While this repeats some of the information in my earlier post, this was a draft of an article which I wrote some years ago as an introduction to fiberglass technology.
A primer on FRP technology
FRP (Fiberglass Reinforced Plastic- the technical name for 'fiberglass' construction- sometimes also called GRP) had become the primary way that pleasure craft have been built since the late 1960’s. There are a lot of ways to build a FRP boat and a lot of variations on each method. The three most common are Monocoque, cored and framed. You often hear people use the term ‘Solid Glass Construction’. This is actually a very vague and not a terribly precise description of the structure of a FRP boat. As the term ‘Solid Glass’ hull or construction is typically used to mean a boat that does not have a cored hull. A non-cored hull can be monocoque (the skin takes all of the loads and distributes them), like many small boats today and larger early fiberglass hulls or framed as most modern boats are constructed today. A cored hull is a kind of sandwich with high strength laminate materials on both sides of the panel where they do the most good and a lighter weight center material. Pound for pound, a cored hull produces a stronger boat. Cored hulls can also be monocoque or framed.
Framing helps to stiffen a hull, distribute concentrated loads such as keel and rigging
loads, and reduce the panel size, which helps to limit the size of the damage caused in a catastrophic impact. Framing can be in a number of forms. Glassed in longitudinal (stringers) and athwartship frames (floors and ring frames) provide a light, strong and very durable solution.
Molded ‘force grids’ are another form of framing. In this case the manufacturer molds a set of athwartship and longitudinal frames as a single unit in a mold in much the same manner as the rest of the boat is molded. Once the hull has been laid up the grid is glued in place. The strength of the connection depends on the contact area of the flanges on the grid and the type of adhesive used to attach the grid. This is a very good way to build a production boat but is not quite as strong as a glassed in framing system.
Another popular way to build a boat is with a molded in ‘pan’. This is can be thought of as force grid with an inner liner spanning between the framing. This has many of the good traits of a force grid but has its own unique set of problems. For one it adds a lot of useless weight. It is harder to properly adhere in place, and most significantly it blocks access to most of the interior of the hull. Pans can make maintenance much harder to do as every surface is a finished surface and so it is harder to run wires and plumbing. Adding to the problem with pans is that many manufacturers install electrical and plumbing components before installing the pan making inspection and repair
of these items nearly impossible.
Glassed-in shelves, bulkheads, bunk flats, and other interior furnishings can often serve as a part of the framing system. These items are bonded in place with fiberglass strips referred to as ‘tabbing’. Tabbing can be continuous all sides (including the deck), continuous on the hull only, or occur in short sections. Continuous all sides greatly increases the strength of the boat but may not be necessary depending on how the boat was originally engineered. The strength of the tabbing is also dependent on its thickness, surface area and the materials used. When these elements are wood they can often rot at the bottom of the component where the tabbing traps moisture against the wood.
The strength of laminate (in either cored or non-cored panel) depends primarily on lay-up quality, kind of fibers used, number of laminations, and orientation of cloth. But also it depends of how carefully the laminate is handled and the ratio of resin to laminate. Glass and carbon fibers before they are laid up are quite brittle, and folding the dry laminate can break some of the fibers in the laminate. This weakens the material substantially. Historically, production manufacturers would cut multiple layers of each piece of laminate to be used in the manufacture of a particular boat and then fold the pieces and store them in a pile until they were needed. This of course created weakened lines
within the fabric. Most quality production builders avoid folding the laminate today.
When it comes to the actual fibers, there are a number of properties that are considered:
Strength in tension- (Tensile Strength) The point at which the fibers can be pulled apart,
Strength in compression- (Compressive Strength) the strength at which the fibers crush,
Elongation (deflection properties)- This is the amount that a material changes length for a given pull or push on the fiber. This is usually given as the Modulus of elasticity (E), which is the length in inches that a square inch of the material elongates or compresses per pound of force. When we deal with FRP there is often a different E for tension and compression. Since the resin is typically responsible for taking a large portion of the compressive loads but have almost no tensile strength, the focus is usually on the E (sub) t or the Modulus of elasticity in tension for the given fiber.
Orientation: The direction or directions that the fibers are oriented within the fabric. Also how the fabric is made. Flat fibers oriented the same direction (tows) and woven roving where the fabric is essentially straight are very strong ways to use fiber. Woven fiberglass cloth has a lot of kinks in the yarns and so are actually weaker and stretchier. Mat is not terribly strong because it uses short length fibers. Bi-axial and Tri-axial cloths use linear fibers either perpendicular or else oriented 120 degrees apart. This allows careful alignment of the fibers with the tensile stresses in the load mapping.
Abrasion resistance: The ability to withstand exposure to a rough surfaces once the resin and/or the gelcoat has worn through.
