and classical
wave-making profiles shown in Figure 15d. The steady flow of 15a favors low
viscous drag, including smooth wake conditions aft of the stern.
It is further noted from Figure 15c that when TH is in motion, the hydrostatic
pressure on the TH's undersurface decreases as the flow moves rearward along
the entire underbody at a negative angle β, which favors by means of a continuous
favorable gravitational pressure gradient, low friction drag, more effective when β
is of the order of 10° or more. However, underbody slope discontinuity can also
be accepted in certain cases; for example, for meeting displacement for cargo
requirements, as shown in Figure 13c. Although the underbody area providing
propulsive pressures is smaller, the exit angle e is much larger than β in Figure
15c, and the effective propulsive thrust may be even larger in Figure 13c, provided
the flow remains attached.
There is another special characteristic of TH which is its ability to change its
altitude nose up with increasing speed up to an angle equal to β, without incurring
in drag due to angle of attack, obtaining the benefit of reduced wetted area on the
hull's sides, of significance at the higher speed range. Change of trim can be
induced by the hydrofield itself, by the flap, by change of fuel position, or by
auxiliary surfaces.
Another characteristic of TH and TH is the absence of suction forces from
usual longitudinal curvature in profile on the hull in the displacement mode, which
prevents added apparent weight usual in displacement hulls. Therefore, TH has
less drag.
The TH and Viscous Effects. The previous figures have described the
archetypes TH (geometry of hull) and TJH (three-dimensional hydrofield principally
in respect to elimination of wave-making drag. However, vessels have also friction
drag, lateral stability requirements, fabrication process, etc. These impose certain
requirements to TH which are acceptable for TH.
Figure 141 addresses the need to reduce wetted area for a given
displacement, and avoiding excessive local speeds with sharp curvatures.
Specifically, a TH having vertical side surfaces such as 61 in Figure 141, and TH's
undersurface 62, may have sectional shapes as follows:
• At the forward station, a U-shape with the lower portion deeply
submerged arc being circular or quasi-circular.
• At midbody station 68, the cross-section of the deeper submerged
portion is elliptical, or quasi-elliptical.
• At stern station 69, the cross-section of the submerged portion is
flattened elliptic, or flattened quasi-elliptical.
The circle to ellipse family mentioned above is convenient, because of its
relative mathematical simplicity in smooth transition, both being from the conic
family. However, TH operates not in conic space, but in a gravitational field.
According to this writer's study, there are better curves identified below:
• A catenary section is gravitationally derived and is well suited.
However, the forward vertical axis needs rotation reaching the stern.
Another shape for section of TH is the hyperbolic spiral with the
corresponding Cartesian axis being rotated gradually, reaching at the stern with
rotation and inversion.
Finally, if a conic family is used, because of its simplicity, a double
longitudinal axis should be considered with a vee-relation between axis meeting
near the forward end of TH.
In profile, body of TH in Figure 141 can have a lower surface element 66
which when parallel to water surface, towards maximum volume for a given wetted
area, has an upwardly inclined section 64 between stations 68 and 69 (see also
Figure 13c).
TH and Form Stability. It is important in respect to the waterplane of
Figure 12 — hydrostatic and hydrofield — and TH shape of Figure 141, to understand
its unique features relative to lateral stability. This is done by comparing certain
features of TH to two archetypes of conventional ships, all of equal maximum
beam and overall length, as follows.
Figure 16a shows a diamond waterplane archetype, which could be
considered to have great lateral or form stability. This is attained an obviously
large wave-making, from its midbody "curvature" and from a large entry angle of
the bow.
Figure 16b shows my DISCOVERY that the TH's triangular planform,
although it has the intended half entry angle compared to Figure 16a and no lateral
wave-making sources, has egual lateral stability. This is shown by comparing the
equal lateral planform area segments AL of Figures 16a and 16b, which have
equal moment arms to the respective centerline.
TH. Longitudinal Trim and Pitch Stability. A unique feature of TH in JH
is the inherent ability to trim by the bow. In this procedure, bow draft increases with
increasing weight (cargo) while keeping near fixed draft at the stern, as shown in
Figure 15c with waterplane WP1 and WP2, a special feature for low drag.
