17.1 Optics and Magnification

The ASNT NDT Level II visual-testing general exam lists Optics as official VT topic 5. The items are not asking you to design a microscope. They are asking whether you know what a lens and a mirror actually do, why magnification is never free, why a 10× hand lens is the default weld-visual confirmation tool rather than a 30× bench microscope, and why a reticle measurement is void the moment you change working distance or zoom.

Vision, lighting, and contrast were the previous chapter. This section is the hardware that sits between the surface and the eye — or between the surface and a remote sensor that is standing in for the eye.

Lenses and image formation

A convex (positive, converging) lens is the shop magnifier. Light from a near object is bent toward the axis. If the object sits inside the focal length, the eye sees a virtual, erect, magnified image. That is a simple loupe or hand lens. If the object sits outside the focal length, a real inverted image can be formed — the start of a compound microscope or a borescope objective.

A concave (negative, diverging) lens produces a virtual, erect, reduced image. It is a corrector or an eyepiece element, not the weld inspector's pocket lens.

A compound system (microscope, rigid borescope) uses an objective to form a real intermediate image and an eyepiece to magnify that image for the eye. The stamped magnification is the product of the stages. It is not a license to treat the image as a measuring scale.

Focal length

Focal length (f) is the distance from the optical center, or the principal plane, to the focal point. For a simple magnifier, shorter f means higher magnification.

Two textbook approximations appear in training outlines. With the image at infinity (relaxed eye), M ≈ 250 mm / f. With the image at the conventional near point of 250 mm, M ≈ 1 + 250/f. A lens with f ≈ 25 mm is nominally a 10× loupe. Those formulas explain the stamp on the barrel. They do not turn the stamp into a calibrated measuring instrument. Object distance, the inspector's accommodation, and any spacer or contact stand change the true magnification.

Mirrors

Plane mirrors give access, not power. They let you see the back of a fillet, the root of an open joint, or the far side of a stiffener. They reverse left and right. They do not magnify. A dirty, pitted, or scratched mirror is a resolution loss, the same as a filmed lens.

Concave mirrors concentrate light in a lamp reflector and, at close range, can add some magnification. Convex mirrors give a wide field and a reduced image — useful for awareness in a vessel, useless for sizing a 0.4 mm undercut.

Remote tips often carry a folding mirror or a side-view prism. That adapter changes perspective. A groove that is obvious in a side view can vanish in a forward view down a tube. Treat the adapter as part of the optic, not as a free extra.

Field of view versus magnification

This is the trade the exam extracts in one sentence: raise magnification and you shrink the field of view.

You cannot have high power and a wide field. A candidate who "just turns up the zoom" on a videoscope loses the next indication, loses the surrounding geometry that tells a toe from a scratch, and then writes a report on a postage stamp of metal.

Working distance

Working distance (WD) is the gap from the front of the optic to the in-focus object. It falls as magnification rises. A 10× Hastings-style triplet sits roughly an inch from the toe. A 30× optic is almost on the metal. A rigid borescope or videoscope has a specified WD or a focusable distal objective; the tip that is sharp at 8 mm is soft at 25 mm.

Working distance matters for three exam reasons:

1. Access and lighting. Too close and the lens collides, shadows its own light, or picks up spatter. Too far and the image is out of focus and the true resolution collapses.
2. Measurement. Optical magnification of a simple lens changes with object distance. A reticle calibrated at one WD is wrong at another.
3. Coverage. A short WD plus high mag is a tiny patch. You have not examined the weld; you have examined a spot.

Depth of field

Depth of field (DOF) is the axial range that still looks acceptably sharp. It shrinks as magnification rises and as the aperture opens (smaller f-number, more light, less DOF).

A 10× loupe focused on a weld toe will not keep the cap crown and the far HAZ equally sharp. A videoscope zoomed in on a pit will not keep the adjacent wall in focus. That is physics, not a bad inspector.

Practical rule: search at low magnification with enough DOF to see the lie of the surface; stop and confirm at higher magnification; then drop back and keep searching. Do not inspect an entire girth weld at 20×. You will miss the next indication because it is no longer in the field, and you will misname grind texture because you have no context.

Distortion and other image lies

Optics do not deliver a perfect map of the metal.

