📘 Module 6 — Skirt Design & Materials

6.1 Base Design Principles

The Base as the Structural Foundation

The hovercraft base — also called the platform or chassis — is the single most consequential structural decision a team makes. Every other component mounts to it: lift fans, thrust fans, battery packs, wiring harnesses, speed controls, and switches. The base must be stiff enough to hold all these elements in precise alignment while simultaneously being light enough that the craft can levitate. Getting this balance right is not guesswork; it requires intentional engineering from the first sketch.

A poorly designed base causes cascading failures. If the base flexes under load, fan mounting angles shift, directing airflow in unintended directions and reducing both lift and thrust efficiency. If it is too heavy, the lift fan must work harder to generate the cushion pressure needed to suspend the vehicle, consuming power that could otherwise drive the craft forward. If it is difficult to modify, the team loses critical testing time drilling new holes, re-routing wires, and re-gluing components between runs.

Core Geometric Requirements

The base must be large enough to contain the skirt attachment perimeter, position the lift fan over a centered aperture, and provide mounting points for the thrust fan at the rear. A rectangular platform is most common because it provides natural symmetry about the longitudinal axis, simplifying center-of-mass calculations.

Typical competition hovercraft bases range from 25 × 30 cm to 40 × 50 cm depending on fan sizes. Larger bases spread the cushion pressure over a greater area, potentially improving stability, but they also increase total mass and aerodynamic drag from the skirt perimeter. Smaller bases are lighter but harder to stabilize.

The lift fan aperture — the hole through which air is blown into the skirt — should be centered on the base or slightly forward of geometric center to account for the rearward weight of battery packs. Aperture size is matched to the fan diameter with minimal clearance (1–3 mm) to reduce recirculation of pressurized air back up through the gap between fan blade tip and base edge.

Common Base Materials

Foam core board (3–5 mm thick foam sandwiched between paper facers) is the single most popular base material for beginner and intermediate teams. It offers a compelling combination of low density (~0.3 g/cm³ effective), easy workability with a hobby knife, and good compressive stiffness in the plane of the sheet. A 30 × 40 cm foam core base weighs approximately 20–30 g. The limitation is out-of-plane shear: foam core delaminates under bending loads if mounting hardware is not distributed over a face plate or washer.

Basswood sheet (2–4 mm thick) provides significantly higher stiffness and resistance to fastener pull-through. Basswood has a density of roughly 0.37 g/cm³ and a Young's modulus of approximately 9–12 GPa along the grain, compared to balsa's 3–6 GPa. For the same sheet size, a 3 mm basswood base weighs about 40–55 g — heavier than foam core but substantially stronger. Teams using larger, heavier fan assemblies often prefer basswood for its ability to resist warping under motor vibration.

Balsa wood occupies a middle ground: lighter than basswood (density ~0.12–0.20 g/cm³) but significantly weaker. Thin balsa sheets can crack under vibration or when impacted by the track surface. Balsa works well when combined with cross-grain lamination or when reinforced with carbon fiber tow at stress concentrations.

3D-printed PLA allows custom geometry — integrated fan ducts, clip mounts for JST connectors, recesses for battery packs — but adds mass quickly. A 3 mm wall PLA base for a medium-sized craft can weigh 80–120 g. Used selectively for brackets and mounting hardware rather than the full base, PLA offers excellent value.

Mounting Provisions

Every component should mount to the base with defined, repeatable attachment points rather than adhesive alone. Hot glue hardens in seconds and seems convenient, but it creeps under sustained vibration and heat. For fan mounts, M2 or M3 machine screws through the base into a plywood backing plate distribute load over a larger area and allow for disassembly. For battery packs, a hook-and-loop strap anchored to basswood rails keeps the battery removable for charging while preventing lateral sliding during acceleration.

Plan wire routing before finalizing the base geometry. Wires that cross the lift fan aperture or lie against the skirt attachment edge create snag points that tear the skirt. Route power and signal wires in channels bonded to the top surface, secured with cable staples or tie-wraps at 5–6 cm intervals.

Designing for Iterability

The most overlooked principle in base design is ease of modification. Teams that iterate rapidly — testing different fan positions, skirt styles, and battery placements between runs — consistently outperform teams with more rigid (literally and figuratively) platforms. Build the base so that fan mounting holes can be patched with adhesive-backed foam and re-drilled without compromising the structure. Standardize fastener types (all M2, or all M3) to reduce the number of tools needed at the test bench.

Leave deliberate blank areas on the base surface for future additions. A foam core platform with 5 cm of clear area forward of the lift aperture can accept a ballast slug, a second switch, or a third JST connector with minimal rework.

Common Mistakes

• Gluing the battery pack directly: Makes charging impractical and prevents weight redistribution for stability tuning.
• Undersizing the lift aperture: Creates a nozzle effect that raises static pressure but reduces total airflow volume, actually decreasing cushion height.
• Building the base too thin: A 2 mm foam core base will bow under motor vibration, misaligning the lift duct and leaking air at the fan-to-base interface.
• Ignoring the skirt attachment edge: The perimeter of the base where the skirt is taped or glued must be smooth and continuous; any steps or gaps cause local air leaks.

A well-designed base is almost invisible in performance terms — the craft levitates, the fans run true, and the wires stay out of the way. That invisibility is the goal.

6.2 Weight vs. Performance Trade-off

No Mass Limit Does Not Mean Mass Doesn't Matter

The 2026 Science Olympiad Hovercraft rules impose no explicit upper limit on vehicle mass. This is a trap for inexperienced teams. The absence of a rule does not mean mass is free — it means the physics themselves punish excess weight, often more severely than any point deduction would.

Understanding why mass is the enemy of hovercraft performance requires connecting three physical principles: cushion pressure, friction force, and available thrust. These principles interact in a way that makes mass reduction one of the highest-leverage engineering decisions a team can make.

Cushion Pressure and Required Lift Force

A hovercraft levitates when the air pressure trapped in the skirt volume equals the weight of the craft divided by the cushion area:

P_c = W / A_c

where P_c is cushion pressure (Pa), W is vehicle weight (N), and A_c is the enclosed cushion area (m²). For a craft with a 0.12 m² cushion footprint (roughly 30 × 40 cm):

• At 300 g total mass: W = 0.300 × 9.81 = 2.94 N → P_c = 2.94 / 0.12 ≈ 24.5 Pa
• At 500 g total mass: W = 4.91 N → P_c = 4.91 / 0.12 ≈ 40.9 Pa
• At 800 g total mass: W = 7.85 N → P_c = 7.85 / 0.12 ≈ 65.4 Pa

These required cushion pressures are not trivially achievable. A typical 65 mm diameter DC fan operating at 7.4 V generates a stall pressure of 70–120 Pa. A craft requiring 65 Pa is operating near the stall limit of its lift fan — any skirt leak, surface irregularity, or minor fan degradation causes complete hover failure. A craft requiring only 24 Pa has substantial margin and will hover reliably even with imperfect sealing.

Airflow Requirements Scale with Mass

The fan must supply not only the required static pressure but also a continuous flow to compensate for air escaping at the skirt hem. The required volumetric flow rate Q depends on both the gap height h at the skirt hem and the perimeter L of the cushion:

Q ≈ L × h × v_exit

where v_exit is the velocity of escaping air (related to cushion pressure by Bernoulli's equation: v_exit = √(2 × P_c / ρ_air)). For ρ_air ≈ 1.2 kg/m³:

• At P_c = 24.5 Pa: v_exit = √(2 × 24.5 / 1.2) ≈ 6.4 m/s
• At P_c = 65.4 Pa: v_exit = √(2 × 65.4 / 1.2) ≈ 10.4 m/s

The heavier craft's fan must supply 63% higher exit velocity — and thus much greater power — just to maintain the same hover height. Power scales as the cube of velocity in the regime where flow is constrained by the fan curve. This is the core reason mass reduction delivers outsized performance returns.

Ground Friction During Hover

Even while hovering, a hovercraft experiences friction. The skirt hem contacts the track surface, and this contact friction is proportional to the residual normal force not supported by air pressure. When cushion pressure is slightly below the required value — which happens whenever the fan is not at peak efficiency — the skirt hem bears part of the vehicle weight, and the resulting friction force is:

F_friction = μ × (W − P_c × A_c)

For a heavy craft marginally hovering, this term can be substantial. A 700 g craft where air pressure supports only 90% of the weight still has 70 g × 9.81 m/s² ≈ 0.69 N resting on the skirt hem. With a friction coefficient μ of 0.3 (typical for Tyvek on painted concrete), this yields F_friction ≈ 0.21 N — a significant drag on a vehicle whose entire thrust may only be 0.3–0.5 N.