Another unique feature of TH is the inherent decoupling of the longitudinal
position of its buoyancy center (LCB) from the longitudinal position of its waterplane
centroid, thus decoupling the center of gravity of the boat in respect to the area
centroid CA of the waterplane shape, as shown in Figure 15a, with a large "arm"
1 subunit or a2, not feasible for conventional hulls. This decoupling damps out
pitch motion in a rough sea, improving comfort and performance, also of interest
for the aerodynamics of sails and hydrodynamics of hulls of sailboats in a choppy
or heavy sea.
Moreover, it is a unique characteristic of TH that for a given CG, attitude
remains constant for a large range of speed, holding the same negative angle β.
Remarks on TH and TH. Although the transonic hydrofield does not
respond to conventional parameters of conventional hulls, such as speed-to-length
ratio (speed in knots divided by square root of boat length in feet) according to
which hull speed is 1.35, nor does it respond to displacement-to-length ratio
(displacement in "tons" divided by to third power of quantity hull length in feet
divided by 100), nor maximum draft-to-length ratio, nor to ratio of boat length to
maximum beam; the corresponding naval calculations can be made for TH
nonetheless, to (a) permit comparisons of test results of TH with conventional hulls,
and to (b) delineate properties and weights for TH vessels in which the unique
benefits of TH excel, by way of example and not of limitation. For example:
• By placing maximum beam at stern, L/Bmax as low as 4.3 can be
used, keeping a total entry angle (side to side) of 6.5°, which means
approximately half that of a refined conventional displacement boat
with fine entry.
• By using a large ratio of draft to length, of the order of 0.1 or more,
large payloads can be carried in displacement modes retaining
supercritical and subcritical flows.
• By using low length-to-beam ratio, lateral from stability is retained with
maximum beam adjacent the stern for fine bow entry.
• By using a large deep draft-to-length ratio, exceptionally low drag can
be attained with ratio of displacement-to-length displacement in tons
well above 100 and exceeding 200.
It is pertinent also to clarify certain qualitative differences between TH and
prior art. For example, TH as in Fig. 14h and the planing hull as in Fig. 14k can be
regarded to have similar waterplanes. The fundamental and qualitative difference
is that for TH triangular planform 34 is at a negative angle to water surface
providing a propulsive force as in Fig. 13a or 13c in the displacement buoyant lift
regime, whereas quasi-triangular water planform 54 in Fig. 14k generates a drag
force as in Fig. 11 in the planing dynamic lift regime. These are totally different and
opposite regimes.
The specifications, claim substance, and drawings pertaining to TH and TH
described in the preceding pages of this document pertain to hydrodynamics, to
corresponding archetype bodies, and to hydrodynamic methods. The inventor
recommends that the dynamics of TH and TH are investigated by a licensed
manufacturer of a TH in full size. This results must establish the necessary
improvement for achieving seaworthiness and safety of TH. Small scale testing
of models is not sufficient for safe, full size commercial application of TH.
Moreover, the structural integrity, mechanical reliability, adequacy of boat control
seaworthiness, and any other specific design and safety factor that affects any
specific boat that could be built in relation to the patent are obvious not covered by
this document, and is the sole responsibility of the engineers, naval architects,
marine organizations, etc. which undertake TH fabrication, and its users.
Accordingly, licenses and/or purchasers of the patent are obligated to assume all
responsibility and liability of the TH vehicles and to advise the clients of this fact.
Within the general archetype and preferred shapes of TH and TH described
above, this writer's design can be varied with unique refinement and special
features needed for concrete practical applications, without departing from the spirit
of the invention.
Claims
1. A transonic hull portion having a bow, a stern, and a length, a substantially triangular waterplane, side surfaces extending from said bow to outboard edges of said stern with a deep draft forward and a shallow draft to the rear, said hull portion being immersed in a water body in a displacement mode having a remote reference water level and a supercritical transonic three-dimensional hydrofield in direct contact with and translating and supporting by means of buoyant forces, the submerged surfaces of said transonic hull portion translating at high speed in said water body in said displacement mode, said hydrofield being further characterized in having (a) a substantially triangular planform at water level with substantially rectilinear longitudinal sides with the base of said triangular planform corresponding substantially to the stern of said hull portion and the pointed end forward; (b) lateral hydrodynamic rays of small elevation and narrow width in cross-section extending along said sides of said planform and continuing downstream of said base of said hydrofield as generally straight rearward extensions of said rays; (d) with a generally flat wake extending downstream from said base of said planform, said generally flat wake being contained between the portion of said rays located to the rear of said base, and having a water level therebetween substantially equal to that of said remote waterplane outboard of said rays.
