Phase A Bench-Test Protocols | Spartan Design VRC

SECTION 0 / 5

Why These Tests Exist

Three Phase A bots have open geometry questions that block confident field testing. Each protocol on this page targets one of those questions with structured data collection.

The Three Open Questions

Bot	Open question	Blocks	Protocol
Skimmer 2822A	Will the tube at Position B catch a vertical drop from the SG11-raised loader chute exit at 11.85″ off tile?	Path B (the architecture's identity)	Section 1
Osprey 2822E	Does the 6.5″ chassis cavity nest correctly on all three goal heights (3.25″, 5.77″, 8.77″)?	Goal-aligned deposit reliability	Section 2
Pelican 2822C	What's the lowest safe arm angle that keeps the lift within the SG2 24″ horizontal envelope?	Software stop floor + driver training	Section 3

Spoonbill 2822D is not on this list because its blocker is a build decision (12″ arm vs 14″ arm, with software-stop lockout), not a measurement that requires bench testing. Once the team commits to the 12″ arm + software stop at +30°, Spoonbill is ready to proceed.

📊

When to run these tests. After the bot is assembled enough that the tested subsystem works, but before mounting on a competition field. Pre-scrimmage. The tests need ~30–60 minutes each plus setup time. Run them on a workbench with the loader/goals mocked in cardboard or with real game elements if available.

What You Get From a Pass

Passing a protocol unlocks a specific next-step claim the team can make in EN4 and to judges:

Skimmer pass → "Path B works at the rate of [85%+] across all element orientations; we selected fix [X] over alternatives [Y, Z] for reasons [...]"
Osprey pass → "The chassis cavity nests correctly on all three goal heights with the driver's natural approach precision; we set the cavity tolerance at [final dim] after measuring [start dim]"
Pelican pass → "We measured the SG2 envelope at 9 arm angles and committed to a +30° software-stop floor with mechanical hard-stop backup; the lift cannot violate SG2 even under auton fault conditions"

The data collected during these tests is high-value notebook content. Document the rejected alternatives with the same care as the accepted design — that's where judges score iteration depth.

Equipment Needed Across All Three Tests

Item	Quantity	Used for
Tape measure (12-ft)	1	All protocols
Protractor or angle gauge	1	Pelican SG2 (arm angles)
Yardsticks or straight edges (24″+)	2	Pelican SG2 (boundary marks)
Painter's tape (masking-style)	1 roll	All protocols (floor marks)
Cardboard sheet (24″×24″+)	1	Skimmer (loader mock)
Phone or camera (top-down + side-view capable)	1	All protocols (EN4 evidence)
Real game elements: pins (10), cups (10), goals (3 sizes)	1 set	Skimmer + Osprey
Whiteboard or laminated tally sheet	1	All protocols (results)
EN4 notebook	1	Final transcription

💡

If real game elements aren't available yet, the Onshape Cup Walkthrough produces a 3D-printable cup placeholder, and the team can hand-cut hex pins from polycarbonate rod (~1.5″ hex stock, scaled to 1.4″–3.16″ tapered profile). The placeholders are good enough for bench testing geometry; not authoritative for inspection.

SECTION 1 / 5

Protocol 1 — Skimmer Path B Tube Catch

Tests whether the polycarb tube at Position B (arm horizontal-back, tube mouth at 12″ off tile, 3″ past chassis back edge) can reliably catch elements dropped from the SG11-raised loader chute (exit at 11.85″ off tile per Figure A9). 100 trials total across three candidate fixes; element orientations include hex-corner cases.

Why This Test Matters

Path B is the architecture's identity. If it fails, Skimmer becomes a single-feed bot (Path A roller only) and loses its strategic differentiator vs the rest of the fleet. Three candidate fixes are documented; this protocol determines which one ships.

⚠

Three candidate fixes to test (build all three; test all three before deciding):

A — Funnel cap: 3D-printed or polycarb-cut funnel scoop redirecting vertical drops into the horizontal tube. ~1.5″ deep, 3.5″ opening, throat matches tube ID.
B — Perpendicular tube mount: tube rotated 90° around arm axis so mouth points UP at Position B. Will require redesigning Position A.
C — Passive-pivot wrist: tube on a free bearing at arm tip with counterweight/spring keeping mouth pointing UP regardless of arm angle.

