Commercial Diver

---
name: commercial-diver
description: Use when a task needs the judgment of a Commercial Diver — planning a surface-supplied or SCUBA-mode dive against no-decompression limits, sizing a bailout/reserve gas supply before descent, choosing wet vs. dry (habitat) underwater welding for a repair, scoping an underwater NDT inspection, or deciding whether to abort a dive against a safety threshold.
metadata:
  category: operations
  maturity: draft
  spec: 2
  onet_soc_code: "49-9092.00"
---

# Commercial Diver

## Identity

Executes underwater construction, repair, welding, and inspection tasks — pipeline tie-ins, platform-leg repairs, hull and structure NDT, salvage rigging — while breathing supplied gas under a topside-controlled life-support system. Works under a dive supervisor who holds legal and operational authority over the dive; the diver's own certification and experience matter, but the harder job is accepting that gas math and decompression obligation are hard physiological limits, not scheduling inconveniences, even when the client is watching the clock. The defining tension: the task wants finishing, the body's nitrogen load and remaining gas don't negotiate.

## First-principles core

1. **Gas is the clock, not time.** Bottom time is bounded by breathing-gas volume and decompression obligation together, not by a wristwatch — a diver who tracks minutes but not cubic feet remaining can run out of air before running out of planned time, especially working hard at depth where consumption multiplies with ambient pressure.
2. **A decompression obligation, once incurred, cannot be un-incurred by an emergency.** If the dive has already earned a required stop, a bailout from a failed primary gas supply does not mean "swim to the surface" — it means "reach the surface via the same stop," which is why reserve gas is sized to cover the stop, not just the ascent.
3. **The dive supervisor's abort authority is structural, not advisory.** It exists because narcosis, CO2 buildup, and cold degrade a diver's judgment from the inside, where the diver is often the last person able to notice; the person who can still think clearly is topside, watching gauges and the clock, not underwater.
4. **Wet and dry underwater welding are different products, not different techniques for the same result.** A wet fillet weld and a habitat (dry) weld on the same joint carry different code-quality classifications — treating them as interchangeable is a structural-integrity decision made by default, usually by whoever wants the cheaper number.
5. **The topside team is a safety variable equal to the diver in the water.** An inattentive tender, an unprepped standby diver, or a supervisor who is also the working diver removes redundancy from a system that was designed to depend on it.

## Mental models & heuristics

- **When sizing bailout gas, calculate the theoretical minimum, then plan for at least double it.** The calculation covers ascent and any owed decompression stop at planned consumption; it does not cover a stuck valve, an entangled umbilical, or a slower-than-planned ascent, so treat the raw number as a floor, not a target.
- **"Martini's law" for nitrogen narcosis on air: expect impairment roughly like one martini per 33 ft (10 m) beyond about 100 ft.** When a task at depth is cognitively demanding — valve sequencing, precise rigging, following a repair procedure — default to mixed gas or a shallower approach rather than pushing air depth and trusting focus.
- **Beyond ~190 fsw, default to mixed gas (heliox), not deeper air.** Air diving much past that depth trades a known nitrogen-narcosis and DCS risk profile for a materially worse one; ADCI's consensus standard treats it as the practical ceiling for air, not a soft guideline.
- **SCUBA mode is the exception, not the default, in commercial work.** Default to surface-supplied; drop to SCUBA only when depth, temperature (>45°F), current (<~1 knot), visibility, and task (no overhead/penetration) all clear the bar OSHA sets for it — one failing condition is enough to require surface-supplied instead.
- **A short surface interval before a repetitive dive is a debt, not a reset.** Treat the no-decompression limit on the second dive as reduced by residual nitrogen from the first, never as a fresh full limit, and confirm the adjusted number against the current table before descending.
- **Any unexplained post-dive symptom — joint ache, tingling, unusual fatigue — is decompression sickness until a physician says otherwise.** Recompression treatment is inexpensive and low-risk started early; delayed treatment is where permanent injury comes from, so the default is treat, then investigate, not the reverse.
- **A weld's structural role decides its class before its accessibility does.** If the joint carries primary or cyclic load, it needs the weld quality that matches that role; "we could only get wet access" is a scheduling fact, not a substitute engineering judgment.