Thrust-to-Weight Ratio

Available thrust from the thrust fan is essentially fixed by fan selection and battery voltage. If total available thrust is 0.4 N:

• 300 g craft: T/W = 0.4 / 2.94 ≈ 0.14 — reasonable acceleration
• 600 g craft: T/W = 0.4 / 5.89 ≈ 0.07 — sluggish, may not overcome static friction
• 900 g craft: T/W = 0.4 / 8.83 ≈ 0.045 — may fail to move at all

The relationship is linear, but because heavy craft also have higher friction loads, the effective performance degradation is superlinear with mass.

Where Does the Mass Go?

For typical Science Olympiad hovercrafts, the mass budget breaks down approximately as:

The battery pack is consistently the single heaviest component and the most difficult to reduce without sacrificing run time. Teams should resist the temptation to use oversized battery packs for "extra insurance" — a 6-cell NiMH pack may weigh 150 g when a 4-cell pack at 100 g provides adequate voltage and capacity for the competition window.

Structural Integrity vs. Mass: Finding the Floor

Mass reduction has a lower bound set by structural integrity. A base so thin it flexes under motor vibration will misalign the fan duct and lose more performance than the saved grams are worth. A skirt so lightweight it tears on the first run costs the team the entire competition. The design target is minimum mass consistent with reliable operation, not absolute minimum mass.

Practical rule of thumb: test the structure by applying a 2× load factor (twice the expected operating forces) and confirm no permanent deformation. If a 30 cm foam core base bows 2 mm when the fan runs, stiffen it with a 3 mm basswood spine bonded along the centerline before reducing sheet thickness further.

Common Mistakes

• Overbuilding the structure: Using 6 mm foam core or multiple layers of balsa "for safety" adds mass without proportional strength benefit.
• Using AA batteries instead of custom packs: Four AA alkalines weigh roughly the same as a proper NiMH pack but deliver lower voltage and higher internal resistance.
• Ignoring skirt mass: A heavy fabric skirt adds 30–50 g and increases drag; ultra-thin plastic sheeting is almost always the right choice.
• Not weighing the craft during development: Teams that track mass at every build iteration make consistently better tradeoff decisions than those who weigh the craft only at competition.

6.3 Center of Mass & Stability

Why Center of Mass Governs Hovercraft Behavior

A hovercraft in flight is supported by a distributed pressure field — the air cushion acts like a fluid foundation spread across the entire skirt footprint. Unlike a wheeled vehicle where the chassis is rigidly linked to the ground through discrete contact points, a hovering craft can tilt freely in any direction. The only restoring forces against tilt are the differential pressure distribution within the cushion (which depends on skirt geometry) and, in some designs, the craft's inherent geometric stiffness from a peripheral jet skirt.

This means that the location of the center of mass (CoM) relative to the center of cushion pressure (CoP) directly controls how the vehicle sits — and moves — during a run. A craft whose CoM is perfectly centered over its CoP floats level and accelerates symmetrically. Any offset between CoM and CoP produces a gravitational moment that tilts the craft, compresses the skirt on one side, and creates an asymmetric pressure distribution that drives steering deviation.

The Stability Condition

For a hovercraft to float in stable trim, the vertical line through the CoM must pass through the center of the cushion pressure polygon. In static equilibrium:

ΣM_CoP = 0 → Σ(m_i × d_i) = 0 for all components i

where m_i is component mass and d_i is the signed horizontal distance from the cushion center. This is simply a balance-of-moments condition, identical to the analysis of a lever or beam.

For a rectangular base with length L (fore-aft) and width W (lateral):

• Longitudinal CoM should be located at the geometric center longitudinally, or slightly rearward of center (by ≈5% of L) to pre-load the thrust fan slightly against the track surface and improve lateral control.
• Lateral CoM must be centered precisely. Even a 1–2 cm lateral offset in a 400 g craft creates a moment of 0.004–0.008 N·m that causes measurable steering deviation over a 3-meter run.

Heavy Components and Their Placement

The battery pack dominates the CoM calculation because it represents 25–40% of total vehicle mass. A 4-cell NiMH pack (typically 100–130 g) placed 3 cm off-center in a 350 g craft creates a lateral moment of approximately:

M = 0.115 × 9.81 × 0.03 ≈ 0.034 N·m

This moment must be balanced by the craft's cushion pressure restoring effect. For a typical peripheral bag skirt, the restoring stiffness is roughly 0.01–0.05 N·m per degree of tilt. A 0.034 N·m unbalanced moment thus causes 0.7–3.4 degrees of permanent tilt — enough to measurably compress the skirt on the heavy side and increase drag on that side, causing gradual course deviation.

