13.3 Radiographic Physics, Geometric Unsharpness, Density & IQI Sensitivity

Key Takeaways

  • Industrial radiography relies on differential radiation absorption governed by the Beer-Lambert attenuation law (I = I0 e^(-μt)); radioisotopes (Iridium-192, Cobalt-60, Selenium-75) offer portable monoenergetic gamma emission, while X-ray machines provide controllable continuous Bremsstrahlung spectra.
  • Radiographic film optical density (D = log10(I0 / It)) must be maintained within tight code windows: 1.8 to 4.0 for X-ray radiographs and 2.0 to 4.0 for gamma radiographs per ASME Section V and AWS D1.1, with density across the weld area not varying beyond -15% to +30% of the IQI body density.
  • Image Quality Indicators (IQIs) verify radiographic sensitivity; standard inspection demands 2-2T sensitivity for hole-type plaques (resolving the 2T hole on a plaque whose thickness is 2% of specimen thickness) or equivalent essential wire resolution under ASTM E747 / ISO 19232.
Last updated: September 2026

13.2 Volumetric Radiography (RT): Geometric Unsharpness & Film/Digital Imaging

Quick Answer: Industrial Radiographic Testing (RT) evaluates internal volumetric soundess by directing ionizing radiation through a weldment onto photographic film or a digital detector. Differential absorption follows the Beer-Lambert law ($I = I_0 e^{-\mu t}$). Image penumbra is controlled by the geometric unsharpness formula ($U_g = \frac{f \cdot d}{D}$), where code limits mandate $U_g \le 0.020\text{ in}$ ($0.51\text{ mm}$) for steel under $2\text{ in}$ ($50\text{ mm}$). Optical density must fall between $1.8\text{ and }4.0$ for X-ray and $2.0\text{ and }4.0$ for gamma rays. Image Quality Indicators (hole-type plaques or wire IQIs) verify standard 2-2T sensitivity. While volumetric flaws (porosity, slag, tungsten) are readily identified, planar lack of fusion and cracks require beam alignment within $\pm 5^\circ\text{ to }10^\circ$ of the flaw plane.


1. Physics of Industrial Radiography: Sources, Spectra & Attenuation

Radiographic examination relies on high-energy electromagnetic photons possessing wavelengths between $10^{-4}\text{ and }10\text{ nm}$. As radiation penetrates steel, photons undergo absorption and scattering primarily through the Photoelectric Effect (predominant at low energies, $< 100\text{ keV}$), Compton Scattering (predominant across industrial welding energies, $100\text{ keV to }5\text{ MeV}$), and Pair Production (occurring at high energies, $> 1.022\text{ MeV}$).

                    VOLUMETRIC RADIOGRAPHIC EXPOSURE GEOMETRY

                               Source / Focal Spot (f)
                                      +-----+
                                      |  *  |
                                      +--+--+
                                         |  \
                                         |   \
                    Source-to-Object     |    \
                     Distance (SOD = D)  |     \
                                         |      \
                                         V       V
                     ====================[===WELD===]==================== Plate
                     ====================================================
                     Object-to-Film (d)  |        |
                     --------------------+--------+----------------------
                                         [==FILM==]
                                   Source-to-Film Distance
                                       (SFD = D + d)

X-Ray Machines vs. Radioisotope Gamma Sources

  • Industrial X-Ray Generators: X-rays are produced when high-velocity electrons accelerated across an electrical potential ($kV_p$) strike a heavy metal target (tungsten). The deceleration produces a continuous Bremsstrahlung spectrum superimposed with sharp characteristic emission lines. The tube voltage ($kV_p$) dictates photon energy (penetrating power), while tube current ($mA$) and exposure time ($s$) dictate total photon flux (intensity).
  • Gamma-Ray Radioisotopes: Gamma rays originate from the spontaneous nuclear decay of unstable radioactive isotopes, yielding discrete, monochromatic photon energy lines.
Radiation SourcePhysical Form & Half-Life ($t_{1/2}$)Principal Photon EnergiesSteel Thickness Inspection RangeHalf-Value Layer (HVL) in Steel
Iridium-192 ($^{192}\text{Ir}$)Metallic pellet; $t_{1/2} = 73.83\text{ days}$Multi-line spectrum: $0.21\text{ to }0.61\text{ MeV}$ (average $\approx 0.38\text{ MeV}$)$0.25\text{ to }2.5\text{ in}$ ($6.4\text{ to }65\text{ mm}$)$\approx 0.50\text{ in}$ ($12.7\text{ mm}$)
Cobalt-60 ($^{60}\text{Co}$)Metallic pellet; $t_{1/2} = 5.27\text{ years}$Two distinct lines: $1.17\text{ MeV}$ and $1.33\text{ MeV}$ (average $1.25\text{ MeV}$)$1.5\text{ to }7.0\text{ in}$ ($38\text{ to }175\text{ mm}$)$\approx 0.85\text{ in}$ ($21.6\text{ mm}$)
Selenium-75 ($^{75}\text{Se}$)Solid salt pellet; $t_{1/2} = 119.8\text{ days}$Multi-line: $0.066\text{ to }0.40\text{ MeV}$ (average $\approx 0.22\text{ MeV}$)$0.15\text{ to }1.0\text{ in}$ ($4.0\text{ to }25\text{ mm}$)$\approx 0.30\text{ in}$ ($7.6\text{ mm}$)
Industrial X-Ray (300 kVp)Vacuum tube; continuous electrical operationContinuous spectrum up to $300\text{ keV}$ peakUp to $2.0\text{ in}$ ($50\text{ mm}$)Variable with filtration; $\approx 0.25\text{ in}$ ($6.4\text{ mm}$)

