Advanced learning material on hull shape design?

[laukejas]

Hey guys. Some of you know me, I've been designing boats for some ~12 years now, around 80 hulls in total, most of which remained on paper, but I've built 3 boats so far, one more is being built right now, and 2 more of my designs built by others. I consider myself a self-taught amateur, but I've always taken this hobby very seriously. I've finished Skene's Elements of Yacht Design, Gougeon Brothers On Boat Construction, Gerr's The Nature of Boats and over a dozen other books, watched hundreds of video tutorials/instructionals, consulted with great many people, even coded a buoyancy/stability physics calculator for evaluating hull designs, but I feel like I've plateaued now, and I'm unsure on how to continue learning hull shape design.

The main subjects I'm interested are:

1. Hull shape optimization for reduced resistance in displacement mode;
2. Application of the fluid compression and particle deflection theory for hull design;
3. CAD techniques for creating parametric hull designs and linking them with CFD simulations.

In my research, it seems that there is a lot of theoretical info on hull/water interaction, but I can find very little on how to actually apply it when designing a hull, how to evaluate predicted performance, and how to make changes to the hull shape until predictions converge on the required performance.

In the old days hulls were designed with pencil and ruler, but nowadays I expect pretty much everyone (like me) is using CAD software and CFD simulations. For the old-school techniques there are plenty of tutorials. But very little on how to actually design a hull in CAD software. I am especially interested in advanced techniques for parametric CAD design (as opposed to NURBS-based modelling like with Rhino). Most of these ~80 hulls that I designed, I did in SolidWorks, but my entire path was self-taught and improvised. There are so many ways of laying out a parametric hull design (lofting through stations, lofting through waterlines, lofting through chines, dozens if not hundreds of ways to define and control shape variables, etc.). I often feel like I'm re-inventing the wheel.

I haven't found any books on this subject, and the video tutorials I've found on Youtube are orders of magnitude more primitive than the stuff I already discovered on my own. DMS channel has some really good theoretical info, but again, never showing how to actually apply it on a CAD design.

Also, since most of my designs are small sailing dinghies or yachts aimed at racing, I am very much interested in very specific design techniques to reduce hull resistance and utilizing modern CFD calculations to create parametric design studies to converge on the optimal shape. I am very much intrigued by the particle deflection theory, but again, I can find very little info on how to apply it to a specific design.

Any suggestions - books, videos, anything else - would be really appreciated.

 

[starcmr]

For advanced hull design optimization, consider "Principles of Yacht Design" (Larsson & Eliasson) for applied hydrodynamics. Explore CAESES, OpenFOAM, Rhino + Grasshopper with Orca3D, or Siemens NX for parametric hull modeling and CFD. Research papers on CFD-driven hull optimization (ResearchGate, ScienceDirect) can provide cutting-edge insights. Communities like SNAME and HydroComp forums offer discussions with professionals. Automating shape iteration using MATLAB, Python, or OpenFOAM can refine hull resistance studies. For particle deflection theory, reaching out to computational hydrodynamics experts or attending industry workshops may yield practical applications.

 

[Ad Hoc]

Hi laukejas,

Hmmm…I think you have fallen into the trap that most armatures do. Assuming there is some magic or black art related to hull design and aspects related to it, because it feels so far away from what you know, or have seen.

Sorry to burst your bubble, but there is no real magic to it.

One can break id down to simple length-displacement ratio (LD). This is the mass/weight of the boat as a ratio over its waterline length. It is a non-dimensional value that is:
L/(displacement)^1/3

L in m, displacement in tonne

If you have a ratio that is low, i.e. around 4-5, as can been seen in the graph below, the drag, and hence resistance is high. Whereas if the LD ratio is high, the resistance/drag is less.

There is a plethora of data out there all stating the same…it is not rocket science…just data.
And the same whether a mono or a catamaran as also noted here:

And when you investigate hulls of the same LD ratio, and alter the basic parameters, such as Beam to Draft ratios (B/T)….the change in resistance is also minimal - just very minor 'sweet spots'. In other words, hull shape plays a very minor role, compared to the LD ratio.

So, it really depends upon whether you think there is some magical black art that awaits and is explained by the use or enhanced by computers, and waste hours/days chasing that rainbow….or just want the basic facts and accept there is nothing magical, except the simple fact that the more weight there is for a fixed length, the greater the drag.

Does this mean changes to hull shape etc do not have an influence, no, but the effect is less and less to being almost non-existent when you have a high LD ratio, but has ‘some’ influence when you have a low LD ratio.

