16.3 Elements of Lighting

The ASNT NDT Level II visual general exam lists Elements of Lighting as official VT topic 3. Lighting is not "bring a flashlight." It is the physical input that turns a surface discontinuity into luminance contrast at the eye or camera. Topic 3 asks whether you can tell illuminance from luminance, what the inverse-square law and angle of incidence do to a reading, where the common 100 fc / 1000 lx number comes from, and why a legal lux value can still produce a useless inspection when glare, shadows, or color temperature are wrong.

ASNT does not publish a Level II lighting secret. Attribute numbers to ASME V Article 9, AWS procedures, ASTM practice, or the employer's procedure.

Illuminance versus luminance

Illuminance (E) is the luminous flux incident on a surface, per unit area. Units are lux (lx, lumens per square meter) and footcandles (fc, lumens per square foot). The conversion used in the shop is 1 fc ≈ 10.76 lx, which is why 100 fc is often written as 1000 lx in SI (the arithmetic product is about 1076 lx; codes round to the 1000 lx / 100 fc pair). Illuminance is what a visible-light meter (photometer) reads when you lay it on the weld.

Luminance (L) is the light leaving a surface toward the observer, per unit area per unit solid angle. Units are candela per square meter (cd/m², sometimes still called nits). The eye and the camera see luminance, not illuminance. Two surfaces under the same 1000 lx can look completely different: a freshly ground steel face is bright; tight mill scale or black oxide is dark. The illuminance meter can pass while the crack in the oxide still has almost no luminance contrast.

Do not swap the words. Codes and procedures specify a minimum illuminance at the examination surface. They do that because illuminance is what you can measure with a shop meter. They are not claiming that every surface at 1000 lx is equally visible.

Illuminance is a white-light / visible quantity. UV-A irradiance (μW/cm²) used for fluorescent MT and PT is a different measurement on a different meter. Topic 3 is VT lighting; do not answer a white-light VT stem with a black-light number.

Inverse-square law and angle of incidence

Inverse-square drop

A compact flashlight, an LED torch, and a small work lamp behave like point sources once you are more than a few source-diameters away. For a point source, illuminance falls with the square of distance:

E ∝ 1 / d²

Double the standoff, quarter the illuminance. Triple it, get one-ninth. That is why a torch that easily makes 4000 lx at 0.5 m is only about 1000 lx at 1.0 m and about 250 lx at 2.0 m — below the common 1000 lx minimum. Inverse-square applies to LEDs. Measuring at the lamp face and assuming the weld is equally lit is the classic fail.

Angle of incidence

Angle of incidence is the second reduction. For a given source, the illuminance on the surface scales approximately with cos θ, where θ is the angle between the incoming beam and the surface normal.

• Near-normal lighting (θ small): even illumination, good for color and stain, fills shallow grooves so they can disappear.
• Grazing lighting (θ large, beam nearly parallel to the surface): long shadows from undercut, overlap, pits, and tool marks. Excellent for texture; easy to create a shadow line that looks like a crack.
• At θ = 60° from the normal, cosine is 0.5, so the same source at the same distance delivers half the illuminance it would deliver straight-on.

Use both angles on purpose. A Level II who never moves the light is inspecting one contrast condition and calling it the whole surface.

A flood or a large diffuser is closer to an extended source. Inverse-square is less brutal because different parts of the source stay at different distances, but you still measure at the surface. Ambient shop light is an extended source that may meet 1000 lx on a bench and fail inside a nozzle.

Typical inspection illuminance

ASME V Article 9 commonly requires a minimum light intensity of 100 fc (1000 lx) at the examination surface/site for visual examination, with the light source, technique, and light-level verification demonstrated and documented. That is a method-article number, not an ASNT exam secret. A referencing procedure may require more. If the procedure says 150 fc, 100 fc is not enough.

AWS D1.1 requires visual inspection of all welds and lighting adequate to evaluate the visual acceptance criteria. Many D1.1 and shop VT procedures adopt the same 100 fc / 1000 lx surface figure used in ASME V. Do not invent a different unpublished AWS lux number. Do not treat "the bay looks bright" as a measurement.

Special cases stay special. Translucent VT uses a backlight whose luminance the procedure defines. Remote VT must light the remote surface, not the inspector's keyboard. Fluorescent MT/PT white-light limits (often ≤ 20 lux / 2 fc ambient in the booth) are the opposite problem and belong to those methods.

Glare, shadows, and color temperature

Meeting 1000 lx does not finish the lighting evaluation.

Glare is excess luminance in the field of view that reduces perceived contrast — a specular highlight on stainless, a wet surface, a freshly ground weld, or a bare LED chip imaged in a borescope window. Veiling glare lifts the dark parts of the scene so a crack and the adjacent metal both look washed. Controls: change incidence angle, use a diffuser or bounce card, shade the source, dry the surface, or move the inspector so the specular lobe does not enter the eye or lens.

Shadows hide and invent. An undercut in the umbra of a backing bar or a fitting is unlit, so it has no luminance contrast. A sharp shadow edge from a torch can be mistaken for a linear indication. The control is the same as for glare: move the light. A real groove stays put relative to the surface; a shadow moves.

Color temperature (kelvin) and color rendering index (CRI) decide whether rust, primer, heat tint, and dye look like themselves.

• Warm sources (~2700–3000 K, old incandescent or "warm white" LED) shift reds and hide some rust-on-paint pairs.
• Neutral to daylight sources (~4000–6500 K) are usually better for chromatic VT calls.
• Low-CRI sources (some shop LEDs, sodium-vapor yard lights) collapse hue differences even when lux is high.

Codes generally do not publish a mandatory kelvin number for VT. The procedure may. If it does not, choose a high-CRI, daylight-balanced task light when the call is chromatic, and do not inspect painted steel under parking-lot sodium lamps just because a meter can be made to read 1000 lx.

Flashlight versus flood versus ambient

A flashlight is a legal source if it delivers the required illuminance at the surface, at a usable angle, without disabling glare. It is not automatically adequate because it is bright in your hand.

Measure at the surface

Verify lighting with a calibrated visible-light meter placed on the examination surface, oriented to the same plane you are inspecting. Do not measure at the lamp face. Do not measure at the inspector's forehead or at the monitor. Do not estimate.

If the procedure requires a record, write the reading, the meter identity, and the location. Recheck when you change standoff, when batteries sag, when you move from cap to root, and when you switch from flood to torch. A 1000 lx demonstration on the shop floor at 07:00 does not cover a 2 m flashlight shot inside a column at 15:00.

Worked numbers the exam expects you to reason, not memorize as ASNT secrets: 4000 lx at 0.5 m from a point-like torch is ~1000 lx at 1 m and ~250 lx at 2 m. At 2 m you are below the common Article 9 minimum. Move closer, add a flood, or stop calling that surface inspected.

Lighting is a measured essential variable. A passed Jaeger card under a dying flashlight is not a visual examination.