Orient fibers so the load has somewhere to go
Generated from
content/lms/fabrication-and-composites/04-laminate-and-consolidate-the-part/02-orient-fibers-for-the-load-path.md; edit the source file, not this page.
Source path: content/lms/fabrication-and-composites/04-laminate-and-consolidate-the-part/02-orient-fibers-for-the-load-path.md
Course: Fabricate composite race-car parts with workshop discipline
Module: Laminate and consolidate without hiding defects
Estimated duration: 55 minutes
A composite part is not strong because it looks technical, uses carbon, or has enough resin in it. It is strong when the reinforcement is placed so the important loads can travel through the part instead of wandering through weak, unsupported resin. That is the skill in this lesson: before you wet out the cloth, before you bag it, before you cure it, you decide where the load is coming from, where it must go, and which fibres will carry it there.
The useful starting point is the basic definition of a composite. In the bonded material here, McBeath describes a composite as a matrix and a reinforcement working together to produce better properties than the separate ingredients. For your workbench, the matrix is the resin system and the reinforcement is the cloth or fibre. The resin gives the part shape and binds the reinforcement, but the lesson is not about adding more resin until the part feels reassuringly heavy. The lesson is about giving the reinforcement a job.
Think of the fibre direction as the road system inside the laminate. If the load has a direct road along the length of the fibres, the part can use the strength and stiffness you paid for. If the load must cross fibres, jump between badly chosen layers, or run through a region where the reinforcement has been spread and separated during lay-up, the laminate becomes much less like an engineered composite and much more like an expensive shell with weak traffic routing. McBeath is explicit on the reason unidirectional fabrics matter: they let the fibres be laid in exactly the orientation required by the component, so loads can be fed along the fibre lengths. He also gives the warning that makes this practical instead of decorative: if those fibres are misused, spread, or separated during lamination, the directional strength and stiffness you were trying to create can be significantly lost.
For an intermediate fabricator, fibre orientation is therefore not a cosmetic choice. It is a design decision made during manufacture. Unlike a metal sheet, where the broad material properties are mostly already present before you cut it, a composite laminate lets you create directional properties while you build the part. That is powerful and dangerous. Powerful because the same general family of materials can make a body panel, spoiler, aerofoil, duct, dashboard, chassis panel, or suspension link. Dangerous because the wrong schedule can make a part that looks finished but has no honest load path.
Your first habit is to name the job before naming the material. Do not start with carbon, glass, Kevlar, pre-preg, wet lay-up, or the fabric that happens to be on the shelf. Start with the question: what is this part being asked to resist? A body panel may mainly need shape, local support, and enough stiffness not to flap. A duct may need lightness, smooth compound curves, and resistance to mounting loads. A wing element or mounting may need to keep an aerodynamic surface in position under load rather than bending into a different angle. A beam-like bracket may have one face doing a compression job and the other face doing a tension or impact job. A safety structure may require energy absorption and loading capacity rather than a brittle stiffness number that looks impressive on paper.
Once you name the job, draw the path. Put the part on the bench or sketch it flat. Mark the places where load enters the part: fasteners, bonded edges, brackets, aero pressure, driver contact, ground strikes, or adjacent structure. Then mark where that load is supposed to leave: another mounting point, a rib, a sandwich core, a return flange, a chassis panel, or simply the rest of the skin if the skin is being used as a shell. Between those points, the fibres need continuity in the direction of the work. A ply that stops short of the actual loaded area may still make the surface look covered, but it does not complete the route. A beautifully wetted layer that points the wrong way may be clean workmanship, but it is not load-path workmanship.
This is also where you separate this lesson from the sibling lessons. Clean wet lay-up matters, pressure molding matters, vacuum-bagging with the correct bleeder and breather stack matters, and resin cure matters. Those lessons decide whether the laminate is consolidated and cured well enough to become structure. This lesson comes earlier. It decides whether the structure you are curing is arranged to do the right job in the first place.
The most useful sub-skill is load-path marking. Before cutting cloth, make a simple load map on the pattern or on masking tape laid over the tool. Use arrows for the main load directions and small labels for the kind of load you expect. Keep the language plain: bending here, pull between these mounts, aero load across this face, impact risk on this edge, support needed in this broad panel. You are not doing finite element analysis in the paddock. You are forcing yourself to explain why each ply exists.
