Celebrating Sixteen Years Online.
Please also visit Kasten Yacht Design web site.
Home | Intro | Custom Yacht Design | Stock Designs | Motor Yacht Gallery | Sailing Yacht Gallery | Prototypes Gallery | Plans List
Articles | Our CAD Design Stream | MAXSURF For Marine Design | News..! | Design Team | Site Map | Site Search | Contact Us
Please see our AVAILABLE BOAT PLANS web page
Designing Boat Structure
Working With the ABS Rule
Copyright 2011 Michael Kasten
I have often been asked how boat scantlings are determined. For the most part such questions fall into one of the following general categories:
- How is structure determined for boats...?
- What are "appropriate" scantlings for an ocean-worthy vessel....?
Of course there is not a "one size fits all" answer to the second question, since boats differ greatly in size and therefore have different loads that will be imposed upon the structure. I have written the following article in order to describe my own approach to determining a boat's structure - which amounts to being an answer to the first question. The following article is not intended to be a treatise on designing in aluminum, but rather a general overview of the process and a few of the variables to be considered, most of which are applicable to boats built in any material.
As a criterion for structure, I prefer to use the applicable American Bureau of Shipping rules wherever possible. This is because the various ABS rules are based on actual calculations, as opposed to "look-up" tables as are found in many other scantling rules. Via the ABS rules we are able to calculate a minimum plate or panel thickness, as well as a minimum Section Modulus for each member. This "calculated" approach allows considerable flexibility in terms of scantling choices, plate and panel thickness, and the spacing of internal frames, bulkheads, and longitudinal stringers.
The exception however is that for plank-on-frame wooden vessels, unfortunately the ABS stopped publishing their wooden vessel rule in the 1960's..! As a result, for plank-on-frame wooden yachts Germanischer Lloyd's, British Lloyd's Register, and Norway's Det Norske Veritas presently offer the best guides to wooden vessel structure.
For metal or composite boats, among the many different rules published by Classification Societies that address a boat's structure there appear to be many good options. Even so, I prefer using the various ABS rule, which offers many excellent advantages that are described below.
Although the following article uses aluminum boat structures as an example, the following methodology is also directly applicable to steel or composite structure (fiberglass).
The alloys commonly used for boat extrusions (flat bar, pipe, angle, etc.) are predominantly 6061 T-6, or occasionally 6063 T-6 or 6063 T-5 if there is an availability problem with 6061. A new 6082 T-6 alloy has been introduced that offers slightly improved strength, but which can be quite difficult to find in many of the common extrusion shapes (pipe, flat bar and others).
For plate (and any shapes that will be NC cut from plate, such as frames) the most available alloys in the US and Canada are 5086-H116 (the most commonly used alloy with the highest corrosion resistance) and 5083 H-116 (higher as-welded strength, but slightly less corrosion resistance; less commonly specified).
For my designs, if the vessel will be built in North America, I prefer to specify 5083 or the new 5383 alloy from Pechiney / Alcoa. If it will be built in the EU, my preference is to specify the even newer 5059 alloy developed by Corus in Germany. The Pechiney 5383 alloy combines the higher corrosion resistance of 5086 with the higher as-welded strength of 5083. The even newer Corus 5059 alloy improves on both the corrosion resistance and the as-welded strength of Pechiney's 5383.
With regard to material strength, it should be noted that each of the ABS rules for aluminum vessels (the ABS Motor Pleasure Yachts rule, the ABS Offshore Racing Yachts rule, or the ABS Aluminum Vessels rule) specify a variety of minimum un-welded and as-welded strengths. In general though, wherever a published allowable strength is higher in a given ABS rule, it is usually compensated for by different head pressures being calculated for each region (hull bottom, hull sides, deck, house structures, tanks, etc.), or different credits being applied locally.
It has been a long-standing point of confusion among the various ABS rules that there is not a single "agreed upon" allowable strength which applies across all of the ABS rules for each of the alloys. Each "supplemental update" to the ABS rules has further complicated this issue, making the ABS published allowable strength data for aluminum alloys a total mess. In fact I will go so far as to say that it is probable that the widely varying ABS aluminum strength tables have been the result of internal squabbles and / or industry lobbying at ABS.
