11.1 Film, Digital Detectors, Density, and Image Quality Indicators

The ASNT NDT Level II radiographic general exam lists Darkroom Facilities, Techniques and Processing and Indication, Discontinuities and Defects as official RT topics 2 and 3. Image quality sits under both headings and under every technique item: if the radiograph cannot show the required image quality indicator (IQI) at a legal density, you do not have a valid examination to interpret.

This section is the detector-and-proof chapter. Film classes and speeds, computed radiography (CR) and digital detector arrays (DDA), optical density, the densitometer, hole-type versus wire IQIs, the commonly taught 2-2T language, and the triad of contrast, definition, and noise are how you demonstrate that the image was capable. Product-form discontinuities and radiation safety belong in the next chapter. Darkroom chemistry and artifacts belong in 11.2. Here the job is to know what the detector recorded and whether that record is good enough.

ASME Section V, Article 2, Radiographic Examination, is the usual U.S. construction-code process article for film radiography of welds and components. ASTM E94/E94M is the general radiographic guide. ASTM E1742/E1742M is the widely cited practice for radiographic examination (often aerospace and precision hardware). ASTM E1025 covers hole-type IQIs. ASTM E747 covers wire IQIs. ASTM E1815 classifies industrial film systems. CR and DDA practices live in documents such as ASTM E2033, E2445, E2597, E2698, and E2737, and in the digital appendices ASME Section V publishes for radiographic examination. Those documents are industry and code references, not ASNT exam publications. Attribute every numeric window to the standard or to the employer's written procedure. If a procedure tightens a value, the procedure wins.

Film as a detector: classes, speed, and screens

Industrial radiographic film is a silver-halide emulsion, usually coated on both sides of a polyester base and used with lead intensifying screens (or, when the procedure names it, without screens or with fluorescent screens). Photons that reach the emulsion form a latent image. Wet processing converts that latent image into metallic silver. The more radiation that reached a given area, the darker that area is on the processed negative.

A film system is not "the box on the shelf." ASTM E1815 classifies the system — film plus screens plus processing — into classes commonly taught as Special / I / II / III (and a W class in the standard's own table). The practical teaching is:

Speed is how much exposure the system needs to reach a stated density. Faster is not automatically better. A fast, grainy system can hide a tight crack that a slower Class I system would have resolved. A slow, fine-grain system that forces a long gamma exposure on a scaffold may invite motion unsharpness that destroys the definition you thought you bought. Match the class to the procedure, the energy, and the geometric unsharpness already set in Chapter 10.

Lead screens do two jobs: they emit photoelectrons that intensify the image (especially at higher energies), and they filter scatter. Front-screen and back-screen thicknesses are procedure and energy choices, not an unpublished ASNT number. Dirty, scratched, or poorly contacting screens print as artifacts (11.2) and as a loss of definition. Fluorescent screens, when a procedure allows them, raise speed dramatically and usually hurt definition. Do not substitute them for lead on a critical weld because the exposure clock is running.

Film storage is part of image quality. Heat, humidity, background radiation, and expired emulsion raise base-plus-fog. Fog eats contrast before you ever expose the part. Keep boxes on edge, away from process radiation and heat, and inside the manufacturer's expiration window.

Digital detectors: CR and DDA

The Level II general exam still expects film vocabulary, but digital detectors are now ordinary shop tools. Two families matter.

Computed radiography (CR) uses a photostimulable phosphor imaging plate (IP) in a cassette. The plate stores a latent image. A scanner's laser reads it, software builds a digital radiograph, and an eraser prepares the plate for reuse. CR keeps much of the film workflow — flexible cassettes, lead screens, the same source-side IQI rules — but the "density" you used to read with a densitometer becomes a gray value controlled by scan settings, photomultiplier gain, and display windowing.

Digital detector arrays (DDA), often called flat panels, convert the beam to a digital image without a storage plate. Indirect panels use a scintillator plus a photodiode array. Direct panels convert x-rays in a semiconductor layer. DDAs enable short exposures, frames that can be averaged to raise signal-to-noise ratio (SNR), and real-time or near-real-time alignment. They also introduce dead pixels, gain/offset nonuniformity, and ghosting — artifacts Section 11.2 treats as the digital cousins of processor marks.