2. A transonic hull portion according to Claim 1 in which the traverse distance of said flat wake downstream of said base continues to increase linearly with increasing distance aft of said stern.
3. A transonic hull portion according to Claim 1 in which the hydrofield surfaces corresponding to and supporting said submerged transonic hull portion forward of said stern between said hydrodynamic rays is depressed below said reference waterplane, with said depression having a deep draft forward with a narrow entry angle, and a shallow draft adjacent said stern, thereby establishing the depressed region of said hydrofield which corresponds to said submerged transonic hull portion.
4. A transonic hull portion according to Claim 3 in which the cross-sectional shape of said depressed hydrofield corresponding to the immersed surface portions of said hull portion has substantially vertical principal water side surface portions adjacent and below said hydrodynamic rays along the principal length and depth of said depressed region.
5. A transonic hull portion according to Claim 4 in which the lower portion of said principal water side surface portions are connected in the transverse direction by a bottom water surface, having a substantial portion of its area being generally flat and horizontal in a traverse direction.
6. A transonic hull portion according to Claim 1 in which the ratio of said high speed expressed in knots to the square root of said length expressed in feet is greater than 1.25.
7. A transonic hull portion according to Claim 5 in which the water flow direction in said hydrofield flows inwards and downwards below the lower limits of said principal water side surface portions, with said flow progressing rearwardly along said bottom water surface, said bottom water surface being upwardly inclined in side view at a shallow angle, said flow emerging into said flat wake aft of said base and between said hydrodynamic ray extensions at generally the same elevation as that of the remote water level.
8. A transonic hull portion according to Claim 1 in which said sides of said water planform adjacent said forward pointed end form an angle no greater than approximately 14°.
9. A transonic hull capable of operating at high speed in a displacement mode characterized in having a submerged displacement volume which projected in planform has a substantially triangular shape at the waterplane having an apex at its bow and a base adjacent its stem, with the left and right sides of said waterplane being substantially rectilinear over substantially the principal length of said waterplane, with said submerged volume having right and left sides which are approximately triangular in side view with a base forward establishing a deep draft adjacent said bow and an apex at a shallow draft adjacent said stern.
10. The hull of Claim 9 in which the underbody surface of said transonic hull in its principal length is inclined at a small negative angle to the water surface, said underbody surface having a substantial portion of its area generally flat and horizontal in the traverse direction and a rear edge exiting substantially at dynamic waterplane level, with the pressure of buoyant forces on said generally flat portions of said underbody surface being generated by smooth hydrodynamic flow along said underbody surface, said pressure forces having a propulsive force component pushing said transonic hull forward, whereby the ratio of vehicle weight to pressure drag is improved.
11. The structure of Claim 10 adapted to operate in waters of limited depth with said underbody surface having a modified forward portion of limited draft and area approximately parallel to the water surface, and a rearward portion of much greater area than the area of said forward portion inclined to the water surface at a negative angle larger than said small negative angle, with the propulsive pressure force component in said rearward portion being increased in proportion to the increase of negative angle, and with the draft of the forward portion being less than said limited water depth, as required for practical use of said hull in said water.
12. A transonic hull having anti-wave-making subduction and anti-planing propulsion when in displacement regime at high speed, characterized in having generally flat sides along the principal length of said hull with substantially no twist below its waterplane and inclined at a very steep angle to said waterplane throughout substantially a principal length of the hull, with the planform of the corresponding waterplane being substantially triangular on a principal length of said hull, with substantially rectilinear sides over said principal length, with the base of said triangle adjacent the stern, and an undersurface having its forward end with a deep draft substantially deeper than the draft at said stern.
13. The structure of Claim 12, in which the submerged portion of said hull facing downwards has cross-sectional shapes including approximately elliptic portions, with major axes vertical near the front, and major axes horizontal adjacent the stern, said cross-section being further in that the side surface portions adjacent the waterplane from midbody towards the rear are substantially vertical, and the corresponding downwardly facing surfaces include approximately flattened elliptic portions in traverse cross-section adjacent said stern.