Setup

Loader Chute Mock

Cardboard or PVC tube, 3.5″ inner diameter, 6″ long, with bottom exit facing down
Mount to a table edge or stand so the chute exit hangs over the floor at exactly 11.85″ ± 0.1″
Per Figure A9, the SG11-raised chute exit is offset 4.02″ laterally from the rest position — mount the mock so this offset matches your bot's positioning

Skimmer Positioning

Power on Skimmer; drive the arm to Position B (arm horizontal back, tube horizontal pointing away from chassis-front)
Verify tube mouth altitude: 12″ off tile ± 0.2″
Verify tube mouth horizontal: 3″ past chassis back edge ± 0.3″
Position chassis under the loader mock so the chute exit is roughly above the tube mouth (account for the 4″ lateral offset of SG11-raised)

Bench setup — loader mock + Skimmer at Position B

Loader mock at 11.85″ off floor (per Figure A9 SG11-raised exit height). Skimmer arm at Position B with tube mouth at 12″ off floor, 3″ past chassis back edge. Element drops ~4″ from chute exit toward tube mouth. The geometry is correct; the open question is whether the horizontal tube mouth catches a vertical drop.

Procedure

For each candidate fix (A, B, C), test each element orientation 10 times. Total 100 trials per candidate, 300 trials across all three candidates.

Element Orientations

Group	Orientation	Trials
Pin	Vertical, base flare down (body-first)	10
Pin	Vertical, base flare up (neck-first)	10
Pin	Horizontal, on a hex flat	10
Pin	Horizontal, on a hex corner	10
Pin	At 45° angle (worst case)	10
Cup	Right-side-up (clear bottom)	10
Cup	Upside-down (clear top — cup is symmetric per Fig A6)	10
Cup	On its side	10
Pin-on-cup	Combo, cup right-side-up	10
Pin-on-cup	Combo, cup upside-down	10

Outcome Categories

Outcome	Definition
CATCH	Element fully enters tube; cinch pneumatic can grip without manual repositioning
MARGINAL	Element enters but rests against tube wall; requires light shake/nudge to seat (cinch may slip)
MISS	Element bounces off tube exterior, lands on floor, or sticks in the chute

Pass Criteria

CATCH rate	Verdict	Action
≥ 85%	SHIP	Fabricate the production version; mount to Skimmer; repeat 20 trials on-robot to confirm
70 – 84%	ACCEPTABLE BACKUP	Document edge cases that fail; consider as fallback if primary fix doesn't hold up under match conditions
< 70%	REJECT	Architecture-level change may be required; escalate to coach

⚠

If all three candidates fail (< 70%): the architecture may need a 5.5W active wrist motor (one additional motor) to actively rotate the tube mouth UP at Position B and back to horizontal at Position A. This is a real fallback option but adds motor budget cost (5.5W) and increases programming complexity. Coach call.

Results Template

SKIMMER PATH B CATCH RELIABILITY TEST Date: __________________ Tested by: __________________ A funnel B perpend. C passive wrist Pin vert base-down __/10 __/10 __/10 Pin vert base-up __/10 __/10 __/10 Pin horiz hex flat __/10 __/10 __/10 Pin horiz hex corner __/10 __/10 __/10 Pin 45 degrees __/10 __/10 __/10 Cup right-side-up __/10 __/10 __/10 Cup upside-down __/10 __/10 __/10 Cup on side __/10 __/10 __/10 Pin+cup right-side-up __/10 __/10 __/10 Pin+cup upside-down __/10 __/10 __/10 ───────────────────────────────────── Total CATCH __/100 __/100 __/100 CATCH % __% __% __% SELECTED FIX : ______________________________________ Rationale: __________________________________________ Rejected: ___________________________________________ Photos: A____.jpg B____.jpg C____.jpg

SECTION 2 / 5

Protocol 2 — Osprey Front-Cavity Fit on Goals

Tests whether the 6.5″ × 6.5″ chassis front cavity correctly nests around the 5.61″ footprint of all three goal sizes (alliance 3.25″, short 5.77″, tall 8.77″ per Figure A7) when Osprey drives forward into deposit position.

