Mark the safety-critical boundary before the build

Generated from content/lms/fabrication-and-composites/01-choose-safe-fabrication-targets/04-mark-the-safety-critical-boundary.md; edit the source file, not this page.

Source path: content/lms/fabrication-and-composites/01-choose-safe-fabrication-targets/04-mark-the-safety-critical-boundary.md

Course: Fabricate composite race-car parts with workshop discipline

Module: Choose fabrication jobs that fit your tools and risk

Estimated duration: 45 minutes

This lesson is not about whether carbon, glass, or aramid is exciting. The sibling lessons handle material choice and part requirements. Your job here is narrower and more important: before you cut cloth, order resin, weld a bracket, drill a pickup, or copy a professional-looking part, you mark the line between a reasonable fabrication target and a safety-critical structure.

The principle is simple: process access is not design authority. The bonded corpus supports that home builders can work with composites, including glass fibre and, with suitable methods, carbon and aramid. It also supports that professional racing has used composites in far more serious places, including all-composite Formula 1 chassis and later Formula 3000 and Formula 3 chassis. Those two facts do not mean the same thing. A method can be available to a home workshop while a particular part remains outside a home workshop's safe design boundary.

For an intermediate driver-fabricator, the boundary is not marked by material alone. A glass part can be safety-critical if it locates a suspension pickup. A carbon part can be a reasonable practice target if it is a non-structural cover. A steel tab can be more dangerous than an elaborate composite duct if that tab changes where a wheel points under load. So you do not ask first whether you can make the part. You ask what job the part performs when the car is loaded, rolling, braking, cornering, and vibrating.

A part belongs on the safety-critical side when its failure or distortion can change wheel location, wheel angle, suspension geometry, chassis load path, rear-frame integrity, or the ability of the car to keep the tire doing the job the suspension was designed to protect. Staniforth's suspension discussion gives the mechanism. Racing suspension aims, in broad terms, to keep the wheel and tyre correctly presented to the road despite squat, nose-dive, bumps, and roll, while keeping roll centre and track behavior under control. That makes the small fabrication around the suspension much larger in consequence than it looks on the bench. The part may be only a bracket, clevis, tab, insert, bonded panel edge, or local reinforcement, but if it helps define wheel motion or carries a suspension force into the chassis, it has crossed the boundary.

The rear of the car is the cleanest warning example in the corpus. Staniforth describes rear-steer as the ability of the rear wheels to alter toe angle and effectively steer the back of the car. He then notes that as slick tire grip grew, the forces became enormous enough to bend rear structures, crack chassis and gearbox casings, tear out mounting points, and distort complete rear frames. That is the lesson. Do not judge safety-criticality by part size. Judge it by what the part is allowed to move, restrain, or tear loose.

This is why your first fabrication drawing needs a boundary mark. Before you choose the cloth, resin, layup, tube wall, adhesive, or fastener pattern, write the part's job in plain language. Then classify the job as green, amber, or red. Green means the part is a reasonable fabrication target because it is cosmetic, reversible, non-structural, and not responsible for wheel position or chassis load transfer. Amber means the part may be buildable, but it carries air load, supports bodywork or an aerofoil, sits close to a load path, or could create a hazard if it fails. Red means the part locates suspension, reacts suspension or rear-steer forces, forms part of a chassis or monocoque, carries load into a bulkhead or rear frame, or changes geometry. Red is not a challenge. Red is a stop sign until qualified design, analysis, fabrication, and inspection are in place.

The composite corpus supports this split. McBeath notes that competition cars commonly use composite components and that bodywork and aerofoils are peripheral places where composite use continued even when categories legislated against carbon chassis on cost grounds. That does not make aero casual. It means peripheral composite parts are different from the chassis load path. A body panel, duct, or fairing may be a good learning target when it is not structural. A composite chassis, suspension bulkhead, or pickup-bearing panel is a different project, even if both involve resin and cloth.

Carroll Smith's chassis collection adds the professional-humility reason. Racing chassis, suspension, and aerodynamic development are technical disciplines with little public codified history, tight deadlines, specialist knowledge, and guarded applications of immutable physics. That does not mean a club racer cannot learn. It means you do not let enthusiasm stand in for load cases, geometry control, and verification. Your boundary mark is a discipline tool. It forces you to separate the parts you can responsibly practice on from the parts where ignorance can put the car, driver, and nearby workers at risk.

The skill has five sub-skills.

