Basiliscus Development Plan
Velocity Prediction Program
Time Domain Simulation
Half Scale Protoype
Basiliscus will be a cruising, hydrofoil sailboat, much along the lines of Dave Keiper's Williwaw, but incorporating the experience that has been gained in multihull design in the thirty years since. I plan to build her for my own use, and develop her into a mature ocean-going yacht over the next ten years.
The process for developing the boat starts with defining clear requirements and priorities. These can be divided into four categories: hard requirements, firm requirements, desires, and crazy ideas. One purpose of the engineering analyses is to help sort out which is which, through rationally trading off one desire for another and seeing what the impact is on the hard requirements.
The basic concept is moderate risk - it's been done before. However, it's only been done once, and the concept is not mature enough to be considered low risk. There are very little empirical data available, and none at all for sailing hydrofoils of this size. The budget is also very limited, which rules out tank testing or wind tunnel testing to base the design on experimental data.
So the plan of attack is to depend a great deal on modeling and simulation to explore the basic design parameters and to create a good performing, robust design. However, computer modeling alone is not expected to be accurate or reliable enough, so a subscale, manned prototype will be used to validate the predictions, gain practical experience with sailing hydrofoils, and to serve as a test bed for investigating various foil arrangements.
Only a very limited budget is available, but money will be spent where it can have the most leverage on the total cost. Consulting by a professional naval architect will be used to make up for my lack of offshore experience and structures expertise, and to serve as a cross check on the engineering analyses.
Finally, the method for constructing the boat will be factored in as early in the design as possible, and considerable effort will be made to ensure that the fabrication will be efficient. The boat must be built as cheaply as possible, but no cheaper.
Intended Usage. The usage of the boat will evolve in three overlapping phases. Phase I is the experimental phase. During this period, the boat will be day-sailed to develop the hydrofoils and rig, and to gain experience in sailing a boat of this type. Phase II will consist of club racing and coastal cruising, primarily in the San Juan Islands. Phase III will consist of cruising in the Pacific, from Seattle to Hawaii to New Zealand. Although the anticipated operating area is the Pacific Ocean, nothing in the design should preclude being able to operate world wide.
There is no intention at this time of doing long distance racing, other than coastal races like the Swiftsure Classic, and racing requirements will not be a design driver. Given the unusual nature of the boat, it may require special treatment in handicapping, so there is little point in warping the design to optimize its rating.
There will be four distinct operating modes, corresponding to increasingly severe conditions. The first mode is hull borne sailing in light winds with the hydrofoils retracted. The second mode is foil borne operation with the hulls clear of the water. The third mode, used when the speed of foil borne flight is not prudent, is hull borne sailing with the hydrofoils down. The final mode, survival, is stationary while tethered to a parachute anchor.
Hard Requirements. The boat has three non-negotiable requirements:
Everything else is negotiable, and will be traded off to in order to meet these three requirements.
It must fly. If the boat does not fly, it isn't worth building. There are many fine designs to choose from today, and used boats can be bought for a fraction of the cost of building a new design. The interplay of hull and hydrofoil means that it is best to design the boat from the bottom up to be a foiler, rather than trying to retrofit hydrofoils onto an existing boat. And the satisfaction of having engineered a successful hydrofoil cruiser is as important to me as the sailing of the boat.
The goal is to be able to fly to windward, and the boat should be capable of flying in as wide variety of wind speeds, sea states, and headings as possible. Williwaw did not have enough stability to fly above a reach, and she never had a decent set of sails. Hydrofoils may be of great value in going to windward, as the boat will be lifted above the sharpest wavetops. The ability to travel quickly and comfortably against the wind and waves may open a whole new dimension in cruising. Perhaps even gentlemen will wish to sail to windward. In addition, the ability to perform well closehauled is expected to improve performance on all points of sail, as the boat is expected to be able to achieve a high boatspeed/windspeed ratio.
The ability to fly is the only performance requirement. There is no specification for how fast the boat has to go. However, it is expected that if the performance is sufficient to fly in a wide variety of conditions, that its performance will be satisfactory for cruising and racing. The purpose of the hydrofoils is to raise the average speed of the boat, not necessarily the top speed, with the philosophy that, "It's not how fast you go, it's how much you go fast that counts." Most contemporary multihulls claim top speeds over 20 kt. When these boats do hit 20 kt, it is a wet, wild ride. Until the VPP is operational, it is not possible to guess as to what kind of speeds will be obtained and under what conditions. I am not out to build a 30 kt sailboat. The object of Basiliscus is to make 20 kt as boring as possible.