Practical placement strategy:

1. Battery pack: centered laterally, positioned fore-aft to achieve target CoM location. Most teams place the battery forward of center to counterbalance the rearward weight of the thrust motor assembly.
2. Fans: symmetric about the longitudinal axis. If two lift fans are used side-by-side, they must be equidistant from the centerline.
3. Speed control electronics: small mass, place on whichever side needs trimming. A 15 g potentiometer moved 4 cm can offset 1–2 g of CoM error.

Low CoM for Pitch and Roll Stability

Hovercraft are inherently low-profile vehicles, but the principle of a low center of mass still applies. All other things equal, mounting heavy components (batteries, speed controllers) as low as possible — flush with the base surface rather than elevated on risers — lowers the CoM and increases angular stability.

The moment of inertia about the tipping axis (lateral axis for pitch, longitudinal axis for roll) is:

I = Σ(m_i × r_i²)

where r_i is the distance from each mass to the rotation axis. A low CoM means small r_i for the heaviest components, resulting in lower moment of inertia. This means the craft accelerates rotationally more slowly under any disturbing torque — effectively giving the cushion pressure more time to provide a restoring force before the craft tips beyond recovery.

Consequences of CoM Errors During Competition

Lateral offset (left-right imbalance): The craft curves to one side. At slow speeds this is a minor inconvenience; at high speeds it can cause the craft to exit the track boundaries. Teams often discover this problem for the first time on competition day when the track surface is different from their practice surface, because different skirt friction coefficients amplify or suppress the torque effect of the CoM offset.

Forward CoM offset: Pitch-forward tendency. The craft's nose tilts down, compressing the forward skirt segment and increasing drag. In extreme cases, the nose skirt contacts the track and the craft stops abruptly.

Rearward CoM offset: The craft tips back, reducing forward skirt pressure and allowing air to escape preferentially from the front. This can cause the craft to speed up unpredictably because forward skirt drag decreases, or to spin backward if the skirt fails on the nose side.

Finding and Adjusting the CoM

The simplest CoM measurement tool is a flat surface and two dowels: balance the assembled craft across a single dowel under the lateral centerline to find the longitudinal balance point, then across a dowel parallel to the longitudinal axis to find the lateral balance point. Mark both and compare to the cushion geometric center.

For lateral correction: add small adhesive mass (washers, tape) to the light side, or physically move the battery laterally by 2–5 mm increments, testing hover trim after each adjustment.

For longitudinal correction: slide the battery fore or aft in its mounting tray. Each 5 mm of battery displacement in a 300 g craft with a 100 g battery shifts the longitudinal CoM by approximately:

ΔxCoM = (m_battery / m_total) × Δx_battery = (100/300) × 5 mm ≈ 1.7 mm

Small adjustments have measurable effects — confirm with float tests on a level surface before competition.

Common Mistakes

• Trusting symmetrical appearance over measured balance: A craft that looks symmetric can be 5–10 g heavier on one side due to asymmetric wiring, unequal glue application, or different fan blade count. Always measure.
• Adjusting CoM with external ballast instead of repositioning components: Adding 10 g of ballast to fix a 10 g imbalance doubles the mass penalty. Repositioning the component is almost always better.
• Ignoring CoM shift when changing batteries or skirts between runs: Swapping a depleted battery for a freshly charged one of slightly different mass or size shifts the CoM and requires re-verification of trim.

6.4 Fan Mounting & Airflow Routing

Two Fans, Two Jobs

A Science Olympiad hovercraft uses two fundamentally different fans with incompatible airflow requirements. The lift fan must push a high volume of air straight downward through a duct into the skirt volume, building and maintaining the pressure cushion that suspends the craft. The thrust fan must push air rearward in an approximately horizontal plane to propel the craft forward. Conflating these requirements — or compromising either to serve both — is one of the most common design errors at the novice level.

The two fans often differ in type, diameter, motor characteristics, and mounting geometry. Understanding why helps explain the detailed mounting specifications that follow.

Lift Fan: Axial Orientation and Duct Integration

The lift fan must be mounted so that its rotational axis is perpendicular to the base plane — i.e., pointing straight down. Any angular offset from vertical wastes a fraction of the fan's thrust as a horizontal component, reducing effective lift by cos(θ) where θ is the tilt angle. At θ = 10°, you lose only cos(10°) ≈ 1.5% of lift, but the horizontal component (sin(10°) ≈ 17.4% of total thrust) now pushes against or away from the intended direction of travel, creating an asymmetric pitching moment.

