An ultrasonic pulse from a 5 MHz transducer enters a steel plate. It travels at 5,920 meters per second, hits a 2-millimeter slag inclusion 15 millimeters below the surface, and reflects. The transducer receives the echo 5.07 microseconds later. An A-scan display draws that echo as a single spike on a time-versus-amplitude graph. A B-scan cuts a cross-section through the plate and draws the flaw as a bright dot at 15 millimeters depth. A C-scan looks down from above and maps the flaw as a colored patch 2 millimeters wide at X-Y coordinates matching its position on the part surface. Same transducer, same pulse, same flaw — three fundamentally different representations. Each answers a different question: Is there a flaw? (A-scan) How deep and what shape? (B-scan) Where exactly is it on the part? (C-scan). This article explains the physics behind each imaging mode, the data each produces, and when to use which — so you can select the right scan mode for your inspection task and interpret the results correctly.
What an A-Scan Shows — and What It Does Not
An A-scan is the foundational ultrasonic display: a two-dimensional graph with time-of-flight on the horizontal axis and echo amplitude on the vertical axis. The horizontal axis represents depth — sound travels at a known velocity in the material, so time converts directly to distance. The vertical axis represents how much ultrasonic energy returned from that depth. A large spike at 15 millimeters means something at that depth reflects ultrasound strongly. No spike means the material is homogeneous at that depth.
The A-scan answers three questions with high precision: Is there a reflector? (yes/no, based on whether any echo exceeds the gate threshold) At what depth? (time-of-flight × velocity / 2, typically ±0.1 mm) How large is the echo relative to a reference? (amplitude compared to a known reflector like a 1.5 mm flat-bottom hole at the same depth). A trained operator reads an A-scan in real time and makes accept/reject decisions based on whether an echo amplitude exceeds the DAC (Distance Amplitude Correction) or TCG (Time-Corrected Gain) curve programmed into the instrument.
What an A-scan does not show: the lateral position of the flaw within the transducer beam, the physical shape of the reflector, or any spatial relationship between multiple reflectors. An A-scan is a one-dimensional view along a single sound path. Mapping a weld with an A-scan means raster-scanning the transducer by hand, logging each indication's coordinates on a paper grid, and mentally reconstructing the 3D flaw distribution — a skill that takes years of experience to develop.
How a B-Scan Builds a Cross-Sectional Slice
A B-scan (Brightness scan) adds a spatial dimension to the A-scan. Instead of plotting amplitude as a vertical spike, a B-scan encodes amplitude as color or grayscale brightness and plots it on a 2D grid where one axis is depth and the other is the transducer's linear position. As the transducer moves along a line on the part surface, the instrument builds up a cross-sectional image one column of pixels at a time — each column is essentially an A-scan turned sideways, with echo amplitude mapped to pixel intensity.
The result is a side-view cross-section through the material that shows the part's thickness profile, the depth and approximate shape of any internal reflectors, and the spatial relationship between features. A B-scan of a corroded pipe wall shows the remaining wall thickness as a varying dark band — the operator can measure minimum wall thickness at a glance. A B-scan across a weld shows the weld root geometry, any lack of fusion at the bevel faces, and embedded slag lines in spatial context — all information an A-scan amplitude alone cannot convey.
B-scan resolution depends on two variables: scan increment (how far the transducer moves between each A-scan column) and beam diameter at the depth of interest. A typical encoded manual scanner captures a B-scan at 1 mm position increments with a 3 to 6 mm effective beam diameter. Phased-array instruments generate B-scans electronically by steering the beam through a sequence of angles without physically moving the transducer — called a sectorial or S-scan — producing a cross-section in under a second.
What makes a C-scan different from a B-scan?
A C-scan (Contour scan) looks at the part from above — it is a top-down plan view where X and Y axes map to surface coordinates on the part and color represents the ultrasonic response at each XY position. To build a C-scan, the operator either raster-scans the transducer in a grid pattern across the part surface (manually or with an automated scanner) or uses a phased-array probe that electronically scans an array of elements. At each grid position, the instrument records the maximum echo amplitude within a defined depth gate — or the exact depth to a reflector — and assigns a color based on the value. The result looks like a heat map overlaid on the part geometry: red and orange for high-amplitude indications, green and blue for low, and white or clear for no indication.
