Prototyping and Fabrication
Prototyping is the bridge between design intent and physical reality. A prototype is a testable embodiment of a design idea -- it answers questions that analysis alone cannot. This skill covers the full range of prototyping methods from cardboard mockups to CNC-machined metal parts, with emphasis on choosing the right fidelity level, the right fabrication method, and the right test plan for each stage of the design cycle.
Agent affinity: lovelace-e (materials and fabrication expertise), watt (mechanical systems and testing)
Concept IDs: engr-rapid-prototyping, engr-testing-methodology, engr-data-from-experiments, engr-failure-analysis
Prototyping Strategy
Fidelity Levels
| Level | Purpose | Typical materials | Time | Cost |
|---|
| Proof of concept | Does the idea work at all? | Cardboard, foam, tape, Arduino | Hours | $1-50 |
| Form study | Does it look and feel right? | 3D-printed plastic, clay | Hours to days | $10-200 |
| Functional prototype | Does it perform as designed? | 3D print, laser-cut, machined parts | Days to weeks | $100-5,000 |
| Pre-production | Can it be manufactured? | Production materials, production-intent processes | Weeks to months | $1,000-50,000+ |
The cardinal rule. Build the lowest-fidelity prototype that answers the question. A cardboard mockup that proves a mechanism works in 30 minutes is worth more than a machined prototype that proves the same thing in 3 weeks. Save high-fidelity prototyping for questions that require it: material strength, precision fit, thermal behavior.
What Question Does This Prototype Answer?
Before building anything, write down the question:
- "Will the gear train fit in the housing?" (form study)
- "Can the mechanism produce the required force?" (functional prototype)
- "Does the user understand the interface?" (proof of concept + user test)
- "Will the part survive 10,000 cycles?" (functional prototype with production materials)
A prototype without a question is a hobby project. An engineering prototype has a test plan before the first cut is made.
CAD Fundamentals
Computer-Aided Design is the standard tool for defining geometry before fabrication.
CAD Workflow
- Sketch in 2D: Define profiles with dimensions and constraints.
- Feature in 3D: Extrude, revolve, sweep, loft, or shell the sketch into solid geometry.
- Assembly: Mate parts together with constraints (coincident, concentric, parallel, distance).
- Drawing: Generate 2D engineering drawings from the 3D model with dimensions and tolerances.
- Export: STL for 3D printing, DXF for laser cutting, G-code for CNC, STEP/IGES for interchange.
Parametric vs. Direct Modeling
| Approach | How it works | Best for |
|---|
| Parametric | Features linked by a history tree; change a dimension and everything updates | Production parts with revision control |
| Direct | Push/pull faces without history; faster for quick exploration | Proof-of-concept and form studies |
Common mistake. Over-constraining sketches. A fully constrained sketch turns black/green in most CAD systems. Under-constrained sketches move unexpectedly. Over-constrained sketches refuse to update. The goal is exactly constrained.
3D Printing
Fused Deposition Modeling (FDM)
| Parameter | Typical value |
|---|
| Layer height | 0.1 - 0.3 mm |
| Materials | PLA, ABS, PETG, Nylon, TPU |
| Accuracy | +/- 0.5 mm |
| Strength | Moderate (anisotropic -- weak between layers) |
| Cost per part | Low |
| Build volume | 200x200x200 mm (desktop) to 500+ mm (large format) |
Best for: Proof of concept, form studies, jigs and fixtures, low-load functional parts.
Limitations: Anisotropic strength (layers are the weak point), visible layer lines, limited material properties compared to machined or injection-molded parts.
Stereolithography (SLA/DLP)
| Parameter | Typical value |
|---|
| Layer height | 0.025 - 0.1 mm |
| Materials | Standard resin, engineering resin, flexible resin, castable resin |
| Accuracy | +/- 0.1 mm |
| Strength | Moderate to high (isotropic) |
| Cost per part | Moderate |
| Build volume | 130x80x150 mm (desktop) to 300+ mm (large format) |
Best for: High-detail form studies, snap-fit prototypes, investment casting patterns, dental/medical models.
Limitations: Post-curing required, resin handling (gloves, ventilation), UV sensitivity of uncured parts, higher material cost than FDM.
Selective Laser Sintering (SLS)
| Parameter | Typical value |
|---|
| Layer height | 0.1 mm |
| Materials | Nylon (PA12, PA11), glass-filled nylon, TPU |
| Accuracy | +/- 0.3 mm |
| Strength | High (near isotropic) |
| Cost per part | High |
| Build volume | 250x250x300 mm typical |
Best for: Functional prototypes requiring durability, complex geometries (no support structures needed), living hinges, production-intent parts in nylon.
Limitations: Grainy surface finish, limited material palette, expensive machines (service bureaus common).