The second sub-skill is choosing fibre direction for each ply. Unidirectional fabric is the clearest teaching tool because it makes the choice obvious. If the fibres run lengthwise, the laminate is being given a strong lengthwise job. If they run across the part, the cross direction is being served. If the part needs more than one direction, one ply cannot honestly do everything. You add plies with different orientations because the loads ask for them, not because symmetry looks tidy in a notebook. Woven fabrics still have direction, and the material list in the corpus includes woven cloths, unidirectional glass and carbon filaments, and Kevlar or graphite weaves. The practical lesson is the same: know which direction in the cloth is serving which direction in the part.
The third sub-skill is material-by-location thinking. Van Valkenburgh gives a simple beam example that is worth keeping in your head because it prevents generic carbon thinking. In a serious design, the optimum material can vary within the same part. In his example, the upper layers of a beam may want carbon for compression resistance while lower layers may want Kevlar for tensile and impact properties. In other cases, alternating carbon and Kevlar can exploit both materials. That means the question is not only which fabric is best. The better question is which layer, on which side, in which region, is doing which job.
For your own parts, that often means dividing the laminate into zones. The middle of a panel may need a different treatment from the mounting edge. A broad skin may need a core or ribs because large smooth surfaces are easy to form in laminate but may be too large to support themselves. A local hard point may need plies that carry load out into the surrounding skin instead of stacking thickness only under the fastener area. A long duct wall may need enough directional stiffness to keep its shape while still being formed around compound curves. The bonded material supports this zone-based view: laminates can be reinforced with foam or honeycomb sandwich panels, integral ribs, or stiffeners when surfaces are too large to support themselves, and top-level cars use carbon or aramid over honeycomb in body panels, chassis, and related structures.
The fourth sub-skill is handling the reinforcement without destroying the orientation you selected. This is most obvious with unidirectional fabric. A UD ply gives you a clean, direct load route only while the fibres remain where you placed them. If you drag the brush, pull the cloth crooked, stretch it around a corner until the fibres fan apart, or lift and reset it carelessly, the schedule on paper no longer matches the laminate on the tool. The visible problem is spread or separated fibres. The structural problem is loss of the directional strength and stiffness the ply was supposed to provide. On the bench, that means you slow down where the fibre direction matters most. You support the cloth when moving it, cut it cleanly, transfer it without pulling the fibre bundle apart, and wet it in a way that preserves the line of the fibres.
The fifth sub-skill is not overbuying stiffness at the cost of toughness. McBeath warns that ultra-high modulus carbon can engineer very high stiffness into laminates, but can also yield very brittle components when used over-zealously where the part needs toughness and forgiveness. That warning belongs in this lesson because orientation is not only about making the stiffest possible part. It is about making a part that sends the real loads through suitable material. A wing support that must hold angle under aerodynamic load has one job. An impact structure that must absorb energy progressively has another. A chassis panel that must contribute to torsional stiffness without approaching ultimate strength in normal racing stress has another. Do not let one attractive property become the entire design.
The sixth sub-skill is checking the part against reality. The bonded corpus repeatedly pushes away from overconfidence. Van Valkenburgh says the practical route is often to modify what exists, or build and test, instead of trying to design the ultimate suspension from advanced theory. McBeath notes that even in Formula 1, much composite design work has remained empirical, based on previous experience and knowledge rather than only computerized structural analysis. In aerodynamics, McBeath gives the same development warning from another direction: what works on one car may not work on another apparently similar car, and trial and error are part of the process. For your laminate, that means the schedule is a hypothesis until the part is checked under the kind of load it will see.
Your practical check can be simple, but it must be intentional. If the part is a mounting or aero support, check deflection under a repeatable static load. If the part is a broad panel, support it as it will be supported on the car and compare how it behaves across the unsupported span. If it is a beam-like bracket, look at which face is being compressed and which face is being pulled as you load it. If it is an impact-prone piece, do not judge it only by bench stiffness. If it is a part that lives near heat or vibration, remember the rear wing mounting failures described in the corpus, where failures were attributed to vibration from changes in natural frequency, heat from exhausts, and flexing behavior that drew regulatory attention. The lesson is not that you can diagnose all those mechanisms from a garage coupon. The lesson is that load path, stiffness, heat, vibration, deflection, and legality are connected in real race parts.