As for CFD...unless the CFD is validated against known data and the hull you are analysing is also within the narrow range of the validation, the output of the CFD is highly questionable at best.
Pretty looking colours though!

 

[Skip Johnson]

For long skinny hulls Michlet with Godzilla is a good way to while away some time. Michlet is a shareware program and the Godzilla optimization routine has helped a number of marathon racing canoes carry their crews across the finish line in first place.

 

[BlueBell]

Laukejas,

Good advice above.

Designing to a target boat speed may be the most significant efficiency gain available.
Minimizing wake is the objective.

 

[Skip Johnson]

Actually Godzilla balances wave drag and skin friction to find an optimum, It's tricky if you specify a specific speed, Godzilla will find an optimum but it can be 'peaky' EG there's a distinct dip at the design speed which implies really increased effort above target speed and a bit more effort at a cruise speed. I usually recommend designing for 80% race speed and 20% sprint speed; this seems to work best for long distance races.
Bear in mind that we're talking about differences of a fraction of a percent improvement at best. For marathon canoe races a boat has never won a race but have been a pretty sharp tool for the competitive racers.

 

[jehardiman]

In the old days hulls were designed with pencil and ruler, but nowadays I expect pretty much everyone (like me) is using CAD software and CFD simulations. For the old-school techniques there are plenty of tutorials. But very little on how to actually design a hull in CAD software. I am especially interested in advanced techniques for parametric CAD design (as opposed to NURBS-based modelling like with Rhino). Most of these ~80 hulls that I designed, I did in SolidWorks, but my entire path was self-taught and improvised. There are so many ways of laying out a parametric hull design (lofting through stations, lofting through waterlines, lofting through chines, dozens if not hundreds of ways to define and control shape variables, etc.). I often feel like I'm re-inventing the wheel.

The reason you cannot find a good cheap sandbox integrated CAD and CFD design program for hull form is because there are none. They would be too limiting in design space and are realistically only providing a false feeling of superiority over existing model basin standard series data...which the programs are based on anyway.

Hmmm…I think you have fallen into the trap that most armatures do. Assuming there is some magic or black art related to hull design and aspects related to it, because it feels so far away from what you know, or have seen.

Ad Hoc is correct here. There is no one "best" hull or hull design method. What can be said is that there are "better" or "worse" hull forms given the desired requirements. Changes in the basic design parameters (weight, length, speed, draft, beam, midship location, midship shape, forebody, run, ect.) change, and interact with each other, to effect the residual resistance and propulsion coefficients. This is why a Statement of Requirements (SoR) is so important; there is not one "perfect" hull form, but there is one that best suits the SoR and its weighted requirements (i.e. for a given effective horsepower (ehp) you can't have both maximum speed and maximum carrying capability...one trades off against the other... a choice the computer can't make, but the naval architect must.)

I often call Naval Architecture the last guild system. There is no book where you can "look up" the correct design process; it is taught, either in university or the design office, and learned by studying other naval architects and their designs: good and bad. While there are books that detail how basic vessel parameters interact with each other (Skene's, Larsson & Eliasson, Hydrodynamics in Ship Design by Saunders, Text-Book of Theoretical Naval Architecture by Attwood, Applied Naval Architecture by by Munro-Smith, and a whole bunch from the old Soviet Bloc that don't come readily to mind) there is no single "correct" method to put all the conflicting concepts together. Perhaps you are lingering under yoke of the old Soviet "Design Bureau" mentality where there was a single state approved method.

As for CFD...unless the CFD is validated against known data and the hull you are analysing is also within the narrow range of the validation, the output of the CFD is highly questionable at best.
Pretty looking colours though!

CFD = Colourful Fanciful Drawings.

 

[laukejas]

Appreciate your replies, guys. Let me comment on each.

For advanced hull design optimization, consider "Principles of Yacht Design" (Larsson & Eliasson) for applied hydrodynamics. Explore CAESES, OpenFOAM, Rhino + Grasshopper with Orca3D, or Siemens NX for parametric hull modeling and CFD. Research papers on CFD-driven hull optimization (ResearchGate, ScienceDirect) can provide cutting-edge insights. Communities like SNAME and HydroComp forums offer discussions with professionals. Automating shape iteration using MATLAB, Python, or OpenFOAM can refine hull resistance studies. For particle deflection theory, reaching out to computational hydrodynamics experts or attending industry workshops may yield practical applications.

