Design process

The team compared alternatives with weighted decision matrices. Early matrices assigned judgment scores; later matrices incorporated estimated cable lengths, reach timings, stakeholder opinion and counts of connections or failure points. Those inputs have different evidence strengths. A high score represents the team’s evaluation under a specified method, not measured reliability, comfort or emergency performance. Normalization arithmetic and trial discrepancies have one home in A07.

Version sequence

“Latest supported” means latest in the supplied evidence. Some filenames and internal headings conflict; not every V4 activity can be assigned an exact day. The file called “Sea Gliders Final Presentation” repeats early GA1 normalized text and notes and is not proof of a final working system. Report, scope meetings, check-ins

D01 — Control placement and input

The early comparison favored waist or “fanny-pack” controls: 86/100 against forearm 76 and dive-computer integration 54. The categories were entanglement 20, implementation 25, control settings 15, placement 25 and emergency shutoff 15. These were judgment scores, with accessibility traded against added arm wiring and the complexity of changing an existing dive computer. Early matrices

The April report changed the method and favored forearm controls. The table below preserves the report’s weights and scores.

The forearm choice traded a worse cable-length score for higher placement and implementation scores. The report’s accompanying prose assigns placement 20 points while its table assigns 25; a presentation labels the total as 100. The table has an 85-point basis; reconciled weights are not established. Later wrist development uses a magnetic throttle/input concept and a switch mechanism; T02 records the printed interface problems and T04 records electrical verification limits. Report, pp. 69–70

D02 — Propulsion mounting

Thigh, top-tank and side-tank mounting were compared. The early scores were thigh 70/80, top tank 65 and side tank 49. The revised method retained thigh mounting as the preferred wearable layout, with the scores below. Early matrices, report

The thigh option traded poorer routing/reach scores for stronger stakeholder and centerline scores. The report narrative says 59.93 instead of the table’s 59.39; the discrepancy does not change the stated ranking. This is a concept selection, not an underwater stability or release result. A separate tank demonstration later served the smaller prototype scope; it did not replace the intended wearable product with a proven tank-mounted build. D04

D03 — Battery placement and routing

The early matrix selected a single propeller-attached battery, 74/80 against a waist battery 61 and two attached batteries 50. Its criteria were entanglement 20, emergency release 15, safety 25 and complexity 20. The April normalized comparison changed both weights and inputs and favored a waist battery. Early matrices, report

SD2 reconsidered battery location around equipment interference and implementation. The side-tank option led even though the front waist option had a shorter estimated cable route.

The side-tank choice reflects the combined criteria, not a demonstrated absence of snagging or interference. A07 preserves routing assumptions; T07 defines the missing mounted-system evidence. GA2, slide 11

D04 — Hardware and demonstration scope

V1 selected CubeMars W30 for an intended pair, but the prototype bill of materials contained one kit. SD2 budget discussions led to APISQUEEN propulsion and a different battery. Exact installed model identities remain unresolved, so historical performance comparisons cannot be treated as specifications of the bench hardware. Supplier comparison, system component record

The SD2 showcase plan separated three deliverables: a one-thruster water-tank demonstration connected to control/power hardware, a wearable layout using printed stand-ins, and an empty enclosure sealing demonstration. September scope notes also proposed keeping the tank-demo battery outside the water. These are planned arrangements, not evidence that the demonstrations were completed. This separation let the team pursue propulsion, packaging and sealing within the reduced budget while retaining the full product concept. Scope meetings, GA2, slides 27–30

D05 — Enclosure and sealing architecture

SD2 enclosure concepts moved from a two-O-ring cap toward a threaded sleeve/cap with a single O-ring and a sleeve-to-tube bond. The proof-of-concept wire entry considered grouped wires and epoxy; a later product concept considered individual penetrators. These remain different concepts rather than one established sealed assembly. The threaded approach offered a removable closure and a controllable interface for print iteration. GA2, slides 13–18

The latest source descriptions and conflicting seal/material labels are in System. Manufacturing links the cap revisions, A06 explains dimensional allowances, and T03 states the missing sealing evidence. Neither a CAD label reading “waterproof” nor a thread-fit improvement establishes water exclusion or a pressure rating.

D06 — Materials and padding

The April material matrix weighted water absorption 25%, modulus 25%, compressive strength 20%, density 10% and price 20%. Its reported totals favored polymethyl methacrylate (PMMA) 78.31 and glycol-modified polyethylene terephthalate (PETG) 77.90. These are historical matrix outputs based on researched properties, not verified material ratings or proof that either was printed in the final assembly. The report’s polycarbonate (PC) total is inconsistent with its displayed category scores; the competing sources remain in provenance. Later slide labels identify polylactic acid (PLA) cap/sleeve concepts, while separate coupon notes identify PETG. Report, pp. 28–29, A06

The V4 padding matrix compared neoprene, ethylene-vinyl acetate (EVA) and ethylene propylene diene monomer (EPDM) rubber and favored neoprene. It combined diving compatibility, adaptability, durability, implementation and cost judgments:

The totals reproduce the category sums. The matrix supports a selection rationale; it does not establish padding installation, seawater durability or measured diver comfort. GA3, slide 25, T07