The fan should be centered over the lift aperture in the base, with the fan shroud mated to the aperture edge. The ideal interface is a short cylindrical duct — essentially a collar — that smoothly transitions airflow from the fan exit to the skirt inlet. Without this duct, airflow expands after leaving the fan blades, and a significant fraction never enters the skirt, instead escaping radially under the base edge.

Duct length matters. A duct length-to-diameter (L/D) ratio of 0.5–1.0 is generally sufficient to collimate the airflow from a propeller-type fan. Longer ducts reduce flow volume due to skin friction without meaningfully improving pressure uniformity. Ducts made from foam core cut into a ring and glued to the underside of the base work well and add only 3–5 g.

Critical clearance: Fan blade tip clearance to the duct wall should be 1–2 mm. Larger gaps allow high-pressure air from below the fan to recirculate upward around the blade tips (tip vortex recirculation), reducing flow efficiency. Smaller gaps risk physical contact if the fan hub vibrates under load.

Thrust Fan: Horizontal Orientation and Thrust Line

The thrust fan mounts at the rear of the craft with its rotational axis horizontal and parallel to the base surface. Its exhaust should be directed straight rearward along the vehicle's longitudinal axis. Any lateral angular offset — even 5–10 degrees — produces a yaw moment that causes the craft to turn, which may be intentional (for course correction) but is more often a construction error that requires active compensation.

Mount height matters for pitching moment. If the thrust fan is mounted significantly above the craft's center of mass, the thrust force creates a nose-down pitching moment that compresses the front skirt and increases forward drag. Ideally, the thrust fan's centerline should be at approximately the same height as the vehicle's CoM, usually 2–5 cm above the base surface. For most competition builds with flat bases, this means the fan hub sits 3–4 cm above the base on a minimal riser bracket.

Mounting Hardware and Vibration Isolation

Fan motors generate vibration at their rotational frequency and harmonics. At 15,000–25,000 RPM (250–417 Hz), this vibration can resonate with base structural modes, causing noise, accelerated fatigue at mounting holes, and — critically — gradual loosening of fasteners.

Best practice: use M2 or M3 stainless machine screws with nylon-insert (Nyloc) nuts at all fan mounting points. Thread-locking compounds (Loctite 243) are an alternative but make disassembly more difficult during testing. For direct motor mounts on lightweight foam core bases, distribute the load with a 15–20 mm diameter backing washer or plywood pad bonded to the base around each mounting hole.

Silicone motor mounts — small cylindrical grommets that isolate the motor from the mount plate — can reduce transmitted vibration significantly for teams experiencing base fatigue. These add 2–4 g per motor but extend the life of both the base structure and the motor bearings.

Airflow Routing: Managing Interference

In a dual-fan system, the wakes of the two fans must not interfere with each other. Specific concerns:

1. Thrust fan ingesting recirculated skirt exhaust: If the thrust fan is positioned immediately rearward of the skirt perimeter, it may ingest slow-moving warm air from the skirt hem rather than free-stream air, reducing its effective velocity ratio and thrust. Leave at least 3–4 cm clearance between the rear skirt edge and the thrust fan intake plane.

2. Wiring crossing fan inlets: A single wire passing through the fan inlet can reduce airflow by 5–15% and can be catastrophically ingested. Route all wiring outside the fan swept area; use adhesive wire guides bonded to the base surface to anchor harnesses.

3. Lift duct blocking the thrust path: On narrow platforms, the lift duct housing may extend into the aerodynamic shadow of the thrust fan, requiring a slight lateral offset. Analyze the footprint before drilling and cutting.

Angle of Attack and Steering Fans

Some advanced teams incorporate a steerable thrust fan — a fan mounted on a servo-actuated pivot that allows yaw control. In Science Olympiad's no-servo format, steering is achieved by tilting the entire craft through CoM adjustment or by using asymmetric skirt resistance. Fixed fan mounting at a precise zero-yaw-offset angle is therefore doubly important: there is no active correction mechanism.

To verify zero offset, place the assembled craft on a flat, very smooth surface (glass or a laminate floor) and let it roll forward from a gentle push. Watch the heading over 2–3 meters. Any consistent left or right curve indicates a thrust fan angular offset or CoM imbalance. Correct by shimming the fan mount 0.5–1 mm on the appropriate side.