This is a lot of very useful info, thank you! I have read Principles of Yacht Design several times and I reference it often. However, I haven't tried most of these softwares you mentioned, nor investigated these communities. I should clarify that I'm specifically after applied theory, since I'm not nearly smart enough to make much use of generic research papers on CFD. But perhaps there is something specific enough - I will check.

Hi laukejas,

Hmmm…I think you have fallen into the trap that most armatures do. Assuming there is some magic or black art related to hull design and aspects related to it, because it feels so far away from what you know, or have seen.

Sorry to burst your bubble, but there is no real magic to it.

One can break id down to simple length-displacement ratio (LD). This is the mass/weight of the boat as a ratio over its waterline length. It is a non-dimensional value that is:
L/(displacement)^1/3

L in m, displacement in tonne

If you have a ratio that is low, i.e. around 4-5, as can been seen in the graph below, the drag, and hence resistance is high. Whereas if the LD ratio is high, the resistance/drag is less.

There is a plethora of data out there all stating the same…it is not rocket science…just data.
And the same whether a mono or a catamaran as also noted here:

And when you investigate hulls of the same LD ratio, and alter the basic parameters, such as Beam to Draft ratios (B/T)….the change in resistance is also minimal - just very minor 'sweet spots'. In other words, hull shape plays a very minor role, compared to the LD ratio.

So, it really depends upon whether you think there is some magical black art that awaits and is explained by the use or enhanced by computers, and waste hours/days chasing that rainbow….or just want the basic facts and accept there is nothing magical, except the simple fact that the more weight there is for a fixed length, the greater the drag.

Does this mean changes to hull shape etc do not have an influence, no, but the effect is less and less to being almost non-existent when you have a high LD ratio, but has ‘some’ influence when you have a low LD ratio.

As for CFD...unless the CFD is validated against known data and the hull you are analysing is also within the narrow range of the validation, the output of the CFD is highly questionable at best.
Pretty looking colours though!

Thank you for your comments, I agree with most of it. But I may have given a wrong impression, I certainly don't consider hull design magic or black art, but I do want to understand it better than I do right now. You are correct that LD ratio is the most important factor, but as you know it can often be restricted by other design considerations, and maximizing the "minor" effect of the hull shape, even if we're talking about a few percent of difference in performance, is very important to me.

As for CFD, I'm prepared to believe you; I just thought it's the best approximation/prediction of the real world performance that we have right now if there is no similar hull data to interpolate from.

For long skinny hulls Michlet with Godzilla is a good way to while away some time. Michlet is a shareware program and the Godzilla optimization routine has helped a number of marathon racing canoes carry their crews across the finish line in first place.

Thank you, I will investigate this. I tried Michlet some years back but completely forgot about it. Will give it another chance!

The reason you cannot find a good cheap sandbox integrated CAD and CFD design program for hull form is because there are none. They would be too limiting in design space and are realistically only providing a false feeling of superiority over existing model basin standard series data...which the programs are based on anyway.

Quick correction, I'm not after some magical software to do it for me - I'm after knowledge of ways how to do it myself, hopefully with aid of some software. Like I said, I do most of my designs in SolidWorks, but I try to be flexible here. It's just a tool. What I want is the know-how.

Ad Hoc is correct here. There is no one "best" hull or hull design method. What can be said is that there are "better" or "worse" hull forms given the desired requirements. Changes in the basic design parameters (weight, length, speed, draft, beam, midship location, midship shape, forebody, run, ect.) change, and interact with each other, to effect the residual resistance and propulsion coefficients. This is why a Statement of Requirements (SoR) is so important; there is not one "perfect" hull form, but there is one that best suits the SoR and its weighted requirements (i.e. for a given effective horsepower (ehp) you can't have both maximum speed and maximum carrying capability...one trades off against the other... a choice the computer can't make, but the naval architect must.)

I often call Naval Architecture the last guild system. There is no book where you can "look up" the correct design process; it is taught, either in university or the design office, and learned by studying other naval architects and their designs: good and bad. While there are books that detail how basic vessel parameters interact with each other (Skene's, Larsson & Eliasson, Hydrodynamics in Ship Design by Saunders, Text-Book of Theoretical Naval Architecture by Attwood, Applied Naval Architecture by by Munro-Smith, and a whole bunch from the old Soviet Bloc that don't come readily to mind) there is no single "correct" method to put all the conflicting concepts together. Perhaps you are lingering under yoke of the old Soviet "Design Bureau" mentality where there was a single state approved method.

I believe you, and again, just to correct, I am not after some single magical design process that works for any case; I know there must be many different workflows and methodologies applicable for different scenarios; so I want to learn them so that I can know which to apply in any given situation. It's just that every time I design something, any kind of hull, I have no methods to draw from at all, I am forced to improvise with a help of a few books that I mentioned, without knowing whether this is a good approach for that specific situation or not. I clearly lack proper education here.

As to the second paragraph, this is an issue: in my country there are no universities or design offices with any more knowledge about hull design than I already possess (not exaggerating my pitiful knowledge; it's just that there is nothing whatsoever here in my country), and since this is just a hobby for me, I can't really travel the world to seek out formal education on hull design. I know this is my problem, your advice is still valid, just explaining why I'm after books/videos/any other kind of info accessible without physical presence in another part of the world

 

[Ad Hoc]

... but as you know it can often be restricted by other design considerations, and maximizing the "minor" effect of the hull shape,...

The problem is Design is holistic, it is not an absolute.
As noted nicely here by JEH:

... This is why a Statement of Requirements (SoR) is so important; there is not one "perfect" hull form, but there is one that best suits the SoR and its weighted requirements (i.e. for a given effective horsepower (ehp) you can't have both maximum speed and maximum carrying capability...one trades off against the other... a choice the computer can't make, but the naval architect must....

And that's the crux of it.

In your attempt to perfect or find the "minor" tweak, more often than not, chasing it, will put one, or more, of the many other design objectives off from its own 'sweet' spot.

Take another simple parameter, deadrise on a chine boat.

If your design is always sitting at Fn of 0.8, one can see slightly less resistance with a deadrise of 16deg, v 38deg.
But this then begs the obvious question, unless the vessel is a high speed vessel destined to always be in rough seas, why bother with a 38 deg deadrise, simply to have "slightly" less resistance.
Oh, that's the point less resistance, isn't it???...., well, ok..but now...you have much less internal volume owing to the increase in deadrise....oh, bummer, how do i fit my engines in, without now raising the KG (reducing statical stability and likely its motions too), and not increasing structure weight and freeboard etc too.
And on it goes.

There is no perfect Design, only a trend that satisfies all the competing elements that become - The Design.

Thus, focus on the trends, not the absolutes.

 

[laukejas]

The problem is Design is holistic, it is not an absolute.
As noted nicely here by JEH:

And that's the crux of it.

In your attempt to perfect or find the "minor" tweak, more often than not, chasing it, will put one, or more, of the many other design objectives off from its own 'sweet' spot.

Take another simple parameter, deadrise on a chine boat.

If your design is always sitting at Fn of 0.8, one can see slightly less resistance with a deadrise of 16deg, v 38deg.
But this then begs the obvious question, unless the vessel is a high speed vessel destined to always be in rough seas, why bother with a 38 deg deadrise, simply to have "slightly" less resistance.
Oh, that's the point less resistance, isn't it???...., well, ok..but now...you have much less internal volume owing to the increase in deadrise....oh, bummer, how do i fit my engines in, without now raising the KG (reducing statical stability and likely its motions too), and not increasing structure weight and freeboard etc too.
And on it goes.

There is no perfect Design, only a trend that satisfies all the competing elements that become - The Design.

Thus, focus on the trends, not the absolutes.

Thank you, this is definitely good advice. But I'm not entirely sure what is the final message you are trying to get across - is it that I shouldn't bother with further learning hull shape design? Or just adjust my expectations? Because like I said, I am aware that I'm chasing after fractions of percent in performance, but since primarily I am designing small racing dinghies that may have very strict restrictions on length/width, there are very few requirements that restrict hull shape - the main (and perhaps - only) criteria being stability and performance while maintaining sufficient buoyancy. So even a massive change in parameters like you mentioned (deadrise) can often be acceptable if it brings even a marginal reduction in resistance.

Again, maybe my question is phrased wrong - the resistance is not the final metric to measure performance, it's the boat getting from point A to point B in as little time as possible in as wide range of conditions as is reasonable. And hull shape definitely plays an important part there, right?

At the very least, I think I should try to be aware of all the influence of each parameter, so that even if design criteria won't let me place all of the parameters in the sweet spot, at least I won't accidentally put them in the worst spot.

So, with my expectations being held in check, what would you say is the best source for learning all that stuff? These graphs that you posted here, where did you get them? I thought they were from Principles of Yacht Design, but now that I'm reviewing that book, I can't seem to find them.

 

[jehardiman]

Again, maybe my question is phrased wrong - the resistance is not the final metric to measure performance, it's the boat getting from point A to point B in as little time as possible in as wide range of conditions as is reasonable. And hull shape definitely plays an important part there, right?