This is a weakness among the various ABS rules, and of course it has been quite frustrating. In fact the ABS standards for acceptable strengths in aluminum alloys have been in disarray for quite a long time - and they area very much in need of being coordinated. Thankfully, this issue has to a large extent been addressed in the new ABS Rules for Materials and Welding - 2006, Part 2, for Aluminum and FRP (updated in 2010), now referenced from within each of the other ABS rules that address Aluminum vessels.
Even so, there remains considerable confusion within each of the separate ABS rules as to which strength value is to be applied.
Despite these shortcomings in the ABS rule, the "calculated" methodology in the ABS rule provides many advantages for the analysis of structure.
STRENGTH OF VARIOUS ALUMINUM ALLOYS
In general, structures are designed to the yield strength of the material, plus a margin of safety. For aluminum structures, the “as-welded” yield strength is ordinarily used, versus the "fully annealed" yield strength. You can see in the table below that the as-welded yield never approaches the fully annealed condition. Even so, the ABS rule uses the fully annealed condition for each alloy. As a result, the ABS rule assures a considerable built-in safety factor.
A summary of the various ways of expressing the yield strength of aluminum alloys is as follows:
FULLY ANNEALED YIELD
- 5086 – 0: 95 N/mm^2 - 14 kpsi
- 5083 – 0: 125 N/mm^2 - 18 kpsi
- 5383 – 0: 145 N/mm^2 - 22 kpsi
- 5059 – 0: 160 N/mm^2 - 23 kpsi *
- 6061 – 0: 110 N/mm^2 - 16 kpsi
- 6082 – 0: 110 N/mm^2 - 16 kpsi *
AS WELDED YIELD
- 5086 – H-116: 131 N/mm^2 - 19 kpsi
- 5083 – H-116: 165 N/mm^2 - 24 kpsi
- 5383 – H-116: 185 N/mm^2 - 27 kpsi
- 5059 – H-116: 195 N/mm^2 - 28 kpsi **
- 6061 – T-6: 138 N/mm^2 - 20 kpsi
- 6082 – T-6: 144 N/mm^2 - 21 kpsi **
AS MILLED YIELD
- 5086 – H-116: 195 N/mm^2 - 28 kpsi
- 5083 – H-116: 215 N/mm^2 - 31 kpsi
- 5383 – H-116: 230 N/mm^2 - 33 kpsi
- 5059 – H-116: 270 N/mm^2 - 39 kpsi *
- 6061 – T-6: 240 N/mm^2 - 35 kpsi
- 6082 – T-6: 260 N/mm^2 - 38 kpsi *
All values above are from the ABS Rule, 2006 Annex, except for the following:
* Values as-published in the GL Rule.
** As-welded values interpolated (based on the ABS allowable as-welded yield for other alloys).
We can see from the above that there is quite a variation in strength between, say 5086 and 5383, regardless of the "temper" of the material, and there is also a considerable variation in the strength of these alloys depending on their "temper".
For my own designs, in order to make use of the most conservative strength values I use the fully annealed material strength as originally specified within the ABS Motor Pleasure Yachts and Offshore Racing Yachts rules. For example, in those rules the lowest fully annealed yield strength for alloy 5086 is given as 14 kpsi; and for 5083 as 18 kpsi. Though the newer alloys are not listed in the original versions of those rules, other sources provide their fully annealed yield strength. For 5383 it is 22 kpsi and for5059 it is 23 kpsi. Thus we can readily see the inherently conservative stance taken when using the fully annealed condition.
In addition to using the fully annealed condition, I generally use the lowest strength alloy (5086) for all scantling calculations unless it is KNOWN for certain that a specific higher strength alloy will be available and will be used. This approach allows use of 5086 throughout if that is what's available to the builder. Then if a higher as-welded-yield strength material is used, the vessel will simply be that much stronger, however in all cases the ABS rule will still have been satisfied.
This can result in heavier weight than is necessary if 5383 or 5059 plate happens to be available. Therefore in a weight-sensitive application, there is plenty of room for optimization of the structure to suit the higher fully annealed yield strength of 5383 or 5059 alloys.
THE SCANTLING CALCULATIONS
The original 1994 ABS Offshore Racing Yacht rule applies to sailing yachts up to 100 feet in length, but is now limited to those under 78 feet. The 2000 ABS Motor Pleasure Yachts Rule originally applied to all motor yachts up to 200 feet, but is now limited in scope to yachts from 79 feet to 200 feet (but without any changes to the actual rule). The 1975 ABS Aluminum Vessels rule (yes it is THAT old…!) applies to vessels from 100 feet to 500 feet in length.