The ability to fly may also improve the light wind performance. In fact, the boat will spend far more time hull borne than foil borne, so good light wind performance is a firm requirement. Since the foils lift the hulls completely out of the water at high speeds, and the foils can be selectively deployed to provide additional damping if required when hull borne, it may be possible to optimize the hull design for low speed performance. For example, the prismatic coefficient may be lower than comparable boats and the stern sections may not need to be as full, reducing wetted area.
It must cruise. There are two fine hydrofoil daysailors on the market, the Hobie Trifoiler and the Windrider Rave, and if daysailing were the intended use, it would be far better to acquire a proven one-design.
The purpose of Basiliscus is to travel long distances offshore with a crew of one or more. Although the boat will undoubtedly be raced, it is not to be a stripped all-out racer. Firm requirements include stand-up head room and an enclosed head, and the boat should be "female friendly."
For design purposes, the target capability is to have a range sufficient to sail from Maui to Victoria with a crew of four. The boat should be capable of carrying sufficient supplies for a 60 day passage. An inboard auxiliary is also a desirable requirement.
Based on boats with comparable requirements, such as the Brown Searunner 37, Farrier F-36, Hughes 37C, Shuttleworth Damiana, or Grainger TR40, the length will probably be between 35 and 40 ft. Also based on similar boats, the empty displacement, with hydrofoils, will be approximately 5000 lb. With cruising payload, in addition to sails and crew, of around 4000 lb, the maximum displacement is targeted at 10,000 lb.
Cruising requirements will also dictate much of the required equipment. The boat will comply with offshore racing regulations, as a guide to the minimum safe requirements.
It must be affordable. Tentatively, the money available for design and construction of the boat will be $20K per year, starting in 2001. Farrier estimates that an F-36 can be built today for $90K - $130K, and a Contour 44 (40 ft Lwl) production trimaran has a sailaway price of $335K. Assuming that the price scales as length3 or by the pound, a 36 ft production cruising trimaran could cost approximately $250K today. Therefore, it appears that if Basiliscus can be built for $200K over approximately seven years, its acquisition costs will be affordable. This will require considerable home-building, but there may be room for some professional assistance. 10% - 15% of the cost of the boat is allocated to consulting, and this may delay construction for a year. The above assumes that the boat would be fully paid off upon completion, which may not be strictly necessary, either.
If the projected costs are too high, the boat may be scaled down, say, by designing for a crew of three or two instead of four. Since the use of the boat will evolve toward its ultimate capability, the equipping of the boat for offshore cruising can be phased in over a period of time after launching. On the other hand, the development of the hydrofoils can be expected to incur costs which cannot be forecast at this time.
Maintenance, mooring, and other operations costs must also be factored into the design. The hydrofoil offers some significant savings in maintenance costs, because the boat can be allowed to dry out on its foils, making the bottom accessible for routine cleaning or minor repair without hauling out. Wider beam makes mooring more difficult to find, and more expensive, and may dictate the maximum beam of the boat. The use of exotic materials will be limited, both to reduce building costs and to ensure the boat is repairable outside of the U.S.
Requirements Document. This has been only a partial listing of the requirements. A separate requirements "wish list" will be maintained that breaks the requirements down in more detail.
The engineering analyses are designed to answer the questions,
"How fast will it go?"
"Is the ride comfortable?"
"What are good shapes for the amas?"
"How strong does it have to be?"
"Will it fly?"
"Will it be stable in flight?"
"How much wind is required to fly?"
"How much wind does it take to capsize?"
"Is it susceptible to wave-induced capsize?"
In short, will it meet the requirements - will it fly, will it be a good cruiser, and does it require exotic materials that might make it unaffordable?
Velocity Prediction Program (VPP). The VPP is the primary analysis tool, used to predict the flat water performance of the boat. It solves for the steady-state equilibrium of the boat, considering the forces of the wind on the sails, the forces on the hulls, and the lift and drag of the hydrofoils. The VPP will have to handle all the modes of operation, including hull borne sailing with or without the hydrofoils, and foil borne sailing.
The VPP will use Excel spreadsheets to integrate all of its parts. Major computations will be written in FORTRAN or C and compiled into dynamic link libraries (DLL's). The DLL's will be linked to the Excel spreadsheets. The user interface will be implemented in Excel, and Excel will be used to display the results.