Ummmm, perhaps...if that is the sole point of the SoR, but usually it is not. A boat in the modern world is not about getting from A to B, it is about transporting something, in some condition, from A to B. An open cat could quickly cross the Atlantic, but would you want to? See this old post (and the whole thread is pretty good also). Canting Keel Monos vs Multihulls https://www.boatdesign.net/threads/canting-keel-monos-vs-multihulls.13511/page-5#post-104220

When you get to the end of your knowledge, go study other Naval Architects designs. Place yourself in their position and work out what decisions and trade-offs they made. I know a case where two yacht designers complained about each others design arrangements...until they met....one was 5 foot tall, the other was over 6. (which is why you need anthropometric tables).

 

[Ad Hoc]

....But I'm not entirely sure what is the final message you are trying to get across - is it that I shouldn't bother with further learning hull shape design?...

No, not at all. One never stops learning and reading...

....Or just adjust my expectations?...

Well, yes. Put it into context, and don't over think it.

... but since primarily I am designing small racing dinghies ...

So, can you control the weight of the person sailing? Can you say that any person(s) using the dinghies must not be over 65kg, to get the best performance?
Since we all know - weight kills performance.

There is the law of diminishing returns at play here and you need to understand, as a naval architect, one can only offer the best compromise to achieve the objective.
Since if you calculated that crew of 65kg gains 1 knot, why is this not a requirement?

There are many factors that influence the overall performance, focusing on just one is missing the bigger picture of the holistic view.

Unless of course your SoR is just a simple one liner, like breaking a world speed record, which 99.99% are not designed for such a narrowly defined SoR.

 

[gonzo]

I think the SOR is very focused on a specific race design, limited by rules. In general, those development classes converge into similar shapes. Sometimes it is possible to be creative and design a completely different and faster boat. For example, in the 12 meter class, the wing keel, and the catamaran. Pretty fast, everyone was sailing catamarans. Then the foiler came in. The question to the OP is if he is trying to get a marginal improvement on the existing design, or to study the rules and come up with a revolutionary design.

 

[jameswilson30]

You're definitely far beyond amateur level — your hands-on experience and the sheer number of hulls you've designed is impressive. I can relate to your frustration; transitioning from theory-heavy resources to practical application in parametric CAD and CFD is a leap not often covered in depth.

You’ve probably already explored it, but I’d recommend looking deeper into CAESES for parametric modeling and coupling with CFD. It’s built for exactly what you're aiming for — shape optimization and iterative design. Combining that with OpenFOAM or Star-CCM+ allows you to directly simulate and validate hydrodynamic performance. For parametric workflows, pairing Rhino + Grasshopper with Orca3D gives a lot of flexibility, especially for early hullform development and resistance estimation.

Another overlooked area is automating your iterations using Python or MATLAB scripts. You can even tie in tools like https://calculadoradeehoras.com/ to help you benchmark and schedule simulation windows or design tasks — super handy for managing long-running optimization sessions.

As for the particle deflection theory — you're right, it’s rarely applied in mainstream design guides. I'd suggest browsing ScienceDirect or ResearchGate for naval architecture research papers, or even connecting with CFD pros through SNAME, HydroComp, or Boat Design Net forums.

You're doing seriously advanced work — at this point, industry-level workshops or even online lectures from marine engineering departments might be your best next step.

Looking forward to seeing where you take this next!

 

[Kayakmarathon]

Laukejas,

I started writing my own boat building software when I was in high school. I started out using parametric design and ended up with hollows in the ends of hulls. I experimented with parametric design approach until I realized the solution to avoiding hollows severely limited the designs I do. I then tried a hybrid model using parametrics (scaled families of standard curves) for chines and NURBs for cross sections. The hybrid approach still had hollows.

I read up about NURB based software and thought its user interface was a hindrance. I finally rewrote my boat building software from the ground up using a NURB model with a simple user interface. The model uses 6 patches for the half model: 3 patches describing the bow half of the boat, 3 patches stern half . The lowest of the 3 bow/ stern patches are the "underwater" surface. The middle patches are the "freeboard" surface. The top patches may be used for tumblehome or deck. Each patch is 4x4 points, corners of each patch are on the hull surface.

Hydrostatic analysis is complex since exact solutions for cubic equations don't come easy. Fortunately I sample enough points on the surface and can make linear interpolations. For a given wood strip thickness, NURBs is easier than parametric models to generate accurate lofting stations, although I overbuild the hull and cut away the desired rake angle on a sailboat stern.