Per these scope limitations, it seems most appropriate to consider the ABS ORY rule or the ABS MPY rule for our boats, according to the yacht type. Both rules address steel, aluminum, fiberglass, and plywood construction, and are therefore quite versatile in their application.
Within all of the ABS rules the method for determining scantlings is generally the same, as follows:
First a head pressure is calculated for each region of the boat, based on boat size and dimensions. Alloys are then selected, and the allowable yield strength for the chosen alloy is considered in all subsequent calcs.
Then a hull plate thickness is selected and verified per the rule, based on experience, boat size, usage, location, etc.
With a plate thickness chosen for each region, a long'l stringer spacing is then selected and verified based on what is necessary to support that plate thickness, and based on the location and the resulting head pressure. Then a frame spacing is selected and verified, based usually on what is practical in terms of attachment of the interior and the arrangement of interior spaces (typically double the long'l spacing, or thereabouts, but often more).
Once the long'l and transverse spacings have been chosen, the long'l stringer scantlings are selected, calculated and verified based on the location, plate thickness, and the maximum span between frames according to the prescribed minimum Section Modulus.
Then the transverse frame and deck beam scantlings are selected and calculated on the basis of being at least twice the depth of the long'l stringers, and verified against the prescribed minimum Section Modulus in the rule, which is calculated according to the local head pressure and the local span, and which considers the local plate thickness.
Per the ABS Motor Pleasure Yachts rule, a few simple limits apply: Aluminum plate must be at least 5/32 inch thickness as an absolute minimum. Based on vessel size, head pressure and plate location, greater minimum thicknesses may be prescribed. In some cases a credit might be available, based on the aspect ratio or the curvature of the unsupported plate region. In other cases, such as for tanks on commercial vessels, 1/4 inch aluminum plate is the minimum thickness used.
Also per the ABS MPY rule, the ratio of depth to thickness for any aluminum flat bar frame members (transverse or long'l) must not be greater than 12:1, or a rider bar or flange must be used. A flat bar flange is also limited to a 12:1 ratio (width to thickness). The depth to thickness ratio of web frames with flanges must not exceed 59:1.
In the ABS Offshore Racing Yachts rule and the ABS Aluminum Vessels rule, the depth to thickness ratio is calculated according to a factor based on material strength.
The ABS Motor Pleasure Yachts rule allows a region of plate adjacent to each frame member equal to 80 times the thickness of the plate to be included in the Section Modulus calculation, but limited to no more than half the frame member's local spacing on each side of the member. The ABS Offshore Racing Yachts rule allows a region of plate 100 times the plate thickness to be included in the frame and long'l stringer SM calcs.
My own preference is to limit this credit to 60 times the local plating thickness as a maximum. In any case, this credit assumes the local plate will be attached to the internal member by welding per the ABS calculated welding schedule.
The hull plating thickness required for ocean-worthy aluminum boats depends on the boat size and on the spacing of the internal framing.
For a skiff or pram, 1/8 inch aluminum plate is about as thin as can be welded easily. For larger boats, although the ABS rule allows the use of aluminum plate as thin as 5/32 inch, the minimum thickness I use is 3/16 inch regardless of boat size (except for skiffs and prams).
For the hull bottom and topsides, I consider it best to use a minimum of 1/4 inch thickness for boats of from 30 feet to around 45 feet, length on deck, then 5/16 inch up to around 55 feet, then 3/8 inch up to around 100 feet, etc. Keel sides are generally one size greater in thickness. Decks and houses are typically one size lesser in thickness.
These are only very general guidelines for minimum thickness. The ideal plate thickness depends on the as-welded strength of the alloy chosen, the type of boat it is used on, the location of the plate, and very much depends on the spacing of the internal framing.
If it is desired to increase plate thickness in each region in order to make use of a wider frame or longitudinal stiffener spacing, that election can be made and the ABS calcs will reflect the added strength imparted due to the greater plate thickness.
For aluminum, a few special considerations are imposed. Among them is to provide increased plate thickness in way of stress points such as next to the keel, above the propeller, around the rudder post, and in way of any other fittings that will have high stress (cleats, bitts, mast partners, chainplates, windlass, etc.).
The required dimensions of the internal transverse and longitudinal framing depends on their location and their span, but also depend on the thickness of plate, as noted above.