Common Mistakes

• Tilting the lift fan to "direct" air into the skirt: The skirt distributes pressure without needing angled input; a tilted lift fan wastes energy and creates uneven cushion pressure.
• Mounting the thrust fan too high: Elevates the thrust line above the CoM, creating a nose-down pitching torque that compresses the front skirt and increases drag.
• Allowing the fan duct to narrow toward the skirt inlet: Creates a venturi effect that accelerates air through the duct but reduces total volume flow — reducing, not increasing, hover height.
• Not checking thrust fan parallelism to the base: A fan that is slightly nosed down blows exhaust into the base rather than rearward, wasting thrust and heating the base surface.

6.5 Structural Materials Comparison

The Materials Design Space

Selecting structural materials for a Science Olympiad hovercraft requires balancing five competing criteria: mass, stiffness, strength, workability, and cost. No single material excels in all five. The best teams choose materials strategically — using the lightest option where loads are low, and stiffer, stronger materials only where the structure demands it.

The relevant structural performance metric is not absolute stiffness or absolute strength, but specific stiffness (Young's modulus divided by density, E/ρ) and specific strength (failure strength divided by density, σ/ρ). These quantities tell you how much structural performance you get per unit mass — the fundamental hovercraft design currency.

Foam Core Board

Density: ~200–300 g/m² (sheet areal density) or effective volumetric density ~0.25–0.35 g/cm³
Young's modulus (in-plane): ~1–3 GPa (depends on paper facer weight)
Common thicknesses: 3 mm, 5 mm

Foam core's paper facers carry the bending loads through a sandwich beam mechanism — the foam core provides shear transfer between the two paper layers without contributing much to bending stiffness itself. This makes foam core highly efficient in bending for its areal density.

For an unsupported span of 30 cm (e.g., a base section between two support ribs), a 5 mm foam core panel deflects approximately 0.3–0.5 mm under a 0.5 N center load — acceptable for most builds. The critical failure mode is delamination: when a point load is applied (e.g., a fan motor pulling upward on its mount screws), the paper facer separates from the foam if the load is not distributed by a backing plate.

Best uses: Main base platform, lightweight bulkheads, non-structural covers.
Avoid when: High vibration loads are expected at single-point attachments; or when the material will be exposed to heat from motors.

Balsa Wood

Density: 100–200 kg/m³ (0.10–0.20 g/cm³, varies by grade)
Young's modulus (along grain): 3–6 GPa
Specific stiffness (E/ρ): ~25–40 GPa·cm³/g
Common forms: Sheet (1–6 mm), strip, block

Balsa is the classic lightweight modeling material. Its specific stiffness rivals steel along the grain, making it excellent for structural spars and ribs. However, across the grain it is dramatically weaker — a 3 mm sheet can be snapped by hand perpendicular to grain. This anisotropy means balsa structures must be oriented carefully, with grain running along the primary load path.

Balsa sheet absorbs moisture and warps if stored improperly. A base cut from balsa that was flat in the workshop can bow 2–3 mm overnight in a humid environment, misaligning the lift duct. Sealing all balsa surfaces with thin cyanoacrylate (CA) glue or shellac before final assembly significantly reduces moisture sensitivity.

Best uses: Longitudinal spars, ribs, struts, any member where the primary load is axial (tension or compression along grain).
Avoid when: Multi-directional loads are expected (use plywood instead), or the part is large-area and must resist bending without a specific grain orientation.

Basswood

Density: 350–450 kg/m³ (0.35–0.45 g/cm³)
Young's modulus (along grain): 9–12 GPa
Specific stiffness (E/ρ): ~25–30 GPa·cm³/g
Common forms: Sheet (1–4 mm), strip, square stock

Basswood is approximately 2–3× denser than balsa but 2–3× stiffer and significantly stronger, particularly in shear. Its specific stiffness is comparable to balsa, but its specific strength advantage becomes apparent under impact and vibration loads. Screw and nail pull-through strength is much higher than balsa, making basswood the preferred material when mechanical fasteners are used.

Basswood sheets are cross-laminated at the manufacturing stage (similar to plywood), giving them more uniform properties across grain directions than solid-grain balsa. This makes basswood preferable for base platforms that must resist bending in all orientations.

Best uses: Base platforms when rigidity is paramount, motor mounting plates, structural brackets.
Trade-off: A 3 mm basswood base is approximately 2.5–3× heavier than a 5 mm foam core base of the same footprint.