The VPP requires the development of several modules. These include a geometry module to generate hull offsets, and a hydrostatics module to compute the buoyant restoring forces. The hull hydrodynamics for the VPP primarily consist of estimating drag. The free surface computational fluid dynamics (CFD) tools available can only handle symmetrical hulls, so this becomes a restriction on the hull shapes. Hull aerodynamics will be based on empirical data on bluff bodies from Hoerner's "Fluid Dynamic Drag." Hull hydrodynamic drag will be calculated with the Michlet code.
Sail aerodynamics will initially be based on the IMS VPP sail model or the WB Sails' Java script sail calculator. As the design progresses, the sail rig will be analyzed using Aerologic's Cmarc, a development of the NASA Ames panel code, PMARC. The Cmarc predictions will be used to predict the impact on performance of detailed changes in the rig design.
Hydrofoil hydrodynamics will be estimated using the handbook methods of Hydronautics, Inc. Cmarc will also be used to estimate the forces and moments on whole foil assemblies.
The steady six-degrees-of-freedom equations of motion will be iterated to balance the nonlinear forces and moments. Sail trim will be adjusted to maximize the performance. The solution will be repeated for a variety of courses and wind speeds to build the predicted boat speed polar diagrams. This will answer the questions, "How fast will it go?" "Will it fly?" and "How much wind does it take to fly?"
Time Domain Simulation. The time domain simulation will answer the question, "Will it be stable in flight?" This simulation will be based on the Scilab analysis tool. Two types of time domain analyses will be made.
A simplified, linear version will be used for preliminary studies of hydrofoil stability in flat water. This will use constant coefficient stability derivatives, either estimated directly or obtained by finite differencing of the nonlinear forces from the VPP modules. The linear analysis will be used to identify the modes of motion in six-degrees-of-freedom and to determine the key frequencies of interest. Frequency responses from the time domain linear analysis can also be compared with the results of the seakeeping analysis to estimate the validity of the quasi-steady, constant coefficient approximation.
The force and moment modules from the VPP will be used in the nonlinear version of the time domain simulation. The purpose of the nonlinear simulation is to examine takeoff and landing dynamics and the response of the boat to large gusts. It also will verify the predictions of the linear analysis for large motions, and may be used for comparison with the seakeeping analysis for flight in regular, long-crested seas.
Seakeeping Dynamics Tool. The seakeeping analysis is the most advanced and complex of the analyses. It considers the unsteady motion about the nonlinear trim conditions established by the VPP. The seakeeping analysis will assume linear dynamics, which restricts it to small motions and small sinusoidal wave amplitudes. The tool available for computing the hydrodynamic coefficients, the US Navy's Standard Ship Motion Program, SMP, also requires that the hull forms be symmetrical.
At present, there are few programs on the market which can handle multihull dynamics, and none that are affordable. The typical approach is to use a nonlinear CFD code to compute the coefficients for the whole ship, including both hulls, and then analyze the motion in the same manner as for a monohull. The dynamics code in hand, SMP, can only handle monohulls.
The general approach to extending SMP to multihulls is to ignore the interaction between the hulls, and compute each hull separately. Hopefully, the wide separation of the hulls will minimize these effects, and the large lever arms will magnify the influence of the more fundamental aspects of each hull. Because the motion of an ama about the boat reference center is very different from the motion about the ama's reference center, it will be necessary to take into account the coupling between the hulls. This will be done by applying a transformation to the coefficients for the individual hulls so that instead of relating the forces and moments referred to the hull center of reference to the motion about the hull center of reference, the transformed coefficients relate the forces and moments about the boat's reference center to the motion about the boat's reference center. This should be straightforward for the radiation forces and added mass coefficients. However, the Froude-Krylov exciting forces from the waves will have to take into account the fact that each hull is at a different place on the wave train, and so the transformation of these forces will depend on the wavelength and wave direction to get the proper phasing. Once the transformed coefficients for each hull are in hand, they can be added together to form the coefficients for the whole boat.
An additional change from the SMP exciting forces will be the estimation of the hydrofoil effects. SMP does not take into account the orbital velocity of the waves, because these effects are of minor importance for hulls. However, the orbital velocities have a major effect on hydrofoil lift. These effects act 90 degrees out of phase with the hydrostatic effects.