Why This Test Matters

Cavity geometry is what makes Osprey's mechanically-aligned scoring work without sensor-based pose estimation. If the cavity binds, the driver can't reach deposit position. If it has too much slop, the chain bar's 180° front-rest manipulator misses the goal top when the chain bar releases. Both failure modes look like "the bot doesn't score" and both come from this single geometry.

📊

Cavity nominal dimension: 6.5″ × 6.5″. Goal nominal footprint: 5.61″ × 5.61″ (Figure A7). Side clearance per edge at nominal: 0.45″. Per T5, field elements may vary by ±1.0″ from nominal — so a goal might measure 4.61″ (loose fit) or 6.61″ (no fit at all). Measure your specific test goals before running the test.

Setup

Place each goal on the floor (or workbench) at a known position
Mark goal centerline on the floor with painter's tape (4″ long, perpendicular to approach direction)
Osprey starts 6″ behind goal — mark this starting position with painter's tape (8″ long, perpendicular to approach)
Use the same goal-center mark for each trial in a set; only the chassis starting position varies (offset trials)
For tall and short goals, set them on a stable base (the goal post has a flat bottom but is top-heavy)

Procedure

For each goal type, run 5 trials at the following approach positions:

Trial 1 — Approach straight: chassis centerline aligned with goal centerline
Trial 2 — Offset 1″ left: chassis starts 1″ left of goal centerline
Trial 3 — Offset 1″ right: chassis starts 1″ right of goal centerline
Trial 4 — Offset 2″ left: chassis starts 2″ left of goal centerline
Trial 5 — Offset 2″ right: chassis starts 2″ right of goal centerline

For each trial:

Drive Osprey forward at slow speed toward the goal
Stop when the chassis "feels" the goal nest in the cavity (or binds, or misses)
Sweep the chain bar to 180° front-rest position
Record:
- Nest result: clean / binds / overshoots
- Manipulator position: distance from manipulator center to goal center (target: within ±0.5″)
- Photo: top-down photo with the chassis and goal in frame

Pass Criteria

Goal type	Pass threshold
Alliance (3.25″)	4/5 straight-approach trials nest cleanly + manipulator within ±0.5″ of goal center
Short (5.77″)	4/5 straight + 3/5 each offset (1″ tolerance is realistic during a match)
Tall (8.77″)	4/5 straight + 3/5 each offset — this is Osprey's primary target

If Binding Occurs

The cavity (6.5″) and goal footprint (5.61″) give 0.45″ side clearance per edge. If you see binding:

Diagnosis check	Likely cause	Fix
Re-measure the goal footprint	Goal may be at +0.5″ tolerance from nominal (5.61″ → 6.11″ per T5)	Note the actual measurement in EN4; design tolerance must cover the spec range
Re-measure the cavity	Aluminum frame may be at −0.3″ from nominal (CAD 6.5″ → built 6.2″)	If cavity is ≤ 6.4″, enlarge to 6.75″
Watch the chassis approach	Drive base tilts on approach (uneven wheel suspension)	Fix the drive base before re-testing the cavity

⚠

If binding persists after the above, enlarge cavity to 6.75″ × 6.75″ (0.57″ side clearance per edge). Don't go larger than 7″ — the chain bar's 9″ arm at front-rest needs the cavity's near edge to be close enough for the manipulator to drop directly over goal center.

Results Template

OSPREY FRONT-CAVITY FIT TEST Date: __________________ Tested by: __________________ Cavity dimensions (measured): ______ x ______ inches Tall goal footprint (measured): ______ x ______ inches Short goal footprint: ______ x ______ inches Alliance goal footprint: ______ x ______ inches Alliance Short Tall Straight approach _/5 _/5 _/5 Offset 1″ left _/5 _/5 _/5 Offset 1″ right _/5 _/5 _/5 Offset 2″ left _/5 _/5 _/5 Offset 2″ right _/5 _/5 _/5 ────────────────────────────── Total nests _/25 _/25 _/25 Manipulator over center _/25 _/25 _/25 Verdict: [PASS / FAIL] Cavity adjustments made: _______________________________ Re-test results (if applicable): _______________________ Photos (top-down per goal): ____________________________

SECTION 3 / 5

Protocol 3 — Pelican SG2 Envelope Verification

Measures the total horizontal envelope at 9 lift arm angles (+90° to −30° in 15° steps) to verify Pelican's 12.5″ tower + 14″ four-bar lift stays within the SG2 24″ horizontal expansion limit at every operating angle. Produces the data for the software-stop floor commit.