Digital image quality is not "the IQI looks good on my monitor." Industry practices (ASTM E2445 for CR, E2737 / E2597 for DDA, and ASME Section V's digital radiographic appendices) ask for pixel size, spatial resolution, SNR or normalized SNR, a bad-pixel map, and a quality-control routine. Windowing a 16-bit image can make a mediocre exposure look crisp; that is display contrast, not radiographic contrast. The procedure, not the brightness slider, decides whether the image is acceptable.

Do not treat CR or DDA as a free pass on geometry. Geometric unsharpness, scatter, and energy selection still set subject contrast. A 200 µm pixel cannot invent a 50 µm crack that the beam and the part physics never delivered. Incomplete erasure of a CR plate, a stale DDA flat-field, or a cluster of bad pixels in the weld is the same class of problem as a fogged film: the detector is no longer a valid recorder.

Optical density and the densitometer

Optical density (D) on film is log10 of incident viewer light divided by transmitted light. Density 1.0 transmits 10% of the light, 2.0 transmits 1%, 3.0 transmits 0.1%, and 4.0 transmits 0.01%. Higher density means a darker radiograph.

Many ASME Section V, Article 2 film techniques require the transmitted density through the area of interest (and through the body of the designated hole-type IQI) to fall in a window of about 1.8 to 4.0 for single-film viewing when the source is an X-ray tube, and 2.0 to 4.0 when the source is gamma — the maximum is 4.0 either way, and composite multi-film viewing carries its own lower per-film minimum. Article 2 commonly prints a 1.8 minimum for x-ray exposures, a 2.0 minimum for gamma-ray exposures, a 1.3 minimum per film when several films are viewed as a composite set, and a 4.0 maximum for either single or composite viewing. Those are code numbers, not ASNT secrets. A procedure may raise the minimum, lower the maximum, or require a tighter band through the weld. The procedure wins.

Why the window exists:

• Too low (under-exposed or under-developed): you are on the toe of the characteristic (H&D) curve. Subject contrast is wasted. Thin sections and IQI holes wash out.
• Too high (over-exposed or over-developed): the illuminator cannot push enough light through the ROI. The image is unreadable even if the silver is there. Article 2's 4.0 cap is a viewing limit as much as a physics limit.
• Uneven density across a weld of changing thickness may require a shim, a filter, a multiple-film technique, or a second exposure. Do not "fix" a 1.2 weld-root density by turning up the viewer and hoping.

A densitometer is the calibrated instrument that measures D. Measure through the area of interest and through the IQI body, not through a random dark logo or a lead letter. Zero the instrument as the manufacturer requires, verify it on a NIST-traceable density strip, and treat a densitometer that has not been checked as an uncontrolled gauge. Step-wedge or plaque readings elsewhere on the film are process-control data; they do not replace a reading in the ROI.

Digital images do not have optical density in the film sense. Do not report "density 2.3" on a DDA frame unless the procedure defines an equivalent gray-value or optical-density conversion and you measured it that way.

Hole-type versus wire IQIs

An IQI (historically a penetrameter) is a known absorber placed on the part so the radiograph proves a stated sensitivity. Seeing the IQI does not mean a crack of the same size will be seen. It means the combination of energy, geometry, scatter control, detector, and processing produced enough contrast and definition to show that known, high-contrast detail.

Hole-type (plaque) IQIs are specified by ASTM E1025. A rectangular or circular plaque of material similar to the part has a stated thickness and three holes whose diameters are 1T, 2T, and 4T — one, two, and four times the plaque thickness. The plaque thickness is chosen as a percentage of the nominal specimen thickness in the region of interest. Common designations:

Wire IQIs are specified by ASTM E747. A set of wires (sets A, B, C, D in the standard) of graded diameters lies across the area of interest. The procedure names the essential wire. Visibility of that wire is the sensitivity proof.

Hole-type and wire systems are not freely interchangeable. Conversion tables exist in the codes, but the procedure says which family to use and which essential hole or wire must be visible. ASTM E94, ASTM E1742, and ASME V Article 2 all allow either family when the document's own tables are followed.