Attenuation Mathematics: Beer-Lambert Law, HVL & TVL

Radiation passing through matter attenuates exponentially according to the Beer-Lambert Law:

I=I0eμtI = I_0 e^{-\mu t}

where $I_0$ is the incident beam intensity, $I$ is the transmitted beam intensity, $\mu$ is the linear attenuation coefficient of the material ($\text{cm}^{-1}$ or $\text{in}^{-1}$), and $t$ is the material thickness.

  • Half-Value Layer (HVL): The thickness of absorber required to reduce the beam intensity to exactly $50%$ ($1/2$) of its initial value: II0=0.5=eμHVL    HVL=ln(2)μ=0.69315μ\frac{I}{I_0} = 0.5 = e^{-\mu \cdot HVL} \implies HVL = \frac{\ln(2)}{\mu} = \frac{0.69315}{\mu}
  • Tenth-Value Layer (TVL): The thickness of absorber required to attenuate the beam intensity to $10%$ ($1/10$): TVL=ln(10)μ=2.3026μ3.32×HVLTVL = \frac{\ln(10)}{\mu} = \frac{2.3026}{\mu} \approx 3.32 \times HVL

2. Geometric Unsharpness ($U_g$) Mechanics & Code Limits

Because radiation sources possess finite physical dimensions (focal spot width $f$ in an X-ray tube, or the physical diameter of a radioactive isotope pellet), rays originate from multiple points across the source face. This geometric dispersion casts a blurred shadow boundary—known as the geometric penumbra or geometric unsharpness ($U_g$)—along every feature edge on the film.

                         GEOMETRIC UNSHARPNESS MECHANICS

                            Source Diameter / Spot (f)
                                 |<- f ->|
                                 +-------+
                                /         \
                               /           \
                              /             \
          Source-to-Object   /               \
           Distance (SOD = D)/                 \
                            /                   \
                           /                     \
                          /   Object / Defect     \
     --------------------+------[=======]----------+--------------------
     Weldment Plate             |<--w-->|
     Object-to-Film (d)  /                 \
     -------------------+-------------------+---------------------------
                        |<--Ug->|           |<--Ug->|
                         Blur    Full Shadow Blur
                              Film Plane / Detector

Governing Formula

From similar triangles across the optical projection:

Ug=fdDU_g = \frac{f \cdot d}{D}

where:

  • $U_g$ = Geometric unsharpness (penumbra width on film)
  • $f$ = Source size (effective focal spot dimension of X-ray tube or active diameter of radioisotope capsule)
  • $d$ = Object-to-film distance (OFD, distance from the source-side surface of the weldment to the recording media)
  • $D$ = Source-to-object distance (SOD, distance from radiation source to the source-side surface of the weldment)
  • Note: Source-to-Film Distance is $SFD = D + d = SOD + OFD$.

To minimize geometric unsharpness ($U_g$), the welding engineer must:

  1. Select the smallest practical source/focal spot size ($f$).
  2. Minimize object-to-film distance ($d$) by placing film cassettes in direct contact with the weld root/face.
  3. Maximize source-to-object distance ($D$), balancing penumbra reduction against the Inverse Square Law ($I \propto 1 / SFD^2$), which increases exposure time quadratically.

Code Acceptance Limits for Geometric Unsharpness

ASME BPVC Section V Article 2 (Table T-274.2) and AWS D1.1 Clause 8 impose maximum permissible $U_g$ thresholds based on material thickness:

Nominal Material Thickness RangeMaximum Permissible Geometric Unsharpness ($U_g$)
Under $2.0\text{ in}$ ($< 50\text{ mm}$)$0.020\text{ in}$ ($0.51\text{ mm}$)
$2.0\text{ to }3.0\text{ in}$ ($50\text{ to }75\text{ mm}$)$0.030\text{ in}$ ($0.76\text{ mm}$)
Over $3.0\text{ to }4.0\text{ in}$ ($75\text{ to }100\text{ mm}$)$0.040\text{ in}$ ($1.02\text{ mm}$)
Greater than $4.0\text{ in}$ ($> 100\text{ mm}$)$0.070\text{ in}$ ($1.78\text{ mm}$)

3. Radiographic Sensitometry, Density & Film Viewing

Optical Density Formulation

Radiographic film consists of a polyester base coated with a photographic emulsion of silver halide crystals ($AgBr$). Photons ionize halide atoms, creating a latent image reduced to metallic silver grains during chemical development. The blackening of the processed radiograph is quantified as Optical Density ($D$):

D=log10(I0It)D = \log_{10}\left(\frac{I_0}{I_t}\right)

where $I_0$ is the light intensity incident on the film from a calibrated industrial viewer, and $I_t$ is the light intensity transmitted through the radiograph, measured via a calibrated transmission densitometer.