Solving the equations can proceed in one of two ways. Once the coefficients for the boat are obtained, they can be fed back into SMP and solved in the usual manner (in the frequency domain). The problem here is that SMP assumes the ship is symmetrical, and separates the six degrees of freedom into a three-degree-of-freedom pitch-heave set, and a three-degree-of-freedom lateral-directional set. Although the trimaran has symmetrical hulls and has bilateral symmetry about the ship reference center, the hydrodynamic coefficients will not be for a symmetrical boat because one ama will typically be out of the water. Thus, SMP assumes that some coefficients are zero, when they will be nonzero in the transformed case. It may be possible to modify SMP to produce a true six-degree-of-freedom solution. Otherwise, the equations will have to be solved in a separate program.
The other method of solution is to transform the equations back into the time domain and solve them there using state space methods and tools such as Scilab. This will probably involve the use of convolution integrals, which may be difficult to formulate and computationally expensive to solve. However, this method would make it simpler to add the forces on hydrofoils (for which the added mass effects are small), and to take into account nonlinearities and the effects of wave orbital velocities.
The influence of the unsteady wind will be handled in a similar manner to the influence of the waves. Gust spectra from aeronautical literature will be used to form the exciting forces on the sail rig at the same frequencies as the wave encounter frequencies. The gusts will then appear much like a separate long-crested wave spectrum superimposed on the long-crested or short-crested wave spectra.
Once the motion spectra for long-crested and short-crested sea states are known, the statistics for the motion at various points on the boat will be compared with criteria for motion sickness to evaluate the quality of the ride. This will answer the question, "Is the ride comfortable?"
Although the dynamic analysis will be linearized on the basis of small waves and motions, the results will be extrapolated to qualitatively evaluate the potential for slamming of the ama and hull. This, in conjunction with the VPP, will answer the questions, "Does the ama have a good shape?"
Capsize dynamics. Initial capsize susceptibility will be evaluated using curves of hydrostatic righting moment vs heel angle. These may be compared with the results of similar boats. Various stability indices will be computed for statistical comparison with other designer's practices. The response to gusts can be predicted using the nonlinear simulation. The response to sharp-crested waves may also be estimated, provided that suitable wave profiles can be modeled.
Structural Loads (static & dynamic). The motion from the dynamics analyses and the best estimates from the aero/hydro codes will be combined to determine critical loading conditions. These will be combined with existing ABS guide or other design rules to calculate the global and local loading for use in the finite element analysis.
Weight Estimation. Initial weight estimates will be based on statistical comparisons of similarly sized boats. For example, a Hughes 37C has an empty weight of 3700 to 4000 lb and a displacement of 6800 lb. A Grainger TR40 has a displacement of 8400 lb, which scales to 6100 lb at 36 ft, and implies an empty weight of approximately 4000 lb. A Brown Searunner 37 has "weekend" displacement of 7000 lb and a cruising displacement of 9200 lb, suggesting an empty weight of 5000 lb to 6000 lb. Shuttleworth's 42 ft Damiana has a cruising displacement of 7500 lb and a racing displacement of 5400 lb, implying an empty weight of around 4400 lb for this simply equipped yacht. A Contour 44 (40 ft LWL) has a cruising displacement of 11,000 lb, which scales to 8000 lb at 36 ft LWL. Other statistical sources include Shuttleworth's plot of displacement/length ratio's and articles in Multihulls Magazine. For payload, Brown recommends 1000 lb per berth. Other rules of thumb include a payload of 3000 lb for a 36 ft trimaran cruiser.
Taking the modern, high performance designs' figures with a grain of salt, and allowing for the fact that not all have stand-up headroom, etc., yields the initial target weight of 4500 lb for the boat itself. This also has to allow for the fact that a hydrofoil hull has a number of point loads at the hydrofoil attachments, and the resultant hull reinforcement will add some weight. Adding 500 lb for the hydrofoils (an optimistic weight, based on Williwaw's 400 lb foils) and 1000 lb for sails and four crew yields 6000 lb for the racing displacement, 8000 lb for the mid cruising displacement, and 10,000 lb for the maximum cruising displacement.
As the design progresses, a spreadsheet will be maintained with a bottoms-up estimate of the weight of each major component. During the detailed design, each piece of hardware will be tracked on the spreadsheet.
Mass properties will initially be estimated by allocating the total weight to major components, such as main hull, amas, and mast. The center of gravity and moments of inertia will be estimated by considering each major component to be a simple solid of uniform density. As the bottoms-up weight estimate is established, each piece will have an estimate of its moment of inertia, and the mass properties of the boat will also be built up in detail.