3D-Printed PLA

Density: 1.24 g/cm³ (solid); effective density of printed parts: 0.6–1.0 g/cm³ depending on infill
Young's modulus: 2.7–3.5 GPa
Specific stiffness (E/ρ): ~2.5–3 GPa·cm³/g (much lower than wood)
Common infill: 15–25% for structural brackets; 40–60% for load-bearing mounts

PLA's specific stiffness is significantly lower than wood alternatives, meaning PLA parts are always heavier for the same structural performance in bending. However, 3D printing enables complex geometry — integrated ducting, clip features, weight-saving pockets, snap-fit connectors — that no other material can achieve without expensive machining.

The correct use of PLA in hovercraft design is selective: motor mount brackets with integrated cable channels, fan duct adapters, battery retention trays, and switch housings benefit from 3D printing. Full-platform bases in PLA are almost always heavier than equivalent foam core or basswood alternatives.

Thermal note: PLA softens at ~60°C. Motor temperatures on stall can exceed this. Use PETG (Tg ~80°C) or ABS for parts in direct contact with motors.

Best uses: Brackets, connectors, ducts, custom-geometry mounts.
Avoid for: Large flat structural panels; they will be heavier and less stiff than foam core alternatives.

Carbon Fiber (CFRP)

Density: 1.5–1.8 g/cm³ (solid laminate)
Young's modulus (unidirectional, 0° fibers): 70–150 GPa
Specific stiffness: ~50–90 GPa·cm³/g — highest of any common material
Forms available to students: Pre-cut rods, tubes, woven fabric, pre-preg sheets

Carbon fiber's specific stiffness is 2–4× higher than any wood product. A 0.5 mm carbon fiber plate is equivalent in stiffness to a 2 mm basswood sheet at roughly half the mass. For teams competing at the highest level, replacing basswood spars or base skin panels with carbon fiber can save 15–30 g with no strength penalty.

The barrier is cost and workability. Carbon fiber sheets must be cut with diamond-tipped tools or score-and-snap (for thin fabric sheets). Edges must be sealed to prevent skin cuts from exposed fibers. Pre-preg sheets require oven curing. For most teams, the cost (carbon fiber sheet: $20–50 per board) and safety complexity outweigh the mass benefit unless the team is already achieving near-minimum mass from other sources.

Best uses: High-performance teams needing the last few grams of savings; structural tubes for motor risers; base stiffening spars.

Comparative Summary

Common Mistakes

• Using PLA for entire base: 3–5× heavier than foam core equivalent, often with lower stiffness.
• Using thin balsa without grain consideration: A 2 mm balsa sheet across the grain has the strength of wet cardboard.
• Mixing fastener-incompatible materials: Hot gluing a metal motor mount to foam core without a backing plate will delaminate under vibration within minutes.

6.6 Modular Design for Testing

Why Modularity Wins Championships

The Science Olympiad Hovercraft competition structure — impound before runs, 8-minute time limit, no major mid-competition rebuilds — seems to make modularity irrelevant. Teams complete their testing before competition day and submit a finished vehicle. Yet teams that invest in modular architecture during development iterate faster, test more configurations, and ultimately arrive at competition with a more refined vehicle.

The competitive advantage emerges during the pre-competition phase. A team that can swap fans, skirts, batteries, or motor controllers in 30 seconds rather than 10 minutes can run 3–4 more test cycles in a practice session. Over a month of development, that's 40–60 additional data points informing design decisions. The compounding effect is profound: more iterations → faster convergence to optimal parameter space → higher confidence in final design → fewer surprises at competition.

Lift Fan Modularity

The lift fan should be the quickest component to swap. Ideally, the fan and motor mount as a single unit — "cartridge" style — to a quick-release bracket. Options:

Sliding cartridge with alignment pins: The fan assembly mounts on a 10 mm thick PLA or basswood plate with two 4 mm dowel pins for alignment. Two quick-release latches (3D-printed toggle clamps or simple latch hooks) hold the plate in position. Unlatch, slide out, slide in new configuration, latch. Time: 15–30 seconds. Alignment is repeatable to within 1–2 mm.

Clip-on mounting: Design the fan shroud to clip onto a fixed base bracket via two or three snap fit tabs. This requires careful 3D-printed geometry but enables changing fans without any tools. Time: 5–10 seconds. Risk: repeated clipping/unclipping can fatigue the snap tabs.