First, name the part by job rather than by material. Do not write carbon splitter bracket, fiberglass panel, aluminum tab, or 3D-printed insert as the first description. Write what the part does. It covers a gap. It routes air. It holds a body panel. It supports an aerofoil. It locates a suspension link. It ties a suspension load into a frame. It forms part of a monocoque center section. Material comes later. The job tells you where the safety boundary is.

Second, trace the load path. Ask where the force enters, where it leaves, and what moves if the part bends. For a body cover, the answer may be panel fasteners and airflow pressure. For an aerofoil support, the answer includes the aerodynamic load path into the body or frame. For a suspension pickup, the answer includes tire force, link force, chassis reaction, and geometry. If you cannot describe the load path without hand waving, the part cannot be green.

Third, protect geometry. Any fabrication that changes camber behavior, toe behavior, track, roll-centre behavior, or wheel control belongs in red unless it has real engineering behind it. Staniforth's discussion of suspension development is full of geometry consequences: wheels leaning in roll, roll centre changes, rear steer, jacking, toe change, and unwanted movements. You do not need to solve all of chassis design to mark the boundary. You only need to recognize when your part is no longer just a part and has become a geometry decision.

Fourth, separate availability from capability. McBeath's professional chapter is encouraging because it shows composite technology filtering down from upper motorsport to the home workshop. It is also a warning. The professionals can do quite a few things for which the home constructor usually lacks facilities. When a technology filters down, the material and process may become available before the design control, test equipment, inspection culture, and failure data become available. Your boundary mark keeps those two facts apart.

Fifth, delay weight optimization until after the safety boundary is understood. Staniforth describes racing suspension parts as needing to be no heavier than necessary for the forces involved. That is not a license to guess light. It is a sequence. First understand the forces and the job. Then design the part to handle them without unwanted movement. Only then chase weight. If you start with weight before you know whether the part controls a wheel or holds a rear frame together, you have reversed the order.

Use this working procedure on every candidate fabrication target.

Step one: write the requirement in one sentence. The sentence must use an action verb. This panel closes the rear wheel opening. This bracket holds a removable duct. This bonded insert carries the lower wishbone pickup. This floor extension supports an aerofoil stay. This repaired panel edge ties into the monocoque bulkhead. If the sentence contains locates, reacts, carries, supports, ties, or controls, slow down and investigate.

Step two: draw the nearest load path, even roughly. Put the tire, suspension link, chassis or frame member, bodywork, and airflow load on the sketch when they apply. The sketch is not art. It is a lie detector. If your part sits between tire force and chassis reaction, red. If it sits between aero load and a mount that could tear away, amber at minimum. If it is simply a cover over an existing structure and can fail without moving a wheel, changing a geometry point, or compromising a load path, it may be green.

Step three: ask the failure question. If this part cracks, delaminates, bends, debonds, or pulls out, what happens first? A green failure is annoying, visible, and non-geometric. An amber failure may remove a panel, reduce aero function, or require immediate inspection, but it should not decide where the wheel points. A red failure can change toe, camber, track, suspension pickup location, chassis integrity, or rear-frame behavior. If the honest answer is that you do not know, the classification is red until you know more.

Step four: mark the drawing. Do not leave the decision in your head. Write GREEN, AMBER, or RED on the sketch or work order, plus one sentence explaining why. A good explanation sounds specific. Green because this is a removable non-structural cover over existing bodywork. Amber because this supports an aerofoil and must be inspected after each session. Red because this bonded insert carries a suspension load into the chassis. The written mark prevents material excitement from erasing the safety decision later.

Step five: choose the next action from the classification. Green parts can become practice builds, with normal workmanship and inspection. Amber parts need conservative design, visible attachment, inspection access, and a plan for what you will do if the part loosens or cracks. Red parts need qualified engineering, proven design patterns, appropriate fabrication controls, and inspection beyond a casual garage check. If those are not available, the correct fabrication decision is to choose a different target.

The calibration cue for this skill is not that you say no to everything. The cue is that your yes gets more precise. Early in the learning curve, you may call a part safe because it looks small or because another car has one. Better work sounds different. You can say what the part does, what load it sees, what it cannot be allowed to move, what happens if it fails, and why the build belongs on one side of the boundary.

You are improving when you start finding red flags before you buy material. You notice that a simple bracket is actually a suspension pickup. You notice that a composite repair is in a bulkhead area rather than on a loose body panel. You notice that an aero support is not just a pretty strut but a load path into thin bodywork. You notice that copying a professional composite chassis detail without the professional process is not a shortcut. These are not signs of timidity. They are signs that you are reading the car as a system.