Why This Test Matters

SG2 is a continuous constraint — not just a starting-position check. A bot that passes inspection at rest but extends past 24″ mid-match will be disqualified during the match. The team's auton routine and driver controls must enforce a safe operating angle floor; this test produces the data to commit to that floor.

Setup

Place Pelican on the floor with chassis fully inside a marked 24″ × 24″ SG2 boundary
Mark chassis-front edge with painter's tape on the floor (this is the SG2 reference edge)
Set two yardsticks parallel to the chassis side, 24″ apart, defining the horizontal envelope
Place a protractor or angle gauge so you can read the lift arm angle directly
Phone or camera positioned for side-view photo with yardsticks visible in frame

Procedure

Power on Pelican. Drive the lift to each of the following arm angles (measured from horizontal, positive = above):

Angle	Cycle phase	Expected envelope
+90°	Fully extended vertical	≤ 22″
+75°	High deposit	≤ 22″
+60°	Typical mid-cycle deposit	≤ 22.5″
+45°	Short-goal deposit	≤ 23″
+30°	Lowest planned operating angle	≤ 23.5″
+15°	Marginal — avoid in match	≤ 24″
0° (horizontal forward)	VIOLATION CHECK	~25.5″ → FAILS
−15°	Rest descend	≤ 18″
−30°	Full rest position	≤ 18″

At each angle:

Measure the distance from chassis-front edge to the farthest forward point on the robot (claw tip, lift arm, anything)
Add the chassis depth (18″) to get the total horizontal envelope
Compare to the 24″ SG2 limit
Photograph from the side with yardsticks visible

The Expected 0° Failure

⚠

The 0° (horizontal forward) angle is expected to FAIL. This is by design — it's a violation check to confirm the team understands the constraint. The fix is software lockout in the driver code preventing the lift from going below +30°. The protocol exists to measure the violation magnitude so the team can write the lockout code with confidence.

The Software Lockout Commit

After the test, the team commits to a software stop floor angle based on the data. Two lockout layers are recommended:

Code-only lockout (primary): in the driver-control and auton loops, ignore lift-lower commands when the lift's rotation sensor reads below the floor angle. Pseudo-code: $if (lift_angle_sensor.value() <= STOP_FLOOR_ANGLE && lift_command < 0) { lift_motor.stop(); } else { lift_motor.set_velocity(lift_command); }$
Mechanical hard-stop (backup): install a physical stop on the lift pivot at the same floor angle. If the code fails or the auton routine has a bug, the lift literally cannot rotate further.

Recommended: both layers. Code as the primary protection; mechanical stop as belt-and-suspenders for inspection confidence.

Pass Criteria

Result	Verdict	Action
+30° measures ≤ 24″	PASS	Set software stop floor to +30°; implement lockout; install mechanical backup stop
+30° measures 24.0–24.5″	MARGINAL	Raise floor to +45°; implement lockout at higher angle
+45° also fails	FAIL	Arm or tower geometry too aggressive for SG2; shorten arm or move tower 2″ forward of chassis-back-edge

Verify the Lockout Works

After implementing the software stop, run this verification:

Power cycle the robot
With the controller, attempt to drive the lift below the floor angle (hold the lift-lower button at the stop)
Confirm the lift refuses to move past the stop — should hold position with motor current rising slightly but the lift not moving
Try the same from auton: write a test routine that commands the lift to −30°, run it, confirm the lift stops at the floor angle

Document the lockout verification in EN4 with screenshots of the code and the rotation sensor reading at the stop angle.

Results Template

PELICAN SG2 ENVELOPE VERIFICATION Date: __________________ Tested by: __________________ Tower height (measured): ______ inches [target: 12.5″] Arm length (measured): ______ inches [target: 14″] Claw protrusion past arm tip: ______ inches [pre-test estimate: 2-2.5″] Angle | Distance past chassis front | Total envelope | Pass/Fail +90° | _____ | _____ | ___ +75° | _____ | _____ | ___ +60° | _____ | _____ | ___ +45° | _____ | _____ | ___ +30° | _____ | _____ | ___ +15° | _____ | _____ | ___ 0° | _____ | _____ | ___ -15° | _____ | _____ | ___ -30° | _____ | _____ | ___ Lowest safe operating angle (envelope < 24″): ______ degrees Software stop floor committed: ______ degrees Lockout layers implemented: [ ] Code lockout in driver loop [ ] Code lockout in auton routines [ ] Mechanical hard-stop at pivot Lockout verification: [PASS / FAIL] Photos at each angle: ___________________________

SECTION 4 / 5

Pre-Field-Test Cross-Fleet Checklist

After bot-specific protocols pass, run this checklist on all four Phase A bots before first scrimmage. Each item is a binary check; print this section and tick boxes during pit setup on competition morning.

Inspection-Critical Items

Item	Skimmer	Pelican	Spoonbill	Osprey
R3 18″ cube at start (sizing tool test)	☐	☐	☐	☐
R4 license plates (2 opposing sides, 2–2.5″ tall)	☐	☐	☐	☐
R4 license plates screw-mounted (not rubber band)	☐	☐	☐	☐
R6 single V5 brain	☐	☐	☐	☐
R7 brain power button accessible without disassembly	☐	☐	☐	☐
R8 firmware on VEXos 1.1.5 or newer (non-beta)	☐	☐	☐	☐
R9 competition template programmed (enable/disable works)	☐	☐	☐	☐
R10 total motor power ≤ 88W (count + verify)	☐	☐	☐	☐
R11 drivetrain power ≤ 55W	☐	☐	☐	☐
R25 pneumatics: 2-tank max, 100 psi max (if applicable)	☐	n/a	☐	☐

Mechanism Verification

Item	Skimmer	Pelican	Spoonbill	Osprey
Motor port map matches port-map page	☐	☐	☐	☐
All motors run to manual command (no dead ports)	☐	☐	☐	☐
Drive forward / reverse / turn left / turn right all work	☐	☐	☐	☐
Lift / arm reaches all required positions	☐	☐	☐	☐
Manipulator (claw / tube / pincer) opens and closes	☐	☐	☐	☐
Toggle mech engages bench-toggle target	☐	☐	☐	☐
Sensor readings reasonable (IMU yaw, distance, etc.)	☐	☐	☐	☐

Programming Verification

Item	Skimmer	Pelican	Spoonbill	Osprey
Auton routine compiles without warnings	☐	☐	☐	☐
Auton routine runs to completion on a clean field setup	☐	☐	☐	☐
Auton aborts gracefully on disable signal (no runaway motors)	☐	☐	☐	☐
Driver control passes the 1-minute drive test	☐	☐	☐	☐
SG2 software lockout works (Pelican specifically)	n/a	☐	n/a	n/a
SG2 software lockout works (Spoonbill 12″ arm)	n/a	n/a	☐	n/a

Safety + Driver Awareness

Item	Skimmer	Pelican	Spoonbill	Osprey
S5 safety glasses worn during all testing	☐	☐	☐	☐
S1 no exposed sharp edges or pinch points	☐	☐	☐	☐
SG6 driver knows the 1-pin + 1-cup possession limit	☐	☐	☐	☐
Driver has practiced ditching excess possession	☐	☐	☐	☐
SG12 driver knows endgame vertical limit (18″ in midfield)	☐	☐	☐	☐

📝

Print this section. Pit-crew lead carries the printed checklist on competition morning, ticks boxes per bot, and signs at the bottom. If any item is unchecked at queue-up, the bot doesn't take the field until it's resolved. This is the team's last gate before competition.

SECTION 5 / 5

EN4 Notebook Templates

Each protocol's results should produce one engineering notebook entry. These templates are starting structures — the team rewrites the content in their own words per EN4, but the structure stays.

⚠

EN4 compliance reminder: AI-generated content is prohibited in engineering notebooks. The text in these templates is structural scaffolding only — the team rewrites every section in their own voice based on their actual test results. Submitting these templates verbatim is an EN4 violation.