Half Scale Prototype
The state of the art in CFD today still does not allow accurate estimates to be made in all cases, without calibrating the calculations with empirical data. This is especially true for the simpler, linear techniques available for this project. In addition, hydrofoils have historically required an extended period of development for each configuration. Therefore, some sort of test program will be necessary. Tow tank testing would be useful, but would consume the entire project budget. Instead, a half scale sailing model of the cruiser will be made to serve as a test bed.
The purpose of the prototype is to serve as a hydrofoil test bed and to validate performance predictions. It may also serve to validate handling qualities and balance. The performance predictions will be validated by predicting the performance of the prototype configuration, in order to validate the prediction process. Therefore, strict similarity to the cruiser configuration is not required, however the closer the test bed is to the cruiser configuration, the better, because the results themselves may be scaled up to be a cross check on the cruiser computations.
An additional purpose of the prototype is to gain experience in boat building and to experiment with different materials and fabrication methods. The experience gained in building the prototype will probably pay for itself in improved detailed design of the cruiser and a shorter learning curve in its construction. For this reason, it is desirable that the prototype use like materials for the hull and similar and hardware for such items as foil attachment points.
The cheapest, and fastest, way to produce a half scale prototype is purchase a used beach cat and modify it to represent the cruiser configuration. The beach cat will be turned into a half scale prototype by building a half scale model of the center hull and extending the beams. The existing beach cat rig will be retained, but it may be relocated to match the cruiser's balance. Lateral resistance will come from a centerboard or dagger board in the main hull rather than the dagger boards of the beach cat. Alternatively, the beach cat's dagger boards and rudders may be used when hull borne. Of course, the hydrofoils will provide the lateral resistance and steering when deployed.
For example, say the cruiser has a sail area of 900 ft2, a length of 36 ft, and displacements of 6000 lb to 10,000 lb. This gives it a displacement/length ratio of 57 to 96, and a Bruce number of 1.6 to 1.3. A candidate prototype would be a Hobie 18, which has a weight of 400 lb and a sail area of 240 ft2. Assuming that the new 18 ft hull and beams weigh 300 lb, and a single crew weight of 250 lb, the prototype would have a displacement of 900 lb, a displacement/length ratio of 69 and a Bruce number of 1.6. With two crew on board, the prototype would have a displacement of 1150 lb, a displacement/length ratio of 88, and a Bruce number of 1.5. The latter case is an almost perfect match for the cruiser at 9000 lb, and the two cases bracket the mid cruising weight quite nicely.
Of course, the beam of the half scale prototype can be sized so as to match the stability of the cruiser. Matching other dynamic parameters, such as moments of inertia may not be possible. In general, the dynamics and handling of the prototype should be similar to the cruiser, but will have higher frequencies due to the scaling. It will be as if time moves faster on the prototype, probably by a factor of around 1.4.
Hydrofoils for the prototype are more problematic. Consider possible scaling laws for the hydrofoils. Let:
K = scale factor = 0.5
lwl = length of the waterline
L = lift
W = weight
CL = lift coefficient
rho = density
V = boat speed
b = effective span = area/chord
c = chord
The total lift on the hydrofoils is
L = W = CL * 1/2 * rho * V2 * b * c
To scale from one size to the other, one takes the ratio of the lift on the model to the lift of the full scale boat.
W2/W1 = CL2/CL1 * (V2/V1)2 * b2/b1 * c2/c1
Applying the scale factor, K, one can solve for many of prototype requirements:
lwl2 = K * lwl1 (geometric scaling)
W2 = K3 * W1 (same displacement/length ratio)
CL2 = CL1 (same foil section and angle of attack)
b2 = K * b1 (geometric scaling)
Let the velocity scale by the relationship
V2 = Kp * V1
The exponent, p, determines the type of scaling that will be applied to the hydrofoils. The span of the hydrofoils is fixed because they have to fit within the dimensions of the hulls. But the chord length of the hydrofoils is not constrained, and can be adjusted to fit the type of scaling desired. If the hydrofoil chords scale as:
c2 = Kn * c1
Then the chord scaling can be related to the velocity scaling by substituting the above into the equation for the weight ratio and solving for the chord ratio, yielding the following relationship:
K3 = 1 * K2*p * K * Kn
n = (3 - 2*p -1)
So the chord of the prototype hydrofoils depends on what one assumes for the scaling on boat speed. There are several possible choices that can be made for p, but two are the most promising.