14. Claim 9 in which the speed-to-length ratio of said hull is greater than 1.25 in displacement mode, with speed expressed in knots and the denominator of the ratio expressed in square root of length in feet.
15. Claim 9 in which said hull has a movable adjustable stern flap, which in operational mode makes a small positive angle relative to the hull's negatively inclined undersurface adjacent the hull's stern, establishing propulsive momentum redirection from undersurface flow.
16. A transonic hull having a submerged portion with a sharp bow, a wide stern, and a length therebetween at waterplane level, said hull portion being further characterized in having:
(a) A substantially triangular planform at waterplane with two substantially rectilinear sides extending over a principal length of said waterplane, said sides connecting said sharp bow with the outboard ends of said stern at waterplane adjacent the horizontal base of said triangular planform;
(b) A lower surface having substantially rectilinear longitudinal elements over the principal length of said lower surface, said lower surface having a deep draft adjacent said bow and a shallow draft adjacent said stern; and
(c) Side surfaces on said submerged portion with substantially no twist over their principal length.
17. The structure of Claim 16 in which said submerged portion has traverse cross-sections with principal elements on said side surfaces and said lower surface having approximately rectangular relationship, said relationship extending between a region adjacent said stem and a region adjacent said bow.
18. The structure of Claim 17 in which the envelope of said cross-sections having approximately rectangular relations define, in end view, a substantially triangular envelope shape, with its apex downwards and its base adjacent the waterplane.
19. The hull of Claim 9, inherently characterized in having a longitudinal position of the center of area of the waterplane located at approximately one-third of said principal length of said waterplane measured from said base, substantially to the rear of the hull's center of gravity including the hull's payload, whereby damping in pitch in adverse sea conditions is greatly enhanced.
20. The hull of Claim 9 in which the ratio of said deep draft to the longitudinal distance between said bow and said stern is no greater than approximately 0.08, whereby operations at high speed of said hull in open ocean is made compatible with operation of said hull within waters of limited depth.
21. The hull of Claim 9 further characterized in that each cross-section in the transverse direction upstream of midbody is characterized in having a beam distribution in a vertical direction below the waterplane that is approximately constant over the principal depth of the cross-section above its lower corners, with the cross- area distribution and volume distribution upstream of said midbody of said hull being deeply submerged, whereby the formation of bow waves is minimized.
22. The hull of Claim 12 in which the beam distribution at each principal beam station above said waterplane level remains generally equal to that below said waterplane.
23. The hull of Claim 12 in which uniform beam distribution characteristics at each section extends along the principal length of said hull.
24. The hull of Claim 19, further characterized in said center of area and the hull's center of gravity can be decreased by moving said payload rearwards, thereby engaging a small incremental draft at said stern without shifting said centers of area.
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/US2000/009799 WO2001079054A1 (en) | 1996-03-13 | 2000-04-12 | Transonic hydrofield & transonic hull |
Publications (2)
Publication Number | Publication Date |
---|---|
EP1326773A1 true EP1326773A1 (en) | 2003-07-16 |
EP1326773A4 EP1326773A4 (en) | 2005-11-02 |
Family
ID=21741269
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
EP00922123A Withdrawn EP1326773A4 (en) | 2000-04-12 | 2000-04-12 | Transonic hydrofield & transonic hull |
Country Status (4)
Country | Link |
---|---|
EP (1) | EP1326773A4 (en) |
AU (2) | AU4235800A (en) |
CA (1) | CA2412908C (en) |
IL (2) | IL154454A0 (en) |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
DE112414C (en) * | ||||
US23626A (en) * | 1859-04-12 | tucker | ||
US514835A (en) * | 1894-02-13 | Francis e | ||
DE19614500A1 (en) * | 1996-04-12 | 1997-10-16 | Weber Eckhart | Body of boat |