Using the minimum plate thicknesses outlined above, it is fairly typical for longitudinal stiffeners to be spaced from 12 inches to a maximum of around 18 inches, depending on the plate thickness, location, head pressure, etc. As noted above, it is more or less the case that transverse frames will be spaced approximately twice the long'l spacing. Transverse frames must always be twice the depth of the long'l stiffeners.
If the plate is of lesser thickness than outlined above, or if the service is more severe (such as for a high speed vessel's slamming loads), the stiffener spacing may well need to be less than the above spacings. If the plate is of greater thickness, then the stiffeners and frames may be farther apart.
One of the excellent benefits of using the ABS rule is that one can freely vary the sizes and spacing of the internal structure according to what is readily available, what is the most simple, and in order to accommodate different build strategies.
On small craft under around 70 feet, for the sake of simplicity it is my preference to use flat bar for internal framing. As a result, the 12:1 depth to thickness ratio limit automatically imposes a minimum thickness for each of the long's and frames.
Where it is necessary to exceed that aspect ratio in order to satisfy the minimum required Section Modulus, but a greater thickness is not desired, a rider bar will be used. For example, it is always necessary to use rider bars on floors, since they always exceed the aspect ratio limit. I will use rider bars or flanges for frames if needed, but generally not for long'l stringers.
Even though "T" shapes are superior structurally, for the sake of simplicity I prefer to use only flat bar for frames and for long'l stringers. On occasion for fast boats that must be as light as possible, I might specify a "T" shape for a long'l stringer, but not without considering alternate arrangements or closer spacing of flat bar, etc.
The available "T" bars are given in the link provided at the end of this article. Those "T" sizes are available from Alaskan Copper, and can therefore be considered , however they may not be readily available to all builders.
The Alaskan Copper stock list also shows what they refer to as a "6061 Hull Stiffener" which might also be considered for small boats, but I have not used them.
I have seen some applications of American Standard channel for long'l stringers, but not very often. Also, structural angle is used on occasion for long'l stringers, and might even be fairly common on some kinds of commercial boats.
I tend not to use any angle whatsoever, mainly because all transverse members are NC cut and will therefore be given a rider bar instead if necessary, and because for long'l stringers angle is not stable in bending since it tends to want to collapse with the open angle inward or outward. So even if strong, angle is not very "builder-friendly" especially where there is any amount of curvature. A structural "T" would be preferable in those locations.
For regions of little or no curvature, say possibly for deck stringers, structural angle might be used more often than I imagine, however even there it requires a rather large cutout in the transverse frame, which either adds complexity in order to weld in a patch, or will otherwise reduce the strength of the transverse frame.
I tend not to use "half pipe" sections anywhere, except possibly as an entirely external rub-strake, say at dock level as a bumper. It is tempting to make use of half pipe as an external "keel cooler" however for the sake of achieving minimum wetted surface, it is always preferable to locate any cooling channels entirely inside, say in place of an internal stiffener or as part of the keel box.
On the other hand, I do make extensive use of "full round" pipe sections, which are located at the intersection of hull and deck, and as trim on the top of bulwarks and around other edges. This is both an aesthetic choice, as well as a functional choice. At the intersection of the deck edge with the hull sides, a full round pipe adds considerable strength, and serves as a robust guard. At the top of a bulwark, a full round pipe provides a visual appeal, as well as a well-rounded edge to better hold paint, and to prevent chafe, etc.
In my use, these full rounds are usually specified as "pipe" rather than as "tube" since tube tends to be relatively much less common in North America, therefore tube is less easy to source. Schedule 40 pipe is the most common and the most readily available thickness. For aluminum structures I will sometimes specify schedule 80 pipe – primarily for ease of welding in the smaller diameters.
An advantage with pipe is that for any given nominal diameter, it always has the same O.D. regardless of wall thickness, so that threaded pipe fittings will work on any schedule thickness. Butt-weld ends, elbows and tees are available for pipe, and make for excellent terminations and transitions.
In the EU, and "down under" the situation is quite the opposite, with metric tube being commonly available, and imperial dimensioned pipe being rather difficult to source.
Many builders prefer to use a heavy wall pipe or tube or a solid round rod at all chines. I prefer not to do that because it complicates the welding considerably (double the number of welds along the chine…!). Instead, I prefer to locate the first long'l approximately 3 to 5 inches from the chine corner or plate edge, depending on plate thickness, on both sides of the chine. This stabilizes the weld-zone considerably, and vastly simplifies the assembly and the weld-up.