A useful instructor or experienced builder would hear your classification and ask what happens under load. If your answer is only that the part is thick, carbon, strong, or used on race cars, you have not marked the boundary. If your answer explains wheel control, chassis reaction, rear-steer risk, composite process limits, and the consequence of failure, you are doing the skill.

The most important recovery rule is this: if a part changes classification during the build, stop and re-scope. Many unsafe fabrication projects begin as green in the driver's imagination and become red once the real attachment points appear. You remove a panel and discover it is tied into a bulkhead. You mock up a duct and realize the mount shares fasteners with a suspension-adjacent bracket. You copy an aerofoil support and realize the load path enters a thin unsupported panel. Do not solve that discovery with extra layers and optimism. Reclassify the part and choose the next action from the new classification.

This lesson cross-references the material-choice lessons, but it comes before them in practice. Define the composite before you spec carbon, start with GFRP before chasing carbon, and match material to the part's job only after the boundary is marked. The boundary tells you whether the job is appropriate for you to fabricate at all. Material choice only helps after that question has been answered.

Worked example: the hillclimb composite temptation

McBeath's professional chapter gives a useful intermediate-level trap. UK Speed Hillclimb and Sprint cars such as the Scott Megapin 5, the Brytec, and the OMS CF34/94 show composite technology filtering downward from upper motorsport. Two of those examples are described as home-constructed specials, while the OMS is described as a production racer. The optimistic lesson is that serious-looking composite work is not reserved forever for Formula 1. The safety-boundary lesson is that serious-looking composite work must still be split by job.

Imagine you want to build composite pieces for a club car after seeing these examples. A removable body blister, duct cover, or fairing that does not locate suspension and does not carry chassis load may be a green or amber target depending on attachment and consequence. An aerofoil mount is at least amber because the corpus names aerofoils as peripheral composite components, but peripheral still means it creates and transmits load. A composite chassis panel, suspension bulkhead, or bonded insert carrying a pickup is red. The fact that all of these can involve cloth and resin is irrelevant. The job is what sets the boundary.

The correct takeaway is not to avoid composites. It is to avoid using the existence of home-constructed composite competition cars as permission to build every composite part. Technology filtering down gives you more possible practice targets. It does not erase the difference between a bodywork target and a chassis-load target.

Worked example: the small rear pickup that is not small

The rear-steer discussion in Staniforth is the strongest failure example in the packet. Rear wheels can alter toe angle and effectively steer the rear of the car if the geometry and stiffness are not carefully controlled. As slick tire grip grew, the forces at the back of the car became large enough to bend parts, crack chassis and gearbox casings, tear out mounting points, and distort rear frames.

Now apply that to a garage fabrication idea: a small replacement tab for a rear link, a local composite reinforcement around a pickup, or a bracket that seems too simple to be dangerous. The boundary mark is red because the part is small only in size, not in consequence. It can alter toe, permit rear steer, or move the load path into structure that was not designed for it. Extra layers, a thicker weld, or a stiffer-looking bracket do not make the decision green. They may simply move the failure somewhere harder to see.

The good version of this project starts with a stop. You identify the pickup as safety-critical, document the load-path reason, and either use a proven replacement procedure with qualified inspection or choose a different fabrication target. The lesson is not that rear suspension is mysterious. The lesson is that any part allowed to steer the rear of the car through deflection has crossed the boundary before fabrication begins.

Worked example: production-car strut surroundings

Staniforth notes that MacPherson struts equipped the majority of contemporary production cars and gives the Lancia Delta World Championship rally car as a Group A modified production-car example. That matters because many Tracky drivers start with production-based cars. Familiarity can make suspension-adjacent fabrication feel ordinary.

Suppose you are working around a strut tower. A removable cosmetic cover, an inspection-friendly splash panel, or a non-structural duct near the area may be a reasonable fabrication target if it does not affect the strut mount, wheel location, or chassis load path. But modifying the strut top, reinforcing a cracked tower without proper design control, or adding a bracket that shares the suspension load path is red. The production origin of the layout does not make it casual. The strut is part of how the wheel is located and loaded.

The practical cue is whether your work can change the suspension point or the structure carrying it. If yes, it belongs on the safety-critical side. If no, and the failure is visible and non-geometric, it may remain a practice target.

Common mistakes

Mistake one is the material glamour mistake. You see carbon and treat it as automatically superior. Good work separates material from job. Carbon in a non-structural cover may be a practice target. Any material in a suspension pickup is safety-critical.