Entry Template — Skimmer Path B

TITLE: Skimmer Path B catch reliability bench test DATE: __________ TEAM MEMBERS INVOLVED: ____________________________________ TEST CONDUCTED AT: ________________________________________ PROBLEM IDENTIFIED: The Skimmer v2 geometry (corrected 2026-05-10 to chassis-center tower for SG2 compliance) introduced a Position B mouth orientation issue. With the arm horizontal-back, the tube mouth points horizontal rather than vertical-up, which may not catch elements dropped vertically from the SG11-raised loader chute exit at 11.85″ off tile (per Appendix A Figure A9). PROCESS: Three candidate fixes were proposed: [list]. The team built each from scratch using polycarb / cardboard prototypes and bench-tested them against a loader chute mock positioned at 11.85″ off the floor. Each candidate was tested with 10 trials per element orientation (5 pin orientations, 3 cup orientations, 2 pin-on-cup combos) for 100 trials per candidate. RESULTS: [insert table from Section 1 results template] SELECTED FIX: ____________ RATIONALE: [Describe in your own words why this candidate passed where others failed. What specific orientations did the selected fix handle well that the others didn't? What's the simplest geometric explanation?] REJECTED ALTERNATIVES: [For each rejected candidate, describe in your own words: what was tested, what failed, what the failure mode looked like, why this informs the team's understanding even though the candidate didn't ship] NEXT STEPS: - Fabricate production version of selected fix - Mount to Skimmer; repeat 20 trials on-robot - Document on-robot test in a follow-up EN4 entry - If on-robot test confirms, mark Skimmer Phase A geometry complete LESSON LEARNED: [What did the team learn? Document in your own words.]

Entry Template — Osprey Cavity Fit

TITLE: Osprey front-cavity fit on the three goal heights DATE: __________ TEAM MEMBERS INVOLVED: ____________________________________ TEST CONDUCTED AT: ________________________________________ PROBLEM IDENTIFIED: Osprey's scoring relies on the chassis front cavity nesting around the goal at deposit position (chain bar at 180° front-rest). The cavity is built at 6.5″ \times 6.5″ nominal; the goal footprint per Figure A7 is 5.61″ \times 5.61″. Per T5, both dimensions may vary by \pm1.0″. The team needed to confirm the cavity nests cleanly on all three goal heights across typical driver approach precision. PROCESS: The team set up each of the three goal types (alliance 3.25″, short 5.77″, tall 8.77″) at a known floor position. Osprey was driven forward from 6″ behind the goal at five approach offsets: straight, \pm1″ lateral, \pm2″ lateral. Each approach was repeated 5 times for 25 trials per goal type (75 trials total). For each trial, the team recorded whether the chassis nested cleanly on the goal, and whether the chain bar at 180° front-rest placed the manipulator within \pm0.5″ of the goal center. RESULTS: [insert table from Section 2 results template] VERDICT: [PASS / FAIL] DIAGNOSIS: [If the test failed, describe in your own words what binding looked like. Was it the goal hitting the cavity edges? Was it the chassis tilting on approach? Was it variance in the goal footprint? What measurements support the diagnosis?] ADJUSTMENTS MADE: [Describe in your own words any cavity modifications, including the new dimensions and why this size was chosen] NEXT STEPS: - If passed: mark Osprey Phase A cavity geometry complete - If failed: detail the cavity modification plan and schedule a re-test - Update Osprey CAD model to reflect any cavity changes LESSON LEARNED: [What did the team learn about geometric tolerance vs nominal in the field-element specification? Document in your own words.]

Entry Template — Pelican SG2 Envelope

Linking the Three Entries

The three EN4 entries are independent but tied to a common theme: Phase A geometry validation. Consider adding a summary entry at the start of Phase B that links them:

PHASE A GEOMETRY VALIDATION SUMMARY (link entry) Three geometry questions were resolved before scrimmage 1: 1. Skimmer Path B: [selected fix] passes at __% catch rate 2. Osprey cavity: nests cleanly at __% of trials across all 3 goals 3. Pelican SG2: software stop at +__° prevents 24″ envelope violation This unblocks Phase B mechanism polish: - Skimmer: focus on cycle time optimization - Pelican: focus on driver practice within the angle limits - Osprey: focus on the chain-bar arc timing and the goal-deposit drop - Spoonbill: independently locked in 12″ arm + software stop Open Phase B questions to track: - [team's actual Phase B questions go here]