Laminate materials are chosen for their strengths and weaknesses in each of these properties, as well as, cost, of course. Because a fiber is low stretch it does not mean that it is also high strength, and just because a fiber has high tensile strength it does not mean it has high compressive strength. Resins have their own properties and, while they are far less important to the overall strength of the composite than the fibers in question, the choice of resin makes a very big difference in the ultimate strength of the part, as well as, its fatigue resistance.
The three most common resins are Polyester, Vinylester and Epoxy
. Polyester is a group of resins that can vary quite extensively in their properties. It is the least expensive and the most commonly used resin. It has poor ductility, impermeability and resistance to fatique as well as being very poor in developing secondary bonds. It is often modified to increase or decrease cure times. One iof the best features for production boat building is that polyester will not fully cure until deprived air. This allows muliple laminations with a laminating resin without sanding between laminations. The last lamination is a finishing resin which contains a waxy material that foats to the surface and seals out the air permitting a complete cure.
Vinylester is a family of vinyl modified polyesters. This recent material (since the 1990’s) is a wonderful material. It has excellent ductility and memory, great fatique resistance properties, and is easy to work with. Used heavily in the helmet industry it has come down greatly in price and is being used pretty extensively on even high volume boats.
has a whole range of extremely wonderful properties. It really shines where secondary bonds are important. Unfortuneately it is very expensive and harder to work with than the other resins and so is rarely used. It offers superior secondary bond adhesion (adhesion after the initial fiberglass has set up) and so is the preferred material for fiberglass repair
Looking at the individual fibers.
Carbon: Carbon has two very important characteristics, 1.Carbon has a comparatively high tensile strength but 2. an extremely high Modulus of Elasticity in tension and moderately high compressive E. This means that Carbon fiber composite parts have a lot of strength in bending but more significantly they can take big loads without much changing shape. It is this property that makes carbon so ideal for masts and other spars. It is also a reasonably light fiber. Carbon has some big negatives as well. Carbon is only moderately in resisting fatigue and so can breakdown in situations where it alternatively flexed and un-flexed. One characteristic that is often overlooked is that Carbon fiber conducts electricity and can be electolytically active (i.e. subject to electrolysis) (One popular theory on why Coyote lost her keel was that there was problems with the grounding of 24 volt generator and the carbon fiber attachment of the bulb keel bolting plate was weakened.) Carbon is also not very good in resisting abrasion. These properties makes it an ideal material for short lived race boat parts and light weight spars like windsurfers and spin poles but not so good for a cruising boat hulls or long life items.
Kevlar is one of my favorite materials. This is one very tough material. It has very good tensile strength properties (but not as great as Carbon or S glass). It also has a large E. Unlike carbon it has excellent fatigue resistance and abrasion resistance. It is extremely light and will actually float out of the resin. You must either vacuum bag kevlar or use a fabric with both glass and kevlar in it. You can’t sand a laminate with kevlar in it. Trust me I have tried. The kevlar balls up. The way I have dealt with repairs
over kevlar is to cut the kevlar strands with an Exacto and then finish with a layer of F.G. cloth. Kevlar is amazingly tough to cut or work with. If you drill though a Kevlar boat (Rugosa had a kevlar hull and deck) and you don’t use a sharp drill the kevlar will not cut and will wrap around the bit and drag the drill to a stop. To me it is an ideal material for the exterior laminates for boat hulls. Kevlar is not too great in compression, so it is best used in concert with S-glass, so that the S-glass can take help take compressive loads.
S glass is a type of fiberglass. There is a lot that distinguishes S glass from E glass, but basically, when glass fibers are made there are a variety of ways of doing it. All of the methods result in glass fibers that are not smooth on the surface when seen in a microscope. The roughness is actually small cracks in the surface of glass fiber. The fewer breaks the stronger the tensile strength of the fiber. Also the longer the fiber the fewer the un-restrained ends of fiberglass fiber and therefore the stronger the composite. The process that produces S-glass produces longer, less fractured fibers and then uses that fiber in fabrics that minimizes crimps in the fiber. S-Glass has really good tensile strength but does not come close to carbon or kevlar with regard to elongation. It is a good alternative for the interior of cored hulls where
E-glass is the run of the mill everyday fiberglass laminate. E glass is used in virtually all production boats and has reasonably good properties for most applications. It is the least specific specification and can vary very widely in quality. All early fiberglass boats were made of E-glass. E-glass can have especially poor fatigue qualities and only fair Tensile strength. It has terrible E properties in tension and only so-so E-properties in compression. In other words it is very flexible. While it is initially true this flexure has little to do with the bending strength of the material, in a material that is not very good in fatigue, flexure can be a significant problem.
Kevlar is harder to laminate than the other fibers. It is hard to cut and floats to the surface. It dulls cutting tools and is hard to tool. The key is to use sharp tools to cut the laminate vacuum bag the lamination and use glass mat buffer laminates. Both carbon fiber and Kevlar require Vinylester or epoxy
resins to get any real advantage out of them.
OLD VS NEW FIBERGLASS
One statement you see a lot is “Early boat builders did not know how strong fiberglass was and so made it very thick.” Horse Feathers! This is just plain bunk. The federal government had done a lot of research on Fiberglass and the information was widely available in the 1960’s. As a kid, I had literature on fiberglass that pretty clearly analyzed its properties. Guys like Carl Alberg, who was working for the government designing fiberglass ammo boxes when he was hired by the Pearsons to design the Triton, knew exactly what fiberglass would do. They knew that the e-glass of that era was pretty poor quality and was especially prone to flexing and to fatigue. They attempted to design fiberglass boats to be as stiff as wooden boats of the era. This took a lot of thickness since F.G. was very flexible compared to wood. This was especially true on a pound for pound basis. They also knew that if the boats were not as stiff as wood, there would be major fatigue problems. This put early designers in a bind. If they made the glass boats as thick as a wooden planked hull they would be impossibly heavy. If they did not, fatigue would condemn them to a short life. They mostly chose to compromise. By that I mean they chose to do boats that were not as stiff as the wooden boats they replaced but were heavier. Early glass interpretations of wooden boats were generally heavier and carried less ballast than their wooden counterparts. They were much stronger in bending but not as stiff. As fatigue took place some of these early glass boats became even more flexible which leads to more fatigue, which can lead to a significant reduction in strength.
Coring allowed the hulls to be made much thicker without the weight penalty. In calculating the stiffness of a section, the thickness is to the third power and so small gains in thickness result in big gains in stiffness. Coring allows a boat to be very stiff and strong and thereby reduces fatigue. Its not that coring comes without problems. The core is primarily subjected to horizontal sheer. To visualize Horizontal sheer, (Take a deck of cards and bend them. As you do you’ll feel the cards slide one over the other. That slippage is horizontal sheer.) The core material must be able to withstand the reversing horizontal sheer loadings without fatigue. That is what Balsa core does best. But balsa core can and does rot. It takes a higher density foam to equal the sheer strength and fatigue resistance of Balsa. That said, if you are building for durability, nothing beats medium density foam coring.
There is an oft-quoted statement floating around the internet “Cored laminates are stronger in flat panels, but are weaker when used with curved surfaces.” There is no scientific basis for that statement. When cored materials are applied to curved surfaces the core materials are designed with small stipes that allow the compound bending. When the core is properly vacuum-bagged into place, these stipes fill with resin and greatly increase bonding and the horizontal sheer of the panel. So, while cored laminates are stronger than solid panels on the flat, they are much stronger than solid panels when used on a curved surface. The author of that statement also has some dramatic photos of delamination problems on cored hulls but all of those photos appear to be low-density foam coring, which is almost never used in sailboat construction.
Mat vs. oriented fabric:
Mat (or chopped glass) does a number of things. First and foremost, almost all fabrics are directional. Mat and Chopped glass are not. Directional fabrics are weaker at bias angles that bisect the primary load directions. With good stress mapping you theoretically could use all directional material carefully oriented but because boats are subjected to loads from all different directions there needs to be an offsetting fiber orientation across the bias. Since mat has equal strength in all directions mat helps resist those loads that do not align with the direction of the directional materials. Mat also serves a more practical purpose. Course materials like woven roving, which have a lot of strength and which represent an easy way to build depth quickly have rough laminated textures. Due to this rough surface it is difficult to get a proper adhesion between course laminates without using too much resin. Mat is able to contort to the texture and make a good connection between the course laminates. Mat has another function as well. Resin shrinks as it cures and resins cure over very long periods, as much as years. If you put roving against gelcoat, the thicker resin in the course laminates shrinks proportionately to the thickness of the resin. This results in “print through” where the pattern of the fabric can be seen by sighting down the hull.
We are just now starting to understand the problems with non-oriented materials. In actual testing performed by the US Naval Academy (from a paper presented at the 2002 SNAME Chesapeake Bay Sailing Yacht Symposium), non-oriented fiber reinforcing fabrics were found to be the primary mode of failure in point impact situations. This paper outlined that Naval Academy cutters, which are used in training exercises, are subjected to frequent collisions, but the Academy cannot afford to take them out of usage for long repair
periods. As a result, impact resistance was very critical. In order to test the impact resistance a large pendulum with a massive weight was constructed. On the leading edge of the pendulum was a steel replica of the bow and stem fitting of a Naval Academy cutter. Test panels were constructed that matched both known (prior cutter lay-up schedule and J-24 topsides) and conjectural hull panels. The panels were aged and then tested warm (some resins lose strength when warm). The tests consisted of retracting the pendulum with a forklift and then releasing the restraint cable. The results were very dramatic.
To begin with. Solid hulls did far worse than cored hulls. In examining the panels after the collisions, the failures almost always occurred in the non-direction material being used and not in the core materials. The test sample that faired best used an oriented glass laminate, NO non-oriented materials, vinylester resin, and a high-density foam core. The pendulum never entered the outer laminate and microscopic analysis further destructive testing showed that core was still fully adhered to the skin and that the deformation was within the elastic (memory) properties of the core.
This is bad news for those with older heavier hulls. Through actual testing it has been known that these heavy solid hulls did not have the strength of newer lighter hulls but the failure mode was not completely understood. As mentioned above, it was generally believed that the issues were inferior resins and fibers, poorer handling of the materials, poor resin ratios, and the extensive use of accelerators and fillers. What is implied in the NA testing is that the problem may also lie in the extensive use of non-oriented fiber type laminates. These old heavier so-called solid glass hulls actually used an enormous proportion of non-oriented materials greatly reducing their impact resistance, stiffness, and tendency to resist fatigue.
Everything else being equal, twice the laminates take twice the time to abrade, but heavier cloths are not more abrasion resistant than lighter ones. Kevlar is enormously more abrasion resistant than any other laminate. The other factor is the force of the impact. A lighter boat hits with less force than a heavier boat so the rate of abrasion is greater on a heavier boat. On the other hand there is typically more material to resist this greater impact and abrasion. As far as I know resin has little bearing here.
If one had to design a boat solely to abrade for a day or two against rock it might be thick steel. If that was not your only criteria for designing a boat (in other words you were concerned about sailing ability and motion comfort), then it makes sense to build in FRP with outer layers of kevlar over a medium density foam core over more layers of S-glass and Kevlar.
Here more laminates is not necessarily better. Fiber type and fabric type is most crucial. Proper load distribution is crucial. This means reasonably small panel sizes, good fiber orientation and a bit of luck. Kevlar helps. Resins again have can have a major impact on performance. In the US Naval Academy testing mentioned above Vinylester Resin of a type used to build military and motorcycle crash helmets performed much better than less ductile resins. The high tech fibers, Carbon and Kevlar, need resins that can withstand higher tensile loads without developing small stress cracks. Epoxy
and Vinylester can deflect more without getting the microscopic fractures that are the beginning of the end for FRP.
Polyester is the cheapest and most common resin and as laid up is not impermeable to water. Polyesters vary widely in quality and performance. They are more prone to fatigue problems than other resins. One source of water penetration is the microscopic passages created as polyester fatigues. Early polyesters were particularly brittle and fatigue prone. This problem was further aggravated by the tendency by early boat builders to use accelerators and retardants depending on temperature and the nature of the operation. Another issue is with accuracy of the metering. Early boat builders used pretty imprecise methods to proportion resin. Today metering pumps
make precision metering a piece of cake, but back then mixing was more hit or miss. For example I installed an instrument through hull in a Triton and found a pocket of uncured un-reinforced resin probably a decade after the boat was built.
Vinylester resin does better than polyester so many better boat builders are now using it in the outer laminates and with high tech fibers. Epoxy seems reserved to custom builders and secondary bonds, because it is expense rather than some other flaw.
In any production boat there are several possible barriers to permeation as follows;
Gelcoat: Gelcoat is a thin molded laminate layer that is the first lay-up in a female mould. It consists of a dyed resin. Historically gelcoat was polyester resin, with an opaquing element and pigment that gave the boat its color. Gelcoat is typically sprayed or rolled on the mould. It has no reinforcing in it and so is brittle and should not be applied too thickly. Because it is applied as a continuous membrane and does not have reinforcing to wick moisture, it forms a barrier protecting inner laminations from water penetration. Because polyester is slightly porous, it is not a perfect impermeable membrane.
A recent improvement on polyester gelcoat is vinylester gelcoat. Vinylester is far less permeable than polyester and so water does not pass through a Vinylester Gelcoat as easily. To further improve the impermeability of the laminate, vinylester gelcoat is often combined with a vinylester veil-coat (the veil coat is the first reinforced laminate in the lay-up which occur between the gelcoat and the structural laminate. Since resins shrink after layup, but the reinforcing materials do not, the purpose of the veil-coat is to keep the pattern of the reinforcing weave from ‘printing through’ as the boat cures over a period of time.). Better manufacturers also use vinylester for the last lay-up as well in order to prevent bilge water from permeating the lay-up.
Often a barrier coat is added below the waterline to further seal the hull. Barrier coats are typically painted over the gelcoat. They are typically either epoxy for its combination of adhesion and impermeability or vinylester which is cheaper, slightly less permeable than epoxy and easer to work with.