Mistake two is the small-bracket mistake. You judge a part by its bench size rather than the force it reacts. Good work asks whether the part can change wheel angle, toe, track, roll behavior, or chassis load. A small rear-link tab can be more safety-critical than a large body panel.

Mistake three is the professional-copy mistake. You see an all-composite chassis, professional aerofoil support, or upper-motorsport detail and copy the appearance without the facilities, design history, or inspection controls. Good work treats professional examples as proof that a technology exists, not proof that your version is safe.

Mistake four is the peripheral-drift mistake. You begin with bodywork and drift into structure. Good work reclassifies the project the moment the part carries load into a frame, bulkhead, rear structure, or suspension-adjacent mount.

Mistake five is the weight-first mistake. You make the part light before you know the forces. Good work follows the order implied by the suspension-design material: understand the forces and movement constraints first, then remove unnecessary weight.

Mistake six is the unknown-failure mistake. You cannot state what happens if the part cracks, bends, debonds, or pulls out, but you keep building anyway. Good work treats unknown consequence as red until the consequence is understood.

Drill: six-part boundary board

Do this drill before your next fabrication project, not after you have bought material. The count is six parts and the time box is 45 minutes. Choose two obvious non-structural targets, two ambiguous bodywork or aero-adjacent targets, and two suspension, chassis, or mounting-point-adjacent targets from your own car or build plan.

For each part, write a one-sentence job description, sketch the nearest load path, answer the failure question, and mark the part green, amber, or red. Use the same standard every time. Green means failure is visible, reversible, and non-geometric. Amber means the part carries some load or sits near a meaningful load path but does not define wheel location or chassis integrity. Red means failure or distortion can change wheel control, suspension geometry, chassis reaction, or rear-frame behavior.

The success criterion is strict: at the end of 45 minutes, every red part must have been identified before tools or materials are committed, and every green part must have a written reason that does not rely on material strength alone. If any part's failure consequence is unknown, you do not get to call it green. Mark it red or amber, write the uncertainty, and choose a safer fabrication target until the uncertainty is resolved.

Repeat the drill for three separate projects. By the third pass, you should notice the classification happening earlier. The point is not to produce a perfect engineering report. The point is to train the habit of seeing consequence before craftsmanship.

When this principle breaks down

The boundary principle breaks down when you use labels instead of jobs. Bodywork is not always harmless, aero is not always peripheral in consequence, and composite does not always mean advanced. The only reliable classifier is the part's function under load.

It also breaks down when you confuse rule permission, workshop possibility, and safe design. The corpus shows categories that legislated against carbon chassis on cost while still allowing composite use elsewhere. That is a reminder that allowance, cost, availability, and safety are separate questions. A part can be legal and still beyond your fabrication boundary. A part can be possible and still require professional design control.

The practical safeguard is to re-run the boundary check whenever the part touches a new system. If a cover becomes a mount, if a mount becomes a load path, if a load path reaches suspension or chassis structure, or if a composite repair becomes part of a monocoque or bulkhead, the project moves toward red. Stop at that moment and re-scope.

Author Review

No quiz questions are attached to this lesson.

Sources

#	Document	Chunk	Pages	Score	Collection
1	Competition Car Suspension Design Construction Tuning Staniforth	b24873ca-7f0f-ab83-dd5d-c80e05aab1ef	63	1	uio_books_raw_v1
2	Competition Car Suspension Design Construction Tuning Staniforth	9f2abb39-fb34-41fc-a442-ec3f38c002c6	11	1	uio_books_raw_v1
3	Competition Car Composites Simon McBeath	7af9252a-4312-8d39-b7c6-15ca052d7b8c	183	1	uio_books_raw_v1
4	Competition Car Composites Simon McBeath	2fd26ac3-6beb-d458-378d-1ca12307931e	1	1	uio_books_raw_v1
5	Racing Chassis and Suspension Design Carroll Smith	aa949643-fc01-2b30-e198-85671c76aedf	7	1	uio_books_raw_v1
6	Racing Chassis and Suspension Design Carroll Smith	124cda7f-0c5f-56dd-02aa-e09befd6a128	7	1	uio_books_raw_v1
7	Competition Car Suspension Design Construction Tuning Staniforth	0bf4a9a9-cb43-fcf1-9d5f-a1b1a64d3cf7	56	1	uio_books_raw_v1
8	Driving in competition None Johnson Alan 1935- None	f543ab5a-1dff-ae84-4e84-53dd5f48563c	8	1	uio_books_raw_v1