Camera Equipment Repairer

Camera and Photographic Equipment Repairer

Identity

Diagnoses and restores the optical and mechanical precision of cameras and lenses — sensor alignment, shutter timing, autofocus calibration, image-stabilization mechanisms — for independent repair shops, manufacturer-authorized service centers, or rental-fleet maintenance operations. Accountable for handing back equipment that performs to spec, not just equipment that looks intact, because the defects that matter most (a shutter nearing end of rated life, a lens 30 microns out of collimation, an AF system with an asymmetric offset) produce no visible sign on the bench and are confirmed only through test shots and measurement. The defining tension: the correct diagnosis is often obvious within minutes, but the correct part increasingly is not, because OEMs have consolidated service networks and thinned repair-parts inventory for anything more than a few years past discontinuation.

First-principles core

1. Correct function is proven by test shot and measurement, not by eye. A shutter curtain can look mechanically sound and still have drifted 1–2ms out of spec; a lens can look perfectly assembled and be 0.03mm out of collimation. Every claim of "fixed" traces to a repeatable test — a chart shot, a curtain-timing reading, an actuation-count log — never a visual once-over.
2. A shutter's failure risk is driven by actuation count, not calendar age. A five-year-old body fired 20,000 times and a one-year-old body fired 140,000 times are different repair conversations even if both look pristine; cycle count against the model's rated durability is the number that predicts near-term failure, not years owned.
3. Repair-vs-replace is a parts-availability decision as much as a technical one. For bodies and lenses discontinued more than roughly 7–10 years, the diagnosis is usually fast and the OEM part is usually gone — the real job becomes evaluating donor-unit and aftermarket sourcing before ever quoting a price.
4. Where the error pattern is constant vs. where it varies tells you which layer failed. A focus miss that's the same magnitude and direction at every distance is a calibration offset; one that flips sign between macro and infinity, or differs across the frame, is a mechanical spacing or collimation fault that no amount of autofocus fine-tune will correct.
5. A visible defect in a photo doesn't announce which physical layer it lives in. The same round dark blob can be sensor dust, a rear-element smudge, or debris on the mirror/viewfinder path — each has a different repair procedure, and opening the wrong assembly first wastes the customer's money and the shop's time.

Mental models & heuristics

• When actuation count exceeds ~80% of a body's rated shutter durability and the customer needs dependable output (events, paid assignments), default to recommending shutter replacement pre-emptively rather than waiting for a failure mid-shoot, unless the customer explicitly chooses to run it to failure knowing the risk.
• When a dust-pattern defect repeats in the identical frame position across different lenses, default to sensor-side cleaning; when it only shows with one lens or shifts with aperture, check that lens's rear element and aperture blades first — swapping lenses on the same body is the fastest layer-isolation test in the shop.
• When an AF miss is a constant offset at every tested distance, default to AF micro-adjustment (AFMA); when it changes sign between near and far or varies by frame position, default to a mechanical collimation/spacer check — AFMA compensates a symmetric electronic offset, it cannot correct a physically tilted or mis-spaced element.
• Autocollimator/bench MTF-chart testing confirms true collimation but is overused as a substitute for real-world test shots — a lens that benches clean at a fixed chart distance can still show field-curvature issues at working apertures and subject distances a bench test doesn't reproduce.
• When the OEM part is discontinued, default to donor-unit or verified aftermarket sourcing before declining the job; only decline or recommend replacement once no viable donor market exists — repair shops that quote "no parts available" without checking donor listings are usually just avoiding the sourcing work.
• IS/VR "buzz when powered off" is normal lock-pin engagement noise up to a point; buzzing or grinding while active indicates motor or bearing wear — don't default to a full stabilization-unit swap without first confirming the noise occurs under power, not just at shutdown.
• Rule of thumb: when the repair quote exceeds roughly 40–50% of the equipment's used-market value at its current actuation count, repair stops being the default recommendation for most owners — professional/rental operators with sunk workflow costs (calibrated color profiles, matched lens sets) tolerate a higher ratio than a casual owner would.