Choosing a Print Technology
| Question | FDM | SLA | SLS |
|---|
| Cost-sensitive? | Best | Moderate | Expensive |
| Need fine detail? | No | Yes | Moderate |
| Need strong parts? | Moderate | Moderate | Best |
| Complex internal geometry? | Needs supports | Needs supports | No supports |
| Large parts? | Yes | Limited | Moderate |
CNC Machining
CNC (Computer Numerical Control) removes material from a solid block to create the desired shape. It produces parts from real engineering materials (aluminum, steel, titanium, engineering plastics) with tight tolerances.
CNC Process Types
| Process | How it works | Typical tolerance |
|---|
| Milling | Rotating cutter, workpiece fixed | +/- 0.025 mm |
| Turning (lathe) | Workpiece rotates, cutter fixed | +/- 0.013 mm |
| Wire EDM | Electrical discharge cuts along a wire path | +/- 0.005 mm |
Design for CNC
- Avoid deep narrow pockets. Tool length/diameter ratio should not exceed 4:1 without special tooling.
- Use standard radii. Inside corners require a radius equal to the cutter radius. Specify standard cutter sizes.
- Minimize setups. Each time the part is repositioned introduces error. Design features accessible from as few orientations as possible.
- Specify tolerances only where needed. Tight tolerances cost money. Default tolerance (+/- 0.1 mm) is fine for most features; tighten only at functional interfaces.
Workshop Skills
Even in the age of digital fabrication, manual workshop skills are essential for rapid prototyping.
Core Competencies
| Skill | Tools | When used |
|---|
| Measuring | Calipers, micrometers, height gauges | Dimensional verification of all prototypes |
| Cutting | Band saw, hacksaw, snips, utility knife | Rough shaping of stock material |
| Filing and deburring | Files, deburring tools, sandpaper | Finishing edges and removing sharp burrs |
| Drilling | Drill press, hand drill, step drill | Hole-making in any material |
| Fastening | Taps, dies, wrenches, screwdrivers | Assembly of multi-part prototypes |
| Soldering/brazing | Soldering iron, torch | Electrical connections, metal joining |
| Adhesive bonding | Epoxy, cyanoacrylate, contact cement | Joining dissimilar materials |
Safety
Workshop safety is non-negotiable:
- Eye protection whenever cutting, drilling, grinding, or machining.
- Hearing protection with power tools above 85 dB.
- No loose clothing, jewelry, or long hair near rotating machinery.
- Know the emergency stop location on every machine before using it.
- Never work alone in a workshop with power tools.
Testing Prototypes
Test Planning
Before fabrication, write the test plan:
- Objective: What question does the test answer?
- Setup: How is the prototype mounted and loaded?
- Instrumentation: What is measured (force, displacement, temperature, time)?
- Procedure: Step-by-step test execution instructions.
- Success criteria: What result constitutes pass/fail?
- Data recording: How is data captured and stored?
Common Test Types for Prototypes
| Test | What it reveals |
|---|
| Fit check | Do parts assemble correctly? |
| Load test | Does the structure carry the design load? |
| Cycle test | Does the mechanism survive repeated operation? |
| Drop test | Does the product survive impact? |
| Thermal test | Does performance change with temperature? |
| User test | Can the intended user operate it? |
Failure Analysis of Prototypes
When a prototype fails -- and it should, because that is how learning happens -- analyze the failure:
- What failed? Identify the failure location and mode.
- Why did it fail? Root cause: design error, material deficiency, fabrication defect, or test error?
- What does the failure teach? Update the design, material selection, or analysis.
- What is the fix? Specific design change, with verification plan.
The mindset. A prototype that does not fail is either over-designed (wasted resources) or under-tested (missed knowledge). Failures in prototyping are cheap lessons that prevent failures in production.
Cross-References
- lovelace-e agent: Primary agent for materials, fabrication methods, and manufacturing process selection.
- watt agent: Mechanical prototype testing, thermal testing, and mechanism analysis.
- brunel agent: Prototyping within the broader design cycle, prototype planning at design reviews.
- design-process skill: Prototyping is Phase 6 of the design cycle.
- structural-analysis skill: Predicting prototype structural behavior before testing.
- technical-communication skill: Documenting prototype test results and failure analyses.
References
- Ulrich, K. T., & Eppinger, S. D. (2020). Product Design and Development. 7th edition. McGraw-Hill.
- Thompson, R. (2007). Manufacturing Processes for Design Professionals. Thames & Hudson.
- Chua, C. K., Leong, K. F., & Lim, C. S. (2010). Rapid Prototyping: Principles and Applications. 3rd edition. World Scientific.
- Redwood, B., Schoffer, F., & Garret, B. (2017). The 3D Printing Handbook. 3D Hubs.
- Oberg, E., et al. (2020). Machinery's Handbook. 31st edition. Industrial Press.