Option 1: Froude scaling: Set p so that both boats will lift off at the same fraction of hull speed, or Froude number. By setting the Froude number of the prototype to the Froude number of the cruiser, one can solve for p and n:
Fr = V/(g*lwl)0.5 = V1/(g*lwl1)0.5 = V2/(g*lwl2)0.5 = V2/(K*g*lwl1)0.5
V2/V1 = K0.5
p = 0.5
n = 1
Since n = 1, c2/c1 = K. The model chord is geometrically scaled, just like the span. This means that the chord on the prototype hydrofoils will be half that of the cruiser. Equal Froude numbers mean that the hump speed characteristics of the two boats will be a similar as possible.
Option 2: Constant foil loading. Another approach is to keep the weight carried by each square foot of foil area to be the same:
W1/(b1*c1) = W2/(b2*c2)
W2/W1 = (b2/b1) * (c2/c1)
K3 = K * Kn
n = 2
p = 0
Since p = 0, the takeoff speeds are the same for the prototype and the cruiser. Since the speeds are the same, and the lift coefficients are the same, the absolute pressures on the foils will be the same. These hydrofoils would have the same susceptibility to cavitation and ventilation as the full scale foils. Naturally, it will be more difficult for the prototype to take off, since its Bruce number is similar to the cruiser's and it has to sail to a higher Froude number. Whether or not this is a problem will depend on how slow the cruiser's liftoff speed is and whether it is even within the achievable performance of the prototype.
The major factor that makes this option attractive is that it would allow the use of aluminum foil extrusions already on hand, purchased from the Keiper estate. For a half scale prototype, the foil chord will be one quarter of the cruiser's chord. Preliminary estimates indicate that the cruiser's foil chords will be on the order of one foot, which makes this a good match to the three inch foil extrusions. Unfortunately, the section shape on the extrusions, approximately a Clark Y, is quite different from the custom foil shapes, such as the H005, that will be used on the cruiser. This limits the utility of the existing extrusions.
Therefore, the plan for the prototype will be to start with hydrofoils made from the existing aluminum extrusions, so as to gain experience in building and sailing a hydrofoil. Since ladder foils are being used, it may be possible to double the number of rungs to gain the extra area needed to match the Froude scaling speeds. As time and money permit, Froude scaled hydrofoils will be fabricated and tested on the prototype. This will also provide experience in fabricating custom hydrofoil units, and offer the opportunity to experiment with candidate materials and processes, which will also be relevant to the full scale boat.
Instrumentation of the prototype will be consistent with its low cost nature. A hand held annemometer and GPS would provide the means to measure windspeed/boatspeed as well as absolute boat speed. It may be worthwhile to invest in commercial electronic instrumentation, since it could later be installed on the cruiser itself.
Testing of the prototype will include general handling, and stabilized speed measurements on a variety of headings. Particular attention will be paid to the behavior of the hydrofoils, liftoff speeds under various conditions, and the difference between foil borne and hull borne speeds. Results will be compared to predictions for the prototype configuration. An attempt will be made to resolve differences between prediction and test data so as to obtain factors that can be applied to correct the modeling of the cruiser performance.
The prototype test program should provide partial answers to all of the key engineering questions. The comparison between prediction and test results should indicate the reliability of the estimates and the areas where more work is required. The experience gained with the candidate materials and fabrication processes will help determine if the cruiser is affordable. And the existence of an inexpensive test bed may prove valuable even during the experimental phase of the cruiser itself.
My background is in aeronautical engineering, small boat sailing, landsailing, and crewing in round-the-buoys racing. My engineering experience includes aerodynamics, flight dynamics, and flight testing. I lack offshore experience, and I have not built a boat before, although I have built an experimental landyacht. Therefore, I need advice from someone who is experienced in designing and sailing multihulls offshore, and has the expertise to do the engineering analyses that I cannot, principally in the area of structural design. I have budgeted 10% to 15% of the cost of the boat (or $20K to $30K) to be applied to consulting. The hope is that by doing as much of the groundwork as possible, and documenting the methods and results, that the consulting expertise will be used a efficiently and effectively as possible.
The consulting effort will be divided into two phases; a review of the conceptual design, and detailed design support.
Phase I: Conceptual design review. The purpose of the first phase is to critique the preliminary design concept and to establish a plan for the rest of the development effort. The ideal time for this review will be after the VPP is working and some initial trade studies have been done. At that point, the design should have some credibility and its general characteristics known.