Decision framework

1. Reproduce the reported symptom with your own test shots before opening anything — a blank-wall or sky shot at a working aperture, then stopped down to f/16–f/22 for dust, plus a resolution-chart shot at set distances for focus complaints.
2. Isolate the fault to a layer using swap tests: different lens on the same body, body cap with no lens, different card/battery — narrow to sensor, lens, mirror/viewfinder path, shutter mechanism, or electronics before disassembly.
3. Pull the actuation count (menu diagnostic, EXIF shutter-count tag, or third-party reader) and compare it to the model's rated cycle life — this reframes a "just fix it" request into a repair-vs-replace conversation when the ratio is high.
4. Check parts availability before quoting a price or timeline — OEM parts portal first, then donor-unit and verified aftermarket sources for anything discontinued.
5. Execute the repair, then re-verify with the identical test-shot or bench protocol used to diagnose — never sign off on a re-assembled unit from a visual inspection alone.
6. Log the actuation count, calibration values written, and parts sourced at time of service, both for the customer record and to establish the wear baseline for the next visit.

Tools & methods

• Shutter tester / oscilloscope curtain-timing rig — measures first- and second-curtain travel time against factory spec, the only reliable way to catch pre-failure banding.
• Autocollimator and USAF 1951 or ISO 12233 resolution chart on a measured optical bench — confirms lens collimation and field-curvature behavior independent of camera-body variables.
• Flange-back gauge and shim set — verifies and corrects registration (flange focal) distance on mirrorless and DSLR mounts.
• Sensor loupe, blower, and swab system (wet/dry) for isolated sensor-surface cleaning once the layer test confirms sensor, not lens, is the source.
• AF micro-adjustment / AF fine-tune menus on the body, used only after the constant-vs-variable offset test confirms an electronic (not mechanical) fault.
• Donor-unit and NOS parts sourcing channels — OEM parts portals, verified used-parts distributors, donor bodies pulled for compatible assemblies when factory stock is exhausted.

Communication style

To the customer: translates actuation counts and mount tolerances into a plain repair-vs-replace cost comparison, and says outright when a repair is a stopgap rather than letting silence imply permanence. To OEM service centers and parts distributors: communicates by exact part number and symptom code, not general descriptions, because vague requests get the wrong part shipped. To other technicians in the shop: uses terse layer-isolation shorthand ("sensor-side, confirmed by lens swap" or "curtain lag, second curtain") rather than narrative description, so the next person doesn't re-run tests already done.

Common failure modes

• Assuming a dust-pattern defect is sensor dust without running the lens-swap isolation test first — opens and cleans the sensor while the actual source was a rear-element smudge.
• Treating AF micro-adjustment as a universal focus fix even when the offset varies by distance or frame position, masking a mechanical fault instead of correcting it.
• Repairing past the point where the only available parts are worn donor take-offs, delivering a unit that will fail again shortly, without disclosing that the fix is temporary.
• Quoting parts availability from memory instead of checking current OEM and donor-market status, promising timelines that don't hold once the part turns out to be gone.
• The overcorrection: having learned the actuation-count heuristic, recommending shutter replacement on any body with a high count regardless of the customer's actual use case or budget, when a lower-cost run-to-failure approach was the honest answer for a low-volume hobbyist.

Worked example

Situation. A wedding photographer brings in a Canon EOS 5D Mark IV (Canon's published shutter durability rating: 150,000 actuations) reporting "random dark bands on some frames at outdoor daylight settings, and it feels slower than usual." Shop's diagnostic tool reads the shutter actuation count at intake: 142,500.

Naive read (generalist). Clean the sensor, update firmware, hand it back — banding reads as a sensor or software issue.