• Barrel distortion — center magnified less than the edges — is common on wide-angle videoscope tips. A straight undercut looks bowed. Length along the edge of the frame is not the same scale as length at the center.
• Pincushion distortion is the opposite bow.
• Chromatic aberration puts a color fringe on high-contrast edges. A blue-yellow fringe is not heat tint and is not rust.
• Spherical aberration and field curvature soften the edge of the field. Do not evaluate a faint line that exists only in the soft corner.
• Keystone / perspective from a shallow viewing angle makes a groove look shallower than it is and can hide a tight crack that needs a steeper look. ASME Section V, Article 9 (an industry/code access rule, not an ASNT secret) expects the eye — or the remote view that replaces it — to meet a minimum angle, commonly taught as not less than 30° to the surface. A 5° glance down a pipe is not that exam.

Distortion means you do not size from a still unless the system was calibrated at that WD, zoom, tip, and angle.

Why a 10× hand lens is common for weld visual

Several independent reasons, all exam-relevant, land on the same tool.

1. Resolution. A healthy near eye at 250 mm resolves on the order of 0.1 mm. 10× brings typical weld-toe cracks, undercut texture, and pore mouths into comfortable resolution without a bench microscope.
2. Field of view. You still see ~20 mm of toe, so you can compare the indication to adjacent ripples, spatter, and HAZ color.
3. Working distance. Short, but you can still hold a flashlight in the other hand.
4. Depth of field. Small, but a toe groove can still be held in focus.
5. Code and shop practice. AWS D1.1 visual examination is primarily unaided, with magnification used as needed and commonly limited to about 10× unless the engineer specifies more. ASME Section V, Article 9 permits optical aids for direct and remote visual. Several ASTM surface-method practices confirm with 10×. Those are industry and code facts, not an ASNT-published "only 10×" commandment.
6. Higher power is a liability. At 20× and 30×, grinding marks, weld ripples, and oxide islands start to look like cracks. FOV collapses. DOF dies. You lose the question the Level II is paid to answer: is this a fracture, or is this texture?

The Level II rule is short: search at 1× with the required lighting and access; confirm with 5–10×; do not live at 20×.

Borescope, fiberscope, and videoscope optics

Remote visual lives or dies on which image path you are using.

Common traps:

• A black speckle on a fiberscope that does not move when you rotate the scope about its axis is a broken fiber. A real pit walks around the image as the scope rotates.
• Digital zoom on a videoscope shrinks the field of view without adding optical information. It is not a 10× loupe.
• Side-view versus forward-view tips change what a groove looks like. If you only have a forward view down a header, undercut at a nozzle weld can hide.
• Illumination fall-off at the edge of a wide-angle tip hides the very undercut you were sent to find.

Calibration of measuring reticles

A reticle is a scale or grid in the eyepiece, or an overlay on the video image. It is a measuring instrument only after it is calibrated at the object plane.

Rules a Level II is expected to state:

1. Calibrate against a known scale — a stage micrometer, a certified scale, or the procedure's demonstration piece — at the same magnification, zoom, tip, and working distance you will use on the part.
2. Change WD, zoom, or tip, and the calibration is void. Recalibrate.
3. A stamped "10×" loupe with a printed scale is not a measuring instrument unless you fix the WD with a contact stand or spacer. Optical magnification of a simple lens changes with object distance.
4. On video systems, calibrate both axes if pixels are not square or if the tip is anamorphic. Do not apply a horizontal cal to a vertical depth.
5. After a battery change, a tip change, a software reboot, or a different display, re-verify.
6. Record the artifact ID, the WD, the zoom index, and the date. The procedure owns the interval — not a shop slogan.

A Level II who reports "0.8 mm undercut" from an uncalibrated videoscope overlay has not measured. They have estimated, and the estimate is not an examination record.

Realistic exam scenarios

The inspector raises videoscope zoom "to see better," then cannot find the next pore and cannot keep the toe in focus. Field of view and depth of field collapsed. Search at lower magnification; confirm on the indication; drop back.

A fiberscope shows a black dot that stays put when the scope rotates. Broken fiber, not a pit.

A reticle was calibrated at 20 mm WD and mid zoom; the undercut is then measured at 8 mm WD and full zoom. The number is invalid until the reticle is recalibrated at the new WD and zoom.

Topic 5 language is short: what the optic does, what magnification costs, why 10× is the weld-visual default, and when a reticle is a measurement versus a decoration.