The tasks for the initial consulting effort are:
This phase of the consulting effort is intended to be short, and focused on determining the tasks which must be completed before the second consulting phase. It is also intended to inject realism into the design and focus it on practical matters.
Phase II: Detailed design & building plan. The purpose of the second consulting phase is to complete all the engineering necessary to begin construction. The bulk of the consulting effort will occur in this phase. At this point, the performance and dynamics analyses will be complete, and the design will have gone through several trade studies, and been tested on the subscale prototype. Static and dynamic loads will be available. Most of the drawings will exist in digital form.
The tasks for the second consulting effort are:
The finite element analysis will require the lion's share of the resources for this phase. The ability to do such an analysis will likely drive the choice of consultants. The choice of consultant will depend on the ability to do the desired analysis, multihull background, interest in the project, and price.
In addition to the naval architecture consulting, potential builders will be identified at this time and brought into the discussions. Considerable design changes may be required in order to reduce the costs of building the boat. One goal of the extensive design analyses is to produce a firm design for which changes during building will be minimal.
The purpose of the logistics effort is to address the affordability of the yacht. The key questions that have to be answered are:
"Does the total cost fit within the budget?"
"What is the best allocation of the budget to achieve a quality result?"
"What is the best source of materials and equipment?"
"What will be required to build the boat, in addition to the boat materials themselves (tools, facilities, etc.)?"
"What part of the construction is best done by professionals, and what by me?"
"Who should the builders be, and where should it be built?"
"How will the boat be supported after it is completed?"
Make/buy breakdown. The biggest question regarding make vs buy is the hull structure itself. The current plan is to have the hulls and akas professionally built, then fitted out at home. This uses the skill of the builder where it counts most; uses the builder's facility with its space, tools, and hazardous waste handling (spray booth and the like); and avoids paying shop time and labor for the time consuming but simpler task of installing the hardware.
It also cuts years off the building of the hull structure, which allows me to spend that time working on the design and saving money. I hope that by investing the money, the gains will partially offset the added cost of the professional labor. However, the investment period is not very long (approx. 5 years) and it may only offset the growth in cost of materials. Still, this approach is in keeping with my philosophy of making the best use of professional services where it makes sense. In addition, having a builder involved will provide a consultant that can provide valuable input all through the design and construction.
Secondary structure, such as interior joinery, can be done at home. This will primarily consist of flat composite sandwich panels, which can be formed and vacuum bagged on a flat table.
The mast will be handled the same way as the hulls - professional construction of the mast itself, and home installation of the hardware.
As for the hydrofoils, I will be looking into the possibility of having dies made for custom composite pultrusions. It appears that it may be possible to build all the foils using one section, or possibly one cambered section and one symmetrical section. Pultrusions may be cost effective, given the large linear amount of material in the ladder foils, plus the need to allow for spares, experimental versions, and scrap.
After pultrusions, the next most likely method of fabricating the foils will be composite hand layup in female molds. This can be done entirely at home. All the foil segments are 8 feet long or less, and there are many pieces, making it feasible to construct the tooling for limited series production. The plug may be machined professionally using numerically controlled machinery to ensure a high degree of accuracy.
If the foils are to be made of aluminum, they will have to be professionally fabricated. As with composite pultrusions, it may be possible to have dies made for custom extrusions. As a last resort, the foils will be made of rolled, welded aluminum plate, as were Williwaw's foils. However, this is not likely to yield the desired accuracy in the section profile, and Williwaw continually experienced problems with fatigue cracking of her welds.
Hardware will be bought new through commercial sources. I do not have the machinist experience necessary to make my own, and I don't trust used hardware. The expense of the hardware can be phased in over the intended use of the boat, with a simplified fit-out used to get the boat sailing, and the full sea-going complement added over a few years of inshore use.
Materials sources. I am accumulating a list of possible sources by monitoring the Multihulls Mailing List and talking to sailors in the NW Multihulls Club. I will also be consulting with the builder as to the best sources of materials, and may well be purchasing much through the builder himself, if that results in the lowest net cost.
Facilities. Under the current plan, the builder's facility will be used for construction of the hulls. These will be transported individually home for fitting out under a bow frame & plastic enclosure. Once the hulls are complete, they will be transported individually back to the boatyard for final assembly and launching. This solves the problem of getting a 28 ft wide boat out from behind my house and transported through the half mile of city to the water.