I have not made much use of "I" beams on boats, except as girders on larger craft, however possibly they can be used as compression posts or stanchions. In general I prefer to use pipe or tube for posts and stanchions.
Although they are available on occasion, I have not seen any use of "bulb flats" nor would I specify them. It is possible they are used to a greater extent in larger craft or in military craft.
REVIEW & VERIFICATION
My own review process starts with making sure that all structure is simple, practical, easy to build, and has good access for welding and maintenance. I will then verify the proposed structure per the applicable ABS rule. If regions of plate are in doubt, I will verify plate thickness and internal support according to plate theory per local edge fixity.
If the vessel is very slender, or is shallow in relation to its length, then the ABS Aluminum Vessels rule requires a global longitudinal strength analysis, based on a calculated minimum Section Modulus for the whole vessel. This would not ordinarily be a factor for vessels under around 60 to 80 feet. It is a seemingly complex calculation, but the ABS AV rule uses a fairly simple approach.
In general, it is preferable to locate transverse plate seams at 1/4 the span between frames - at the location of least stress. It is desirable in all cases to locate plate seams away from other stress points, such as hatch or house corners. In general, all house and hatch corners should have a generous radius. It is generally desirable to reinforce transverse plate seams using "sister" long'ls to span the seam, which also helps to minimize distortion during weld-up.
One of the best sources of supply for aluminum alloy shapes and plate on the US West Coast is Alaskan Copper (www.alaskancopper.com). And... any of the ABS rules mentioned here can be downloaded for free at www.eagle.org, in the Marine section.
Probably the best general guide to structure for metal boats is Tom Colvin's excellent book "Steel Boatbuilding" which, even though aimed at building cruising yachts in steel, is entirely applicable to building similar types of boats in aluminum. I find the examples of typical boat structure that Tom Colvin has offered to be very simple and extremely practical.
It should be rather evident from the above that I do not subscribe to the so-called "frameless" approach to metal boat building. For a discussion of the merits of the so-called "frameless" approach, please see my article on Metal Boat Framing. On the other hand, I do strongly favor the use of increased plate thickness in order to minimize the internal framing wherever it is practical to do so.
Further, there is no special requirement that dictates whether the frames or the plate will be erected first. For a discussion of various methods of metal boat construction, please see my article on Metal Boat Building Methods.
The approach of using increased plate thickness in order to limit internal framing is sometimes referred to as the "Strongall" system, which a company in France claims to have "invented." This approach definitely saves labor (and therefore costs) and vastly improves hull fairness. Despite those advantages, increased plate thickness will ordinarily result in a heavier structure and a higher materials cost.
This illustrates one of the biggest benefits of using the ABS rule, which is not a rigidly "prescribed" tabular rule (as is for example the Lloyds rule). In other words, when using a "calculated" approach to boat structure as is inherent in the ABS rules, we have always had the option to freely vary the internal structure and plate thickness, for example to increase plate thickness in order to realize benefits in terms of simplicity of a boat's internal structure.
Come to think of it, this was actually not "invented" in France after all… it is merely a practical approach to metal boat building!
Other Articles on Boat Structure
Metal Boats for Blue Water | Aluminum vs Steel | Steel Boats | Aluminum for Boats
Metal Boat Framing | Metal Boat Building Methods | Metal Boat Welding Sequence | Designing Metal Boat Structure
Composites for Boats | The Evolution of a Wooden Sailing Type
Kasten Marine Design, Inc. Professional Memberships
Member Royal Institution of Naval Architects Member Society of Naval Architects and Marine Engineers Member Society of Boat and Yacht Designers Member Metal Boat Society Member American Boat and Yacht Council
Please see our AVAILABLE BOAT PLANS web page.
Home | Intro | Custom Yacht Design | Stock Designs | Motor Yacht Gallery | Sailing Yacht Gallery | Prototypes Gallery | Plans List
Articles | Our CAD Design Stream | MAXSURF For Marine Design | News..! | Design Team | Site Map | Site Search | Contact Us
- All Web Site Graphics, Layout, and Written Content at this Domain Created by Michael Kasten.
- All Graphic and Written Materials at this Domain Copyright © 1989 - 2013 Michael Kasten.
- All Content Registered with US Library of Congress and US Copyright Office.
- Copyright Violations will be Prosecuted. All Rights Reserved.