  • $D = 1.0 \implies 10%$ of light transmitted ($10^{-1}$).
  • $D = 2.0 \implies 1%$ of light transmitted ($10^{-2}$).
  • $D = 3.0 \implies 0.1%$ of light transmitted ($10^{-3}$).
  • $D = 4.0 \implies 0.01%$ of light transmitted ($10^{-4}$).
                       HURTER & DRIFFIELD (H&D) CURVE

     Optical Density (D)
       ^
   4.0 |                                       .--- Shoulder (Saturation)
       |                                      /
   3.0 |                                    / <-- Linear Region
       |                                   /      (Maximum Film Gradient & Contrast)
   2.0 |                                 / 
       |                    .-----------' <-- Toe (Under-exposed, low contrast)
   1.0 |                   /
       |   _______________/
       +---+---+---+---+---+---+---+---+---+---> Log Relative Exposure

Code Optical Density Boundaries (ASME Section V / AWS D1.1)

  • Single Film Viewing:
    • X-Ray Radiographs: Density must be within $1.8 \le D \le 4.0$ throughout the area of interest.
    • Gamma-Ray Radiographs: Density must be within $2.0 \le D \le 4.0$ throughout the area of interest.
  • Composite Viewing of Double Film Exposures: Density must be within $2.6 \le D \le 4.0$, with neither individual film falling below $1.3$.
  • Density Variation Limitations: The optical density across the weldment area of interest must not vary by more than $-15%$ to $+30%$ from the density measured through the body of the Image Quality Indicator (IQI).

4. Image Quality Indicators (IQI / Penetrameters) & Sensitivity Calibration

Radiographic sensitivity measures the smallest discontinuity that can be discerned on the film. Sensitivity is verified using Image Quality Indicators (IQIs) positioned on the weldment during exposure.

       HOLE-TYPE PLAQUE (ASTM E1025)             WIRE-TYPE IQI (ASTM E747)

          T = 2% of Specimen Thickness                 6 Graduated Wires
         +---------------------------+              +-----------------------+
         | (1T)    (2T)        (4T)  |              | |   |   |   |   |   | |
         |  o       O           ( )  |              | |   |   |   |   |   | |
         |     Lead ID: "25"         |              | Lead ID: "SET B"      |
         +---------------------------+              +-----------------------+
              Hole Diameters:                           Essential Wire:
             1T, 2T, 4T Plaque Thick.                   Must be Clearly Visible

Plaque-Type vs. Wire-Type IQI Systems

  1. Hole-Type Plaques (ASTM E1025): Rectangular metal strips manufactured from radiographically similar alloys. The plaque thickness ($T$) is typically fabricated to equal $2%$ of nominal weldment thickness. The plaque contains three drilled holes with diameters proportional to plaque thickness: $1T$, $2T$, and $4T$.
    • Standard Radiographic Sensitivity (2-2T): The radiograph must clearly display the $2%$ thick plaque and resolve the $2T$ hole (whose diameter equals twice the plaque thickness, $d = 2T$). Resolving the smaller $1T$ hole represents higher sensitivity ($2-1T$).
  2. Wire-Type IQIs (ASTM E747 / ISO 19232-1): Six plastic-encapsulated parallel wires of graduated diameters (Sets A, B, C, and D). ASTM E747 Table 4 establishes the mandatory Essential Wire Number that must be resolved across the weld area based on joint thickness.

Source-Side vs. Film-Side Placement Rules

  • Standard Source-Side Placement: Codes mandate placing the IQI on the source side of the joint (the side facing the radiation source), aligned parallel to and within $1/8\text{ in}$ ($3.2\text{ mm}$) of the weld toe.
  • Film-Side Placement: When physical geometry precludes source-side access (e.g., closed pressure vessels, unpiggable pipelines), the IQI is placed on the film side. To warn the interpreter of lower sensitivity due to reduced penumbra magnification, a lead letter "F" at least $1/4\text{ in}$ ($6.4\text{ mm}$) high must be placed immediately adjacent to the IQI on the film cassette.
  • Weld Reinforcement Shims: When a weld possesses high external crown reinforcement, the base metal under the IQI is thinner than the total weld cross-section. A radiographically similar compensating shim must be placed beneath the IQI to elevate it to the total thickness of the weld plus allowable reinforcement.

Test Your Knowledge

Why does Radiographic Testing (RT) exhibit a notoriously low probability of detection (POD) for planar lack of sidewall fusion in heavy-wall narrow-groove weldments compared to Ultrasonic Testing (UT)?

A
B
C
D