UT Attenuation Calculator
Conventional UT loses roughly 20 dB through 300 mm of carbon steel at 5 MHz — half from material absorption, half from beam spreading. Push that to 10 MHz on the same path and total loss climbs to 38 dB, which is why thick-wall and coarse-grain forgings sit at 2.25 MHz or lower. This calculator separates the two loss mechanisms so you can size the gain budget on a real UT instrument (typically 60–100 dB dynamic range) before mobilising to the field.
How it works
Total back-and-forth loss = 2 × distance × α(f, material) + 20×log₁₀(distance/N) once past the near-field, where α is the material attenuation coefficient in dB/mm and the second term is geometric spreading from the diverging beam. α is itself a function of frequency: in carbon steel α(f) ≈ 0.02×f for longitudinal waves and 0.04×f for shear waves (dB/mm, f in MHz). Austenitic stainless and coarse-grain materials have α 3–10× higher because of grain-boundary scattering. Effective range is the path length where total loss equals the typical 40 dB working budget (above the noise floor on a calibrated DAC).
Formula
L_total = 2αx + 20 log₁₀(x/N) ; α_steel(L) ≈ 0.02f ; α_steel(S) ≈ 0.04f [dB/mm, f in MHz]
L_total = 2αx + 20 log₁₀(x/N) ; α_steel(L) ≈ 0.02f ; α_steel(S) ≈ 0.04f [dB/mm, f in MHz]Worked example
5 MHz longitudinal in carbon steel, 100 mm sound path (50 mm wall, pulse-echo). α = 0.02 × 5 = 0.10 dB/mm. Material absorption (both ways) = 2 × 100 × 0.10 = 20 dB. Near-field of a typical 10 mm probe at 5 MHz is ~21 mm; at 100 mm we are at 5N, geometric spreading = 20×log₁₀(100/21) = 13.6 dB. But geometric loss for back-wall echo is largely cancelled by the back-wall acting as a reflector of the full beam — so practical loss on a back-wall echo at 50 mm wall ≈ 2αx ≈ 20 dB plus a few dB couplant + reflector loss = 23 dB total. Effective range (40 dB budget): solve 2αx = 40 → x = 200 mm one-way = 400 mm sound path; subtract 50 dB for an FBH-sized indication reflector and effective range drops to ~350 mm. This is why 5 MHz works on 300 mm steel and 10 MHz does not.
| Variable | Value |
|---|---|
| input: material | Carbon steel |
| input: frequency | 5 |
| input: distance | 100 |
| input: beamMode | Longitudinal |
| output: totalLoss | 23 |
| output: attenCoeff | 0.10 |
| output: geometricLoss | 13 |
| output: effectiveRange | 350 |
When to use this tool
Use when selecting probe frequency for thick-wall components (typically going DOWN to 2.25 MHz or 1 MHz for forgings >150 mm), when sizing the gain budget for a long sound-path inspection (austenitic SS welds, copper castings, large forgings), when explaining to a procurement reviewer why a low-frequency probe was specified, or when comparing two material lots that show different attenuation on the same calibration block.
Limitations
Where this calculator stops being accurate:
- Coefficients are typical values for fine-grained carbon steel and aluminum. Cast iron, austenitic stainless (especially weld metal), Inconel, and coarse-grain forgings can have α 3–10× higher and must be measured on the actual material, not estimated.
- Frequency exponent for α varies by material: carbon steel ≈ linear in f, austenitic SS often closer to f² (grain scattering dominates), polymers can be f^1.5.
- Temperature changes α materially — austenitic SS attenuation drops ~25% from 20°C to 200°C, which matters for in-service hot-piping UT.
- Geometric spreading assumes a piston transducer in far-field. Focused probes converge then diverge; PA arrays follow their focal plan. Use manufacturer beam plots for non-flat probes.
- Does NOT account for couplant transmission loss (typically 6–12 dB on contact UT, near-zero on immersion).
- Does NOT account for surface roughness loss (rough as-rolled or as-cast surface can add another 10–20 dB on contact UT — use immersion or buffer-coupling for rough surfaces).
Frequently Asked Questions
Why does shear-wave attenuation roughly double the longitudinal value?
In polycrystalline metals, shear waves interact more strongly with grain boundaries because their displacement is perpendicular to propagation (so grain anisotropy matters more). Across most engineering metals, shear α is 1.8–2.5× longitudinal α at the same frequency. This is why angle-beam shear-wave inspection on austenitic SS welds frequently uses 2.25 MHz when longitudinal would have used 5 MHz — the shear-mode attenuation forces a lower frequency to keep enough signal-to-noise across the weld leg.
How do I measure α on a piece of material I have no data for?
Use the back-wall echo decay method per EN 12668-2. Place a 0° contact probe on a flat parallel-sided sample. Record amplitude of the 1st and 2nd back-wall echoes (B1 and B2). α = (20×log₁₀(B1/B2)) / (2 × thickness). Repeat at 3 different thicknesses to confirm linearity. Use the same probe and gain as the field inspection. This gives the in-service attenuation including roughness, couplant, and grain scattering — far more useful than a literature value for that alloy grade.
Why does this calculator separate material absorption from geometric spreading?
Because they have different fixes. If material absorption dominates (low frequency helps), going from 5 → 2.25 MHz cuts loss roughly in half. If geometric spreading dominates (a focused probe helps), switching to a focal-point lens or PA array can recover 10–20 dB at the inspection depth. Showing the two terms separately lets a UT level III pick the right tool. A single combined "loss" number hides the diagnostic.
References & Standards Cited
- EN 583-2:2001 — Non-destructive testing — Ultrasonic examination — Part 2: Sensitivity and range setting
- EN 12668-2:2010 — Probes — Sensitivity and pulse-echo characteristics
- ASNT Nondestructive Testing Handbook, 3rd ed., Vol. 7 Ultrasonic Testing, §3.3 Attenuation
Join NDT Connect — free
The free marketplace for the NDT industry — connect inspectors and the companies that need them.
Free to join · No credit card · Provider profiles verified against ASNT & API rosters · Browse providers
Founder of NDT Connect and Atlantis NDT. 15+ years in industrial inspection across oil & gas, petrochemical, and offshore. ASNT Level III certified across five methods. Drives platform standards for the NDT Connect marketplace.