Placement rules the written exam loves:

• Place the IQI on the source side whenever the procedure requires source-side placement — that is the more conservative location because the IQI image includes geometric unsharpness.
• Film-side (detector-side) placement, when allowed, is a degraded demonstration; codes often require identification (a lead F or equivalent) and may require a thinner essential IQI.
• Put the IQI in the area of interest or on a block of similar material and thickness, not on a handy nameplate across the room.
• If weld reinforcement makes the plaque rock, use shims of like material so the IQI sits on the equivalent thickness the code describes.
• The IQI material must be radiographically similar to the part (same alloy family). A steel plaque on titanium is not a valid sensitivity proof.

2-2T sensitivity language

2-2T is classroom and procedure shorthand, not a mystical ASNT constant. It means:

1. Use a hole-type IQI whose thickness is 2% of the specimen thickness (2T IQI), and
2. The 2T hole must be visible on the radiograph.

That combination is the most commonly taught industrial sensitivity statement. ASME V Article 2, ASTM E94, and ASTM E1742 do not all require 2-2T on every shot. Many construction-code tables step the essential hole or wire with thickness. Aerospace work often requires 2-1T or 2-2T or better. Read the table in the governing document.

What 2-2T is not:

• It is not a promise that a 2%-of-thickness crack will be seen. A crack is a planar, often tight opening. A hole is a high-contrast drilled void.
• It is not a density number.
• It is not satisfied by seeing the 4T hole and squinting at the 2T hole.
• It is not satisfied by windowing a digital display until a wire "appears."

If the stem says the procedure requires 2-2T and the 2T hole is not discernible in the ROI at a legal density, the radiograph is unacceptable for interpretation. Reshoot. Do not interpret "around" a missing IQI.

Contrast, definition, and noise

Image quality is three separable ideas. Mixing them is how candidates miss technique items.

Radiographic contrast is the density (or gray-value) difference between adjacent areas. It is the product of subject contrast and detector contrast.

• Subject contrast comes from the part: thickness change, density, atomic number, and beam energy. Lower energy generally increases subject contrast and decreases penetration. Higher kV or a harder gamma spectrum penetrates more and flattens subject contrast. Scatter reaching the detector also flattens contrast — that is why collimation, lead screens, masks, and air gaps matter.
• Film contrast is the slope (gamma) of the characteristic curve in the density band you actually use. That is one reason the 1.8–4.0 window exists: it is where industrial films are designed to be steep. Under-exposed film sits on the toe and loses contrast even if the part had subject contrast.

Definition (sharpness) is how abruptly density changes at an edge. It is limited by geometric unsharpness Ug (source size, source-to-object distance, object-to-detector distance — Chapter 10), by inherent detector unsharpness (grain, screen contact, pixel size), and by motion. A high-contrast blot with a fuzzy edge will not resolve a tight crack.

Noise is random or structured fluctuation that is not the part. On film it is grain (and screen mottle). On CR/DDA it is quantum noise, electronic noise, and residual patterns after a poor gain/offset correction. Averaging DDA frames raises SNR; using a faster, grainier film lowers it.

Sensitivity — the ability to see a small detail — needs enough contrast, enough definition, and a low enough noise floor. The IQI is a packaged check of that triad for a high-contrast known detail. A radiograph can show a crisp 2T hole and still miss a tight, misoriented crack (11.3). A radiograph that cannot show the essential IQI is not ready for that argument.

Realistic exam scenarios

A single-film x-ray weld radiograph measures 1.4 in the weld and 2.1 on the IQI body sitting on a shim beside the weld. ASME V Article 2 style density is not met in the area of interest. The shot is not acceptable because the IQI happened to be dense enough. Correct exposure, a multiple-film technique, or a filter — then re-read density in the ROI.

A procedure calls for a 2-2T hole-type demonstration. The Level II loads an ASTM E747 wire set because "wires are more modern." Unless the procedure's conversion table and essential-wire line are followed, that is the wrong IQI family. Visibility of some wire does not automatically equal 2-2T.

A DDA image is windowed until the weld looks punchy, but the SNR and bad-pixel checks required by the digital practice were never run. Display contrast is not image quality. The essential IQI and the digital quality metrics both have to pass.

A Class III film system is substituted for a specified Class I system to shorten a cobalt-60 shot. Speed improves; grain and definition usually get worse. If the procedure named the class, the substitution is a technique change, not a convenience.

Image quality is a proof, not a vibe. Density in the ROI, the correct IQI in the correct place, and a detector whose contrast, definition, and noise can support the essential detail — then you have something to interpret.