Good orientation work has recognizable calibration cues. The first cue is that you can explain every ply without reaching for vague words. If the best explanation is that it adds strength, you have not gone far enough. Strength in which direction, between which features, for which load? The second cue is visual: the fibres that matter remain aligned after wet-out and consolidation. The third cue is mechanical: the part resists the load you designed for more directly than the earlier version, without needing unnecessary mass everywhere. The fourth cue is developmental: your test results change in the direction the load map predicted. If you add lengthwise fibres to a beam-like part, lengthwise bending behavior should improve. If you add support through a core, rib, or stiffener to a large surface, the unsupported skin should stop acting like the only structure. If you change only the appearance and nothing measurable changes, you probably decorated the laminate rather than engineered it.
Bad orientation work also has recognizable cues. A part that is thick but limp in the loaded direction has material but no path. A part that is stiff on the bench but brittle in service may have chased modulus where toughness was required. A part that moves enough under aero load to change its function may be too flexible even if it does not break. A part that cracks near a mount may have accepted the load locally but failed to distribute it into the surrounding laminate. A part that varies wildly from one build to the next may have a handling process that disturbs fibre orientation, especially with unidirectional materials.
The intermediate-level standard is not perfection. It is conscious authorship. You should be able to lay the pattern on the bench and say: this ply carries the main span load, this ply spreads the mounting load, this region gets a different material because it needs impact or tensile behavior, this large surface is supported by core or rib because the skin alone is too large, and this test will tell me whether the schedule did what I claimed. That is what it means to orient fibres so the load has somewhere to go.
A final caution: regulations and budget are not afterthoughts. McBeath closes the composite discussion by saying the types of composite and the applications ultimately depend on the rules for your category and the budget available. In club racing and HPDE-adjacent fabrication, that is not a small note. A clever flexible aero part may be illegal. A costly material may be unnecessary if the part only needs a clean GFRP wet lay-up with sensible reinforcement. A high-end process may be outside your capability if you cannot cure it evenly or verify it honestly. The right laminate is the one whose fibres, materials, process, test, and rules all agree with the job.
Worked example: rear wing mounting that must hold angle, not just survive
A rear wing mount is a clean example because failure is not limited to a broken part. The mount can also fail the job by bending enough that the aerofoil no longer sits at the intended angle. The bonded corpus describes rear wing mounting failures at the start of the 1998 and 1999 Formula 1 seasons, with causes variously attributed to vibration from changes in natural frequency, heat from engine exhausts, and attempts to make mountings flex backward at high speed so the aerofoils adopted a lower angle of attack and reduced drag. The regulatory response was a static load deflection test.
For your own fabrication, treat that as a load-path lesson. The mount has to carry aerodynamic load from the wing into the car while controlling deflection. If you only wrap carbon around the visible bracket shape, you may create thickness without a path. First mark where the wing load enters the mounting. Then mark where the load exits into the chassis, endplate, support, or bonded structure. The important plies must run between those areas, not merely around them. If the bracket is tall and loaded fore-aft, the fibres that resist that bending must be placed where they carry the bending load. If the surrounding skin is part of the support, plies must feed load into that skin over enough area for the skin, core, rib, or stiffener to participate.
The check is also borrowed from the example. Do not ask only whether the mount looks solid. Put a repeatable load on it in the direction that tries to change the wing angle and measure deflection. Then inspect whether the load is being carried by the intended laminate route or concentrated at one local edge. If the part survives but moves enough to change the aerodynamic function, the load path is incomplete for the job. If the part becomes stiffer only after you add mass everywhere, your next development step is to improve fibre direction and support placement rather than simply making the whole bracket heavier.
Worked example: a simple beam with different jobs on the two faces
Van Valkenburgh gives a useful beam example: the upper layers may want carbon for compression resistance, while the lower layers may want Kevlar for tensile and impact properties. That example is small enough to carry into the workshop.
Imagine a simple composite support that behaves like a beam between two points. When the part bends under load, the two faces do not have the same job. One face is asked to resist compression while the other is asked to resist tension and possibly impact damage. A one-material, one-direction laminate may still work, but it may not be the best use of the materials. If the upper face is doing the compression work, carbon placed there along the span has a clear reason to exist. If the lower face is exposed to tension and impact risk, Kevlar or a carbon-Kevlar combination may have a clear reason to exist there. If both properties are useful through the thickness, alternating layers or hybrid fabrics may be justified.
The important habit is not memorizing that every beam must use that exact schedule. The important habit is location-specific thinking. Ask which side of the part is doing which work. Ask whether the outer layers are where stiffness or impact resistance is needed. Ask whether a material is being used because it is fashionable or because its properties match the load on that face. A serious laminate plan can vary across the part and through the thickness. That is the difference between building a composite shape and engineering a composite structure.