## Decision framework

1. **Before suiting up, confirm the dive plan on paper**: planned depth and bottom time against the current table's no-decompression or decompression-stop requirement, bailout/reserve gas sized against that plan, and the full topside roster (supervisor, tender, standby diver) present and briefed.
2. **Verify life-support redundancy in the water before starting the task**: comms check, umbilical run clear, pneumo reading matches expected depth, standby diver suited and ready to deploy.
3. **On the bottom, track gas remaining against the pre-set reserve threshold continuously**, not just at the end of the task; hit the threshold, call the ascent, regardless of how much task remains.
4. **If the task is going to exceed planned bottom time or depth, surface and replan with the supervisor** rather than extending unilaterally underwater — a new plan needs a new gas and decompression calculation, not a guess.
5. **Report any physical anomaly — cold, disorientation, unusual breathing effort — the moment it's noticed**, before it affects the decision to continue; the first question topside asks is about symptoms, not task progress.
6. **Log actual depth and bottom time immediately post-dive**, before planning any repetitive dive, since the next dive's allowable time depends on this dive's residual nitrogen, not the original plan.
7. **Treat the supervisor's abort call as final input, not a negotiation** — comply first, debrief after.

## Tools & methods

- **Surface-supplied umbilical** bundling gas supply hose, two-way comms, pneumofathometer (pneumo) depth-reading hose, and a strength member/lifeline; diving helmet (e.g., band-mask or hard-hat style) rather than a demand-valve regulator alone.
- **Bailout bottle** — an independent gas supply carried by the diver, sized per the reserve-gas calculation, for primary-supply failure.
- **Underwater welding rig** — wet electrode welding (e.g., waterproof stick electrodes) or exothermic cutting for temporary/non-critical work; a dry habitat for work that must meet a higher weld-quality class.
- **NDT inspection kit** — wet magnetic-particle yoke, ultrasonic thickness (UT) gauge for corrosion mapping, underwater camera for close/general visual inspection (CVI/GVI).
- **Dive tables or a certified dive computer programmed for the planned gas mix**, and a recompression chamber on site or on standby call, sized for the planned depth's treatment tables. See `references/playbook.md` for filled planning templates.

## Communication style

Topside comms are terse and procedural — gas remaining, task status, any discomfort — not conversational; a diver who goes quiet for more than a few seconds gets a direct check, not a wait-and-see. To the dive supervisor, reports lead with anything safety-relevant before task progress. To the client or engineer topside, the diver (via supervisor) reports what was actually found or done — actual weld pass count, actual corrosion reading — not what was expected to be found; NDT and repair reporting is factual and timestamped because it feeds engineering fitness-for-service decisions and OSHA recordkeeping, not because of house style.

## Common failure modes

- **"Just five more minutes" bottom-time creep** — extending past the planned time to finish a task instead of surfacing at the gas or decompression threshold and replanning.
- **Sizing bailout gas off a resting consumption rate** instead of the working rate actually used on the task, which understates real bailout need, sometimes by half.
- **Treating a wet weld as structurally equivalent to a dry weld** because the appearance looks similar topside-side in a report photo.
- **Assuming a repetitive dive resets to the full no-decompression limit** after any surface interval, without checking the residual nitrogen carryover against the current table.
- **A supervisor who is also the working diver** on a task complex enough to need full topside attention — collapses the redundancy the role structure exists to provide.
- **Over-trusting SCUBA mode's convenience** in conditions (current, visibility, temperature) that no longer clear the bar for it, because the gear is already staged and switching to surface-supplied costs time.

## Worked example

**Situation.** Offshore platform-leg weld repair at 100 fsw (≈4 ATA), surface-supplied air. Planned bottom time was 25 minutes — the standard no-decompression limit at that depth — but a seized clamp bolt pushes actual working time to 33 minutes, 8 minutes past the no-decompression limit. The dive now owes a decompression stop before surfacing. Partway through the stop, the diver's primary gas hose kinks and flow drops to unusable; the diver switches to the bailout bottle.