Tools. Tooling will be simplified due to the use of a professional builder and his facility for the major construction. A flat table with vacuum bagging equipment will be required. Basic hand tools and composite material handling (resin pumps and the like) will be needed. A table saw and band saw will be needed for cutting composite core and wood joinery, and in general the tools required will be those of a basic home woodworking shop. Jigs will be required for fabrication of the hydrofoils. Inexpensive laser pointers or home-construction type laser tools will be used as references for alignment.
Transportation. Transportation of the boat will be eased by transporting the hulls individually. These can be moved on an ordinary flat bed truck or trailer.
Berthing location. Berthing of a multihull in Puget Sound will take some advance planning, since all the marinas have waiting lists that are years long. I will be getting on various waiting lists as soon as construction actually begins. The most convenient location would be the Des Moines Marina, since it is within walking distance from home. But it is small and has a limited number of spaces that can accommodate a multihull. It may be necessary to find a moorage on Vashon or Bainbridge Islands while waiting for a more convenient berth on the mainland.
The construction planning has two purposes. The first is to determine what aspects of the design drive the costs, so that the design can be modified at the early stages where construction considerations can have the biggest impact on the cost. The second purpose is to plan the construction process for execution.
Part fabrication. The design of the boat will be done using CAD, which opens up some possibilities over traditional boat building. In general, I will be investigating shops with numerical machining capability to cut parts for the boat, including tooling and composite cores. This may make it possible to provide the builder with precut pieces, like a kit, with the labor savings making the added expense worthwhile. The labor savings are expected to come from having the pieces readily available for assembly, reduced fitting and rework, and a more accurate hull shape requiring less fairing effort.
Hulls. I plan to consult extensively with the builder and spend a great deal of design time on the construction process itself. Features in the hull design which are difficult to build will be modified or eliminated. Once construction starts, changes to the design will be limited to those which are absolutely necessary to remove obstacles to getting the boat built, or those which result in a net reduction of the cost of building.
The materials to be used for the hulls, and thus the construction process, will be determined as part of the structural design and analysis. For planning purposes, the baseline construction will be of Corcel foam core with s-glass skin and local carbon fiber reinforcement. Although cold molded wood, or even wood skinned sandwich construction is not ruled out. Plywood is unlikely due to the rounded contours. Wood core, such as Western Red Cedar, is attractive in cost, but may be excessively heavy. Cylinder mold and constant camber methods are also unlikely due to their inability to accurately achieve the desired shape. And accuracy in building is important to realize the performance benefits of the engineering and to minimize the labor involved in constructing the hulls through the use of CAD/CAM.
Modern methods, such as resin infusion, will be investigated to see if they are cost effective for this one-off boat. There should be considerable progress in this area by the time the boat is built.
Interior. Interior joinery will be sandwich panels for lightness. A thin wood veneer may be used on one surface for aesthetics, with glass used for the facing skin.
The current design lends itself to having the center section of the boat interior, including the centerboard trunk, flammables locker sole, cockpit seats, bulkheads, engine compartment, and berths all constructed as a single unit using flat panels. If male molds are used, this center section could be constructed as a unit and used as a major part of the mold. If female molds are used, this unit could be dropped into place once the interior of the shell is laid up.
Spars. A rotating spar will be used unless there is an overwhelming advantage shown for a fixed spar. The spar will have a custom section. By the time the boat is built, it is likely that carbon fiber costs will have dropped to the point where carbon will be only real choice for the spar. This will undoubtedly require professional fabrication. As mentioned under Logistics, the spar hardware will be installed at home, although the stays will be fabricated by a professional rigger.
Hydrofoils. Assuming that the foils are made of composites, either pultrusions or laid up, constructing the hydrofoil units will be largely a matter of trimming the sections to length and gluing them together. Fairings at the intersections will both reduce the interference drag and reinforce the joints. A key requirement for the hydrofoil assembly is high accuracy in aligning the foil elements during assembly. This will require extensive jigging. I have a cousin who is a Boeing structural engineering manager, and I will be consulting with him on applicable aerospace practices for constructing the foil units.
Fitting out. As covered above, the vast majority of the hardware installation will be done at home using hand tools, and my own semi-skilled labor.
Final assembly & launching. Final assembly is another area where the builder's skill in ensuring accurate alignment, and the equipment necessary to handle large, awkward hulls, will be of benefit. Depending on the builder and his facilities, this may be a different boatyard from the one where the hulls were originally constructed.
I have not decided at this time whether the hulls will be demountable or not. The boat will definitely not be trailerable. But demountable hulls may be useful for transporting the boat in the future, for making major repairs, or for experimenting with new amas.