A C-scan answers the question "where is the flaw on the part?" — the same question a manufacturing engineer or a maintenance planner asks when deciding whether a flaw is in a critical stress zone or a non-load-bearing area. In aircraft maintenance, a C-scan of a composite wing skin produces a map of delamination, disbond, and water ingress that the engineer compares directly to the structural repair manual's allowable damage limits — often without ever looking at the raw A-scan data. C-scans are also the dominant output format for automated immersion scanning systems that inspect raw material plates, billets, and forgings at the mill.
Phased Array: Running All Three Modes Simultaneously
A phased-array ultrasonic instrument uses a multi-element probe — typically 16 to 128 elements — and electronically controls the timing (phasing) of each element's transmit and receive pulses. By varying the delay pattern, the instrument can steer the ultrasonic beam through a range of angles, focus at different depths, and scan along the array axis without moving the probe. One linear scan of a phased-array probe across a weld simultaneously generates an A-scan for each beam angle, a sectorial B-scan showing the weld cross-section in real time, and — if the probe position is encoded — a top-down C-scan of the entire weld length. The operator watches the S-scan (B-scan equivalent) during scanning to identify indications, then uses the C-scan to map the indication length and the A-scan to verify amplitude against the acceptance criteria. All data is recorded — the inspection can be replayed and re-analyzed offline without re-scanning the part.
Phased-array instruments, including the Waygate USM 100, combine all three scan modes with a 7-inch touchscreen interface and cloud-based data sharing — the raw A-scan, sectorial B-scan, and C-scan top view update simultaneously during scanning. For critical weld inspections requiring full data traceability, phased array has largely replaced conventional single-element UT in industries governed by ASME B31.3, AWS D1.1, and ISO 13588.
When should I use each scan mode?
| Inspection Task | Primary Mode | Why |
|---|---|---|
| Go/no-go thickness measurement at a grid of points | A-scan | One depth reading per point; no spatial imaging needed |
| Weld flaw detection and characterization per AWS D1.1 | Phased Array (A+B+C combined) | Need amplitude (A), cross-section geometry (B), and flaw length (C) for acceptance criteria |
| Corrosion mapping on a pipe or vessel wall | B-scan + C-scan | B-scan shows depth profile per scan line; C-scan merges lines into a wall thickness map |
| Composite delamination and disbond inspection | C-scan | Top-down mapping matches the part geometry; delamination area, not depth, drives repair decisions |
| Bondline inspection — adhesive between two skins | B-scan | Cross-section shows which side of the bondline disbonded and whether the adhesive is absent or degraded |
| On-line tube or bar testing at production speed | C-scan (automated) | Rotating probe + linear feed gives continuous C-scan strip map at meters-per-minute throughput |
| Spot-checking a suspicious indication found by another method | A-scan | Quick amplitude and depth verification; no scanning hardware needed |
How does encoder resolution affect C-scan quality?
A C-scan is only as sharp as the position encoding driving it. With a manual encoded scanner tracking probe position at 1 mm resolution, a C-scan resolves features down to roughly 2 to 3 mm — adequate for corrosion mapping but insufficient for characterizing a 0.5 mm fatigue crack tip. Immersion scanning systems using precision linear stages at 0.1 mm or 0.05 mm step resolution produce C-scans with near-optical image quality, capable of mapping individual plies in a composite laminate or resolving a 0.2 mm inclusion in a bearing race. The trade-off is scan time: doubling the resolution quadruples the number of data points and the scan time. For a production forging inspection scanning at 1 mm versus 0.5 mm resolution, the difference is between a 10-minute scan and a 40-minute scan — and a forging mill running three shifts measures throughput in parts per hour, not pixels per millimeter.
A-scan, B-scan, and C-scan are not competing methods — they are three views of the same ultrasonic data, each optimized to answer a different question. