Sarah Clarity Studios

Chapter 2: Collision‑Free Paths & Clearance Checks

Created by Sarah Choi (prompt writer using ChatGPT)

Collision‑Free Paths & Clearance Checks — Transformation Sequencing (Convert, Dock, Combine) for Mecha

The most common reason a transformation design “falls apart” is not style, not coolness, and not even complexity. It’s collisions. Parts that would scrape, intersect, crush cables, or swing through occupied space break believability instantly—especially once animation begins. Collision‑free paths and clearance checks are the discipline that turns a transformation from a magic trick into a machine.

For concept artists on the concepting side, thinking in collision‑free paths improves readability and helps you design transformations that can be storyboarded cleanly. For production‑side concept artists, clearance logic becomes practical guidance for modeling, rigging, and animation: it defines hinge axes, motion arcs, stop angles, safe states, and “no‑go” volumes. The goal isn’t to calculate engineering tolerances; it’s to build a consistent spatial story where every part has a believable route.

Why clearance is the heart of transformation credibility

Any moving assembly needs space to move through. If the space isn’t there, something else must move first to create it. That is why sequencing exists. A panel opens to create clearance. A joint retracts to avoid a pinch zone. A module shifts outward before rotating. Every satisfying transform moment is built on this logic.

Clearance is also what gives transformations their rhythm. “Open, reveal, rotate, seat, lock” feels good because each action creates the space for the next action. When transformations feel messy, it’s usually because the design lacks those deliberate clearance beats.

Three transformation families, three collision risks

Collision‑free thinking changes slightly depending on whether you’re converting, docking, or combining.

In convert systems (one body to another), the enemy is internal interference: shells colliding with inner frames, limbs intersecting torso cavities, and panels trying to occupy the same space in different modes.

In dock systems (unit to station/carrier), the enemy is approach geometry: wrong approach vector, misalignment, and collision between protrusions during the final “seat.”

In combine systems (multiple units forming one), the enemy is multi‑body interference: two units’ silhouettes colliding during approach, misaligned seams causing binding, and load‑bearing interfaces that cannot seat because secondary parts block them.

A good concept solution starts by identifying the dominant collision risks for your system type.

Clearance checks as an artist tool: think in “volumes,” not lines

Most collision problems arise because artists think in 2D lines rather than 3D volumes. The fix is to treat every moving part as occupying a simple volume as it moves: a box, capsule, cylinder, or wedge. You don’t need a full CAD model to do this. You need a mental “collision bubble.”

When a part rotates, imagine the swept volume of that rotation: a circular arc that becomes a thick donut shape in space. When a part translates, imagine the box it drags through. When a part both rotates and translates, imagine a larger blended volume. If any other geometry sits inside that volume during the move, you have a collision unless something moves first.

This volume mindset is one of the biggest upgrades you can make to transformation design.

The “swept arc” rule: every hinge has a footprint

Hinges are the most common collision source because artists underestimate their arcs. A door doesn’t just “open.” It swings through a radius. A rotating shoulder shell doesn’t “turn.” It sweeps a thick arc that will hit nearby protrusions unless you provide clearance.

A simple concept‑side habit is to draw an arc arrow and a faint ghost shape of the part at the start and end of the move. If the ghost overlaps existing forms, you either change the axis, retract first, or redesign the part.

On the production side, this becomes a rigging constraint: hinge axis placement and stop angles must be consistent with your intended arc.

Convert systems: building clearance states that feel intentional

Conversion sequences usually require one or two dedicated “clearance states.” These are moments where panels open like petals, shells slide apart, or limbs retract slightly to create space. Without clearance states, conversions become impossible origami.

A useful convert pattern is: release cosmetic panels → expand outward (create clearance) → reposition cores (rotate/translate major assemblies) → collapse inward (seat) → close panels → lock.

The “expand outward” beat is key. It gives you a believable reason for why the machine can rearrange itself without scraping. Visually, it also makes the transformation feel more powerful and mechanical.

Clearance corridors: designing intentional “routes” inside the body

Convert transformations become much more believable when you design internal corridors—paths where parts can move through the body without conflict. Think of them as mechanical hallways. A forearm might fold into a torso cavity, but only if there’s a corridor shaped for it.

In concept art, you can hint at corridors with panel seams and recesses that suggest “this space is reserved.” In production, these corridors inform internal geometry, collision volumes, and how tight the rig can be.

If you don’t want to draw internals, you can still imply corridors by shaping outer armor with deeper cavities, stepped layers, and visible “channel edges.”

Dock systems: approach vectors and “funnel geometry”

Docking collisions are usually approach mistakes. The unit tries to dock from the wrong angle, or protrusions hit before the primary interface can seat. The solution is to design an approach vector and support it with geometry.

Funnel geometry is your best visual shorthand. Chamfered lead‑ins, guide rails, tapered collars, and conical sockets all communicate “you can come in slightly off and still align.” They also reduce the required precision of animation.

In concepting, you can show funnel geometry as bold chamfers and nested rings. In production, you can specify the approach axis with a simple arrow and note: “approach along Z; funnel lead‑in handles ±5°.” You don’t need numbers, but you need the concept.

Dock clearance: what must retract before contact

A strong dock sequence includes a “retract state” where fragile protrusions get out of the way. Antennas fold, wings tuck, weapon barrels align, landing feet adjust. This is not just cool motion—it is collision management.

Your dock state chart can explicitly include: “Clear protrusions → Align pins → Seat collar → Lock → Transfer.” If you skip “clear protrusions,” you invite collisions.

Visually, retract states are also great storytelling. They feel like real pre‑dock procedure.

Combine systems: collision‑free choreography between multiple bodies

Combine transformations have choreography problems because multiple units move in shared space. The key is to stage the approach so the largest silhouettes don’t cross.

A helpful combine design habit is to define “lanes.” Each unit approaches in its own lane, then merges at the seam only after the lanes are clear. This can be shown with staggered heights, offset approach angles, or one unit holding position while another seats.

Combine systems also benefit from “pre‑fold states.” Each unit folds into a docking‑ready silhouette that reduces snagging. Those pre‑fold states should look stable and locked—because you don’t want units flapping while they dock.

Seam clearance: the hidden enemy of combining

Even if two units can approach without collision, they can still fail to combine if seam geometry binds. A believable combine seam needs clearance for alignment pins, room for clamps to swing, and a path for the seam to fully close.

A good seam has three layers: a lead‑in (chamfer), an alignment stage (pins/keys), and a final clamp stage (locks). If you design a seam as two flat faces meeting perfectly, it will feel unrealistic and will be hard to animate without visible interpenetration.

In concept art, show stepped edges and nested lips. In production, call out “tongue‑and‑groove” or “nested collar” so modelers build the seam with depth.

Pinch zones and no‑go volumes: where mechanics get hurt

Pinch zones are areas where moving parts could trap cables, crush fingers, or pinch armor layers. They are also areas where collisions are most likely. In mecha transformations, pinch zones often occur near elbows, shoulders, hip skirts, and folding wings.

A useful depiction trick is to design pinch zones as “guarded corridors” with covers, bellows, or clear gaps that indicate safe movement. Even a simple recessed channel implies that the designer planned for motion.

On the production side, pinch zones become collision masks: areas where rigging should avoid intersections and where animated parts need more generous spacing.

Cables, hoses, and flexible systems: clearance checks for the soft parts

Hard‑surface collisions are obvious, but flexible systems cause subtle believability failures. If a torso rotates and the cable bundle doesn’t move or gets cut through armor, the transformation feels fake.

Design flexible corridors. Give looms strain relief near hinges. Provide slack loops where rotation occurs. Provide clamps and brackets in the rigid zones and flexible sleeves in the motion zones. You can stylize all of this, but the logic must remain: flexible parts need space and controlled paths.

In concept sheets, a single callout showing “loom path during transform” can save enormous downstream confusion.

Clearance checks as “drawing checkpoints” (concept-side)

A practical way to integrate clearance without turning into engineering is to adopt three drawing checkpoints.

First, the ghost test: draw the part in start and end positions, lightly. If it overlaps, you need a different path.

Second, the swept volume test: imagine the arc or translation volume and check if anything occupies it.

Third, the seal and lock test: after the move, does the part seat somewhere that looks supported and lockable? If it ends “floating,” you may have solved collision but lost credibility.

These checkpoints can be applied quickly to every major moving assembly.

Clearance checks as “rig rules” (production-side)

For production, clearance checks become simple rig rules. Define hinge axes and stop angles. Define which parts must retract before others can move. Define “collision critical” areas where interpenetration is unacceptable.

You can package this as a small sequence sheet: key states with arrows, plus a short list of dependencies: “A opens before B rotates,” “C retracts before D swings,” “E cannot lock until F seats.” This is how concept art becomes buildable behavior.

If the project supports technical animation, include notes about expected tolerances—like intentionally leaving a visible gap so parts don’t clip in motion.

Using gaps on purpose: the honesty of “clearance margin”

One of the most mature transformation design choices is leaving deliberate gaps. Many artists want everything flush, but flush designs collide. A small, consistent clearance margin makes motion believable and protects animation.

You can integrate gaps aesthetically by turning them into shadow lines, gasket seams, stepped layers, or telescoping collars. When done well, clearance margins look like premium industrial design rather than “we couldn’t fit it.”

Sequencing beats: when to choose slide vs rotate vs fold

Collision‑free paths often come down to choosing the right motion type. Rotations have large swept arcs; slides have predictable volumes; folds combine both. If you have tight space, sliding or telescoping can be more believable than rotating a large shell.

A good rule is: big shells slide or hinge with large clearance; small elements can rotate. Telescoping is especially useful for convert systems because it creates space before rotation. In dock and combine systems, telescoping collars and sliding clamps can seat cleanly without huge arcs.

Common mistakes and how to fix them

A common mistake is designing transformations in a single view without checking other angles. Fix it by doing quick orthographic or top‑down clearance thumbnails for your key states. You don’t need full orthos—just enough to see if the arcs collide.

Another mistake is forgetting thickness. Panels have thickness, and thickness increases collisions. Fix it by drawing edge thickness on moving shells and leaving clearance margins.

A third mistake is designing seams as perfect, flat joins. Fix it by adding chamfers, lead‑ins, and nested lips so parts can seat without binding.

Finally, many designs ignore flexible systems. Fix it by planning cable corridors and adding strain relief at joints.

A reusable mini-system: clearance-first sequencing for any transformation

If you want a repeatable framework, design every transformation with four clearance stages: Expose → Expand → Reconfigure → Seat. Expose means unlock and open covers. Expand means create space—petal panels, telescoping, outward shift. Reconfigure means rotate/translate major assemblies through clear corridors. Seat means nest, clamp, and close.

Then apply a simple rule: every major move must have a visible clearance justification and an end-state support/lock justification. If you can show both, your convert, dock, and combine systems will feel controlled, readable, and buildable—exactly what transformation sequencing is supposed to achieve.