US5711239A (en) * | 1994-04-21 | 1998-01-27 | Petroleum Geo-Services As | Propeller configuration for sinusoidal waterline ships |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US4079688A (en) * | 1976-08-12 | 1978-03-21 | Diry George L | Displacement hull |
DE3022966C2 (en) * | 1980-06-19 | 1986-07-17 | Paul Dr. 1000 Berlin Mader | Hull, in particular for a sailing dinghy |
-
2000
- 2000-04-12 AU AU4235800A patent/AU4235800A/en active Pending
- 2000-04-12 AU AU2000242358A patent/AU2000242358B2/en not_active Ceased
- 2000-04-12 IL IL15445400A patent/IL154454A0/en unknown
- 2000-04-12 CA CA2412908A patent/CA2412908C/en not_active Expired - Lifetime
- 2000-04-12 EP EP00922123A patent/EP1326773A4/en not_active Withdrawn
-
2003
- 2003-02-13 IL IL154454A patent/IL154454A/en not_active IP Right Cessation
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
DE112414C (en) * | ||||
US23626A (en) * | 1859-04-12 | tucker | ||
US514835A (en) * | 1894-02-13 | Francis e | ||
US5711239A (en) * | 1994-04-21 | 1998-01-27 | Petroleum Geo-Services As | Propeller configuration for sinusoidal waterline ships |
DE19614500A1 (en) * | 1996-04-12 | 1997-10-16 | Weber Eckhart | Body of boat |
Non-Patent Citations (1)
Title |
---|
See also references of WO0179054A1 * |
Also Published As
Publication number | Publication date |
---|---|
IL154454A (en) | 2009-09-22 |
AU2000242358B2 (en) | 2005-10-20 |
EP1326773A4 (en) | 2005-11-02 |
CA2412908A1 (en) | 2001-10-25 |
CA2412908C (en) | 2010-06-29 |
IL154454A0 (en) | 2003-09-17 |
AU4235800A (en) | 2001-10-30 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Molland et al. | Marine rudders and control surfaces: principles, data, design and applications | |
EP1718518B1 (en) | Transonic hull and hydrofield (part iii-a) | |
US6158369A (en) | Transonic hydrofield and transonic hull | |
CN107672737A (en) | A kind of deformable glider for taking into account rapidity and sea-keeping | |
Uithof et al. | An update on the development of the Hull Vane | |
Kim et al. | Turning characteristics of waterjet propelled planing boat at semi-planing speeds | |
Yanuar et al. | Resistance analysis of unsymmetrical trimaran model with outboard sidehulls configuration | |
US7004093B2 (en) | Low drag submerged asymmetric displacement lifting body | |
Jiang et al. | Effect of hull displacement on hydro-& aerodynamics of a planing trimaran | |
US6843193B1 (en) | Transonic hull and hydrofield (part III) | |
CA2412908C (en) | Transonic hydrofield & transonic hull | |
EP1718520A1 (en) | Transonic hull and hydrofield iii | |
US4907518A (en) | Planing hull for multi-hull sail boats | |
AU2000242358A1 (en) | Transonic hydrofield & transonic hull | |
NZ524162A (en) | Transonic hydrofield & transonic hull | |
Savitsky et al. | Hydrodynamic development of a high speed planing hull for rough water | |
WO1997029940A1 (en) | Production vessel with sinusoidal waterline hull | |
CN113165723B (en) | Ship body with raised part in bottom side area | |
Chengzhu et al. | Hull forms and straight forward CFD free running trials of high-speed shuttle vessels | |
Wahidin et al. | Monohull or Twin hull Shape Selection Based on the Performance of Hull Planning on Fixed Displacements | |
Calderon et al. | Transonic Hull: Theory, Validation, Breakthroughs and Applications to Ships | |
Elaghbash | A CFD Study of the Resistance Behavior of a Planing hull in Restricted Waterways | |
Trucchi | Fluid Mechanics of Yacht Keels | |
Lu et al. | Mechanics Analysis of Drag Reduction in Hull Motion under Wind and Waves | |
Loveday | The design of a hydrofoil system for sailing catamarans |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PUAI | Public reference made under article 153(3) epc to a published international application that has entered the european phase |
Free format text: ORIGINAL CODE: 0009012 |
|
17P | Request for examination filed |
Effective date: 20030113 |
|
AK | Designated contracting states |
Designated state(s): AT BE CH CY DE DK ES FI FR GB GR IE IT LI LU MC NL PT SE |
|
A4 | Supplementary search report drawn up and despatched |
Effective date: 20050915 |
|
17Q | First examination report despatched |
Effective date: 20070213 |
|
STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: THE APPLICATION IS DEEMED TO BE WITHDRAWN |
|
18D | Application deemed to be withdrawn |
Effective date: 20070824 |
|
18RA | Request filed for re-establishment of rights before grant |
Effective date: 20080507 |