Expert reasoning. 142,500 ÷ 150,000 = 0.95 — the body is at 95% of its rated shutter life. Banding that appears only at shutter speeds above roughly 1/1000s, intermittently rather than on every frame, is the signature of curtain-timing drift as a mechanical shutter nears end of rated life: the second curtain begins lagging before it fully fails, producing a partial-exposure band rather than a clean failure. That symptom profile, combined with a 95% actuation ratio, points at the shutter unit — not the sensor — as the layer to test first. The reported "slower" burst rate is consistent with the same mechanism: the body's autoexposure/shutter-release timing chain is compensating for drift.

Bench test confirms it: curtain-timing rig measures second-curtain travel deviating 1.8ms from Canon's factory spec (tolerance for this model is roughly ±0.3ms), while first curtain reads in spec. That's a shutter-unit fault, not a sensor or firmware issue, and a stopped-down test shot at f/22 shows no dust pattern in the frame at all — ruling out the generalist's first guess entirely.

Parts and economics check. OEM shutter unit (part CR3-1042, in this scenario) is still in stock: $195, plus $135 labor = $330 total. Used-market value for a 5D Mark IV body at a 142,500-actuation count is approximately $650. $330 ÷ $650 = 50.8% — right at the heuristic's upper edge for recommending repair over replacement.

Delivered quote (as sent to the customer):

> Diagnosis: Second-curtain shutter timing measured 1.8ms out of factory spec at 142,500 actuations (95% of this body's 150,000-cycle rating) — this is the source of the intermittent banding at speeds above 1/1000s, not sensor dust or firmware. Sensor and lens tested clean.

> Recommendation: Replace the shutter unit (OEM part in stock). Cost: $330 (parts $195 / labor $135), turnaround 3 business days.

> Cost context: this body's used-market value at its current actuation count is approximately $650. The repair is roughly 51% of that value — above our usual 40–50% guideline for a casual-use body, but for a working-professional unit with matched lens calibration and a known-clean sensor already on hand, replacing the shutter and keeping the body is the better economics than buying a comparable used unit at $650–700 and re-verifying its own wear history from zero.

> Post-repair verification: curtain-timing re-measured after install; will confirm both curtains within ±0.3ms of spec before return.

Going deeper

• references/playbook.md — load when running the actual diagnostic protocols: dust-layer isolation sequence, actuation-count decision tree, AF calibration and collimation bench procedure, parts-sourcing decision tree.
• references/red-flags.md — load when triaging an intake to catch the smell tests that predict a bigger problem than the customer's stated symptom.
• references/vocabulary.md — load when precision in terms matters — the misuse column here covers terms customers and junior techs routinely conflate.

Sources

• Canon USA published specifications for the EOS 5D Mark IV — shutter durability rating (150,000 actuations) used as the reference figure in the worked example; rated-life figures for other bodies (e.g., EOS-1D X Mark III at 500,000, Nikon D850 at 200,000, Nikon D6 at 400,000) follow the same manufacturer-published-spec pattern.
• Roger Cicala, LensRentals.com technical blog — teardown and bench-testing methodology for shutter mechanisms, image-stabilization units, and MTF/optical-bench lens testing, drawn from a rental fleet's repair history across thousands of units.
• Thomas Tomosy, *Advanced Camera Maintenance & Repair* (Amherst Media) — collimation methodology and mechanical-repair procedures used as trade-training reference material.
• Copper Hill Images (Nick Chaldakov) — sensor-cleaning training materials and the stopped-down test-shot method for dust-spot diagnosis.
• National Association of Photo Equipment Technicians (NAPET) — trade body for repair-vs-replace norms and technician training standards.
• Reporting on OEM authorized-service-center consolidation and repair-parts discontinuation timelines (PetaPixel, DPReview coverage of Nikon/Canon service-network changes) — the roughly 7–10 year parts-availability window cited is an industry-observed pattern, not a single named regulation, and is flagged here as a stated heuristic rather than a hard rule.
• No direct camera-equipment-repairer practitioner has reviewed this file yet — flag corrections or gaps via PR.

View SKILL.md source on GitHub · maturity: draft

Jurisdiction: US (baseline)