Worked example: broad body panel, duct, or dashboard skin that cannot support itself
Laminates are attractive for bodywork, ducting, dashboards, spoilers, and aerofoils because they can form large, smooth, compound shapes. Van Valkenburgh notes that when surfaces are too large to support themselves, they can be reinforced with sandwich panels of foam or honeycomb or with integral ribs or stiffeners. McBeath also points to carbon or aramid over honeycomb in high-level racing body panels and chassis.
The load-path mistake on a broad panel is assuming that a smooth skin is automatically a structure. A wide panel may look complete after lay-up, but if the load path is only a thin skin spanning too far, the part can flap, oil-can, crack at mounts, or feel dead in one direction and flimsy in another. The fix is not always more plies over the entire surface. Often the fix is to decide where the panel needs support and then give the load a route into that support.
For a duct wall, that may mean fibres running along the length where the duct is trying to sag or twist, plus a rib or local sandwich region where the span is too large. For a dashboard or body panel, it may mean using the laminate skin for shape while using foam, honeycomb, ribs, or stiffeners to keep large unsupported areas from carrying the whole job alone. The point is that orientation and support type are linked. A core without fibres that feed load into it is not fully used. Fibres without a support strategy across a large surface may be asking the skin to do too much. A good panel plan makes the skin, fibre direction, and support features agree.
Common mistakes
Mistake one: choosing material before choosing the load path. The bad version starts with carbon because carbon sounds like race car engineering. The good version starts with the part job, the load entry points, the supports, the rules, the budget, and the process you can actually execute. Only then do you choose glass, carbon, Kevlar, hybrid cloth, wet lay-up, pre-preg, core, ribs, or stiffeners.
Mistake two: adding thickness instead of direction. A thicker laminate can still be weak or flexible in the wrong direction if the reinforcement does not carry the load where the part needs it. The good version uses each ply for a named direction or zone. If you cannot say why a ply is where it is, you do not yet have a schedule; you have accumulation.
Mistake three: damaging unidirectional fibres during lay-up. UD fabric is valuable because it can place fibres exactly where the component needs them, but it is also easy to misuse. The bad version spreads or separates fibres while transferring, wetting, or shaping the ply, then assumes the original paper schedule still applies. The good version handles the material gently enough that the finished laminate still shows the intended fibre line.
Mistake four: making a stiff part where a tough part was required. Ultra-high modulus carbon can add very high stiffness, but McBeath warns that over-zealous use can create brittle components when toughness and forgiveness are needed. The good version distinguishes a stiffness problem from an impact or energy-absorption problem. A part near impact risk, a safety structure, or a load case with uncertain angles may need more than the stiffest possible fibre choice.
Mistake five: treating broad skins as self-supporting structure. Laminates form large smooth shapes easily, but large surfaces may need foam, honeycomb, ribs, or stiffeners. The bad version keeps adding skin plies until the panel feels acceptable in your hands. The good version asks where the span is unsupported and creates a support path so the panel is not relying on skin thickness alone.
Mistake six: calling the first finished part done. The bonded sources are clear that trial, empirical knowledge, and build-test learning remain part of motorsport development. The bad version finishes the laminate and assumes the job is complete because the surface cured. The good version defines a check before lay-up: deflection under load, comparison to the previous panel, inspection for fibre disturbance, or another repeatable test tied to the load map.
Drill: the load-path map and two-coupon check
Use this drill at your next fabrication session before making a real part or repair. The count is one part pattern and two small coupons. The duration is about 45 minutes before lay-up plus whatever cure time your resin system requires. The success criterion is that you can explain the job of each ply before wet-out and then see a measurable or feelable difference between the two coupons after cure.
Step one: choose a simple part or repair area. Good candidates are a small bracket, duct tab, body-panel support, splitter edge patch, dashboard mount, or non-critical practice panel. Do not choose a primary safety structure for this drill.
Step two: draw the load map. On paper or tape, mark where load enters and exits. Use arrows for the main direction. Add one sentence naming the job: resisting bending between two mounts, supporting a broad panel span, feeding a mounting load into a skin, or keeping an aero-related surface from deflecting.
Step three: write a two- or three-ply schedule in plain language. Do not write only material names. Write what each ply does. For example: this ply runs along the span to carry the main bending load; this ply spreads the mount load into the surrounding skin; this local reinforcement supports the broad unsupported area with the core or rib.