Double-sided adhesive with quick-detach interface: Use heavy-duty double-sided tape to mount fan assemblies to interchangeable mount plates. Swap the entire plate between tests. Time: 10–20 seconds. Cleanup: old adhesive residue must be scraped off, which adds 30–60 seconds if the geometry is complex.

The lift duct interface must be standardized. If you test different fan diameters (50 mm vs. 65 mm), the duct changes. Design duct adapters that bolt or snap to a central collar rather than gluing the duct to the base. This way a new duct can be swapped in without disturbing the base or aperture region.

Thrust Fan Modularity

The thrust fan is typically heavier and more rigidly mounted than the lift fan, making swapping slower. However, the same principles apply: a quick-release mount bracket lets the team test different thrust fan sizes or motor power levels without extensive teardown.

Battery placement interacts with thrust motor weight. If the team is testing a battery mass range from 80 g to 140 g and a motor mass range from 25 g to 45 g, they want to quickly reposition the rear assembly for CoM balance. Use a slide-rail system: the motor mount plate slides along two 8 mm aluminum rods or smooth wooden rails bonded to the base. Stop-blocks at key positions (for different battery-motor combinations) let the team quickly set the thrust module's fore-aft location.

Battery Pack Modularity

Batteries should be mounted, not glued. Options:

Velcro/hook-and-loop strap: Two adhesive-backed strips of heavy-duty hook-and-loop fastener (rated for 0.5–1 kg per 5 cm × 5 cm area) secure the battery tray. Unbind, lift out, swap for a different battery configuration, rebind. Time: 10 seconds. The battery tray itself is a simple 3D-printed or laser-cut box that holds the battery in place; swapping battery tray + battery together takes 20–30 seconds.

Sliding tray: The battery mounts in a 3 mm PLA or basswood tray that slides into a fixed base bracket via two slots. Slide in, lock with a quick-pin or lever, slide out to swap. Time: 15–30 seconds. Alignment is precise and repeatable.

A well-designed battery system lets the team test:

• 3-cell vs. 4-cell packs (8.1 V vs. 10.8 V nominal) without rewiring
• Fresh vs. discharged batteries (voltage drop impacts speed)
• Different physical locations for CoM tuning

Skirt Modularity

The skirt is glued around the base perimeter, making it the slowest component to swap. However, designing for rapid skirt iteration is still high-value:

Tape-on skirt with removable base rim: Glue a basswood or PLA rim (15 mm tall, 5–8 mm thick) around the base perimeter. The skirt material (plastic sheet, Tyvek, mylar) is taped to this rim with strong waterproof tape. Replacing the skirt requires removing the old tape (3–4 minutes) and applying new tape to the same rim (4–5 minutes). Total swap time: 7–10 minutes. Not fast, but much faster than cutting and re-gluing the skirt base-to-base.

Multiple pre-built skirts: Build 3–4 skirts with different materials, hole patterns, or dimensions. Between test runs, swap the entire craft's base-plus-skirt assembly. Time: 30 seconds if the battery and fans remain connected. This requires building multiple bases, so it's only practical for teams with significant construction time.

Sensor and Control Modularity

Speed controllers and potentiometers rarely fail or need testing variations, so modularity here is lower priority. However, solder + JST connector pairs at all electrical junctions let teams quickly disconnect and reconnect components without desoldering.

A standard JST-PH or JST-XH connector on every motor lead, battery lead, and controller lead means:

• Swapping a burned-out motor takes 30 seconds
• Testing a different speed controller takes 15 seconds
• Disconnecting for transport is safe and repeatable

Cost: $0.02–$0.05 per connector. Time to assemble: 10 minutes per vehicle. Value: immeasurable.

Documenting Modular Variants

As the team tests different configurations, data quality depends on logging what was changed. Maintain a simple table:

This record is invaluable during competition if an issue arises: "We never tested this configuration before, but Configuration 2 from March 1st worked well. Let's revert to that."

Common Mistakes

• Building only one skirt: If the primary skirt fails at competition, the team has no fallback. Build two skirts (one primary, one spare) that can be swapped in under 10 minutes.
• Gluing fans directly to the base: Requires 20–30 minutes to cut free and re-mount a replacement. Use quick-release brackets from the start.
• Not standardizing connectors: A motor that works with motor #1 requires desoldering to fit motor #2. Standardize all connections to JST connectors at every component boundary.
• Over-designing for modularity: If the modular system weighs 20 g more than a fully integrated design, and you only plan 3–4 test iterations, it's not worth it. Balance modularity against weight and complexity.