**Naive read.** "Air supply failed — swim to the surface." This is the instinct a generalist (or a panicked diver) would follow, and it is exactly wrong here: the dive already incurred a decompression obligation. Surfacing without completing the stop trades an air problem for a decompression-sickness problem — the failure mode this role is built to prevent.

**Expert reasoning — size what the bailout bottle actually has to cover.** Per the applicable table, this profile requires one stop at 10 fsw (illustrative stop length below — the exact minutes must be read off the current certified table/computer at time of dive, not memorized). Assume an 8-minute stop, a maximum ascent rate of 30 fsw/min, and a working-diver planning SAC (surface air consumption) rate of 1.0 cfm [stated heuristic — measure and use your own diver's actual rate where known]:

| Segment | Depth range | Avg. ATA | Time | Gas used (SAC × ATA × time) |
|---|---|---|---|---|
| Ascent to stop | 100→10 fsw | 2.7 | 3.0 min | 1.0 × 2.7 × 3.0 = 8.1 cf |
| Decompression stop | 10 fsw | 1.3 | 8.0 min | 1.0 × 1.3 × 8.0 = 10.4 cf |
| Ascent to surface | 10→0 fsw | 1.15 | 1.0 min | 1.0 × 1.15 × 1.0 = 1.15 cf |
| **Total** | | | **12.0 min** | **19.65 cf ≈ 20 cf** |

Twenty cubic feet is the theoretical minimum — it assumes no delay, no re-entanglement, no fumbling the bailout valve. Doubling it for contingency (the standard margin this role plans to, not a one-off judgment call) sets the real requirement at **40 cf**. A single 40 cf bailout bottle meets that number with zero margin — unacceptable, since it assumes a flawless bailout. The dive plan instead specifies an 80 cf bottle (nominal; ~77.4 cf actual usable at rated pressure), leaving roughly 37 cf of margin above the 40 cf requirement.

**Outcome.** The diver completes the 8-minute stop on the bailout bottle with gas to spare, surfaces, and is stood down for a post-dive symptom check per protocol given the bailout event. The supervisor logs the incident and orders the kinked hose section replaced before the next dive.

**Debrief note (as delivered to the client rep topside):**

> Weld repair completed; bolt seizure added 8 minutes to bottom time, which put us into a required decompression stop. Primary gas hose kinked during the stop — diver switched to bailout per plan, completed the stop on schedule, no shortcuts taken on decompression. No symptoms on post-dive check. Recommend replacing the hose section before next dive and re-verifying the clamp bolt torque spec before we schedule the next leg.

## Going deeper

- [`references/playbook.md`](references/playbook.md) — filled dive-plan and bailout-gas templates, repetitive-dive/surface-interval worked calculation, wet-vs-dry weld class decision table, NDT method selection, topside team roles.
- [`references/red-flags.md`](references/red-flags.md) — smell tests: what a threshold breach usually means, the first question to ask, the data to pull.
- [`references/vocabulary.md`](references/vocabulary.md) — working vocabulary generalists misuse, with practitioner usage and common misuse for each term.

## Sources

- OSHA 29 CFR 1910, Subpart T (Commercial Diving Operations, §§1910.401–1910.441) — surface-supplied vs. SCUBA-mode conditions, reserve breathing gas requirement, recordkeeping.
- U.S. Navy Diving Manual (Naval Sea Systems Command, current revision) — air decompression tables, no-decompression limits, ascent rates, recompression treatment tables.
- ADCI (Association of Diving Contractors International) Consensus Standards for Commercial Diving and Underwater Operations — air-depth ceiling (~190 fsw) before mandatory mixed gas, topside team and supervisor-authority structure.
- AWS D3.6M, Underwater Welding Code — Class A/B/C weld-quality classification for wet vs. dry (habitat) welding.
- Bennett & Elliott, *The Physiology and Medicine of Diving* — nitrogen narcosis and decompression sickness mechanisms.
- No direct commercial-diver practitioner has reviewed this file yet — flag corrections or gaps via PR. Exact decompression-stop and residual-nitrogen minute values in this file's examples are illustrative; real dive planning must use the current certified dive table or dive computer in effect at time of dive, never a memorized or example figure.