Step four: make two coupons from scrap or practice material. Coupon A should follow your chosen fibre direction for the main load. Coupon B should deliberately rotate the main fibre direction away from that load or omit the support feature you think matters. Keep resin system and size as similar as practical so the difference you feel is mostly orientation and support.
Step five: after cure, load both coupons by hand or with a simple repeatable weight in the direction you mapped. You are not producing certified data. You are training your eye and hands. Coupon A should better resist the load it was designed for. If it does not, your map may be wrong, the ply may have moved during lay-up, the support feature may be misplaced, or the load case may not be what you thought.
Step six: record the result. Keep the map, schedule, and a note on what changed. This is how empirical composite work becomes disciplined rather than random. You are building the habit that McBeath and Van Valkenburgh both point toward: practical construction informed by testing, previous knowledge, and honest feedback from the part.
When this principle breaks down
The principle does not break down because load paths stop mattering. It breaks down when you pretend a simple load-path sketch is enough for a part whose consequences exceed your ability to verify it. Race car impact structures, monocoques, suspension links, and heavily loaded aero supports can involve crash loading, oblique impacts, heat, vibration, natural frequency, rule compliance, and stiffness targets that are not safely resolved by workshop intuition alone.
The corpus gives several boundaries. CART and FIA impact-structure rules are described as continuously reviewed, with improving standards for impact absorption and loading capacity. Indycar chassis are discussed in relation to oblique collisions with concrete oval walls. Formula 1 composite use extends into chassis, suspension links, bodywork, aerofoils, and wind tunnel models. Rear wing mounting failures show how vibration, heat, deflection, and rules can converge. These examples should make you more serious, not more timid.
For HPDE and club-racing fabrication, the practical boundary is this: use load-path thinking on every composite part, but do not let it license unsupported safety-critical design. For non-critical bodywork, ducts, dashboards, small supports, and practice laminates, this lesson gives you a strong working method. For primary structures, crash structures, suspension links, or aero devices whose movement can affect legality or control, the method is only the beginning. You need rules review, appropriate engineering, suitable materials and process control, and testing that matches the risk.
Author Review
No quiz questions are attached to this lesson.
Sources
| # | Document | Chunk | Pages | Score | Collection |
|---|---|---|---|---|---|
| 1 | Competition Car Composites Simon McBeath | 33166f0f-e752-e86b-241d-4a2c998ac3c2 | 176 | 1 | uio_books_raw_v1 |
| 2 | Race Car Engineering Mechanics Paul Van Valkenburgh | ca7a3241-be1f-1f6f-b111-5291d7865790 | 96 | 1 | uio_books_raw_v1 |
| 3 | Race Car Engineering Mechanics Paul Van Valkenburgh | caef1668-4a67-bed6-3e4b-5c7daa3bb656 | 95 | 1 | uio_books_raw_v1 |
| 4 | Competition Car Composites Simon McBeath | 4cd165c8-25b6-009a-f4b5-4fae9a62b8dc | 12 | 1 | uio_books_raw_v1 |
| 5 | Competition Car Composites Simon McBeath | a0cc1d08-7515-9bbc-fe01-3d5ebc6719bb | 11 | 1 | uio_books_raw_v1 |
| 6 | Competition Car Composites Simon McBeath | 2b41f8db-cdd5-e584-d3e8-3cb469eea226 | 198 | 1 | uio_books_raw_v1 |
| 7 | Competition Car Composites Simon McBeath | 781f8145-6150-097b-9c36-0cf693583e67 | 202 | 1 | uio_books_raw_v1 |
| 8 | Race Car Engineering Mechanics Paul Van Valkenburgh | 41eb93f1-5aee-0fc7-9bd3-fdbe776a3459 | 21 | 1 | uio_books_raw_v1 |
| 9 | Competition Car Composites Simon McBeath | 629cf934-5b41-0aa0-eb70-cec1d94b0bbb | 171 | 1 | uio_books_raw_v1 |
| 10 | Competition Car Aerodynamics 3rd Edition McBeath Simon | c7d0125c-8080-dbcc-df83-3b96d0b84bab | 477 | 1 | uio_books_raw_v1 |
| 11 | Competition Car Composites Simon McBeath | 237c1c01-041e-d102-c244-155ba8d3fbb6 | 8 | 1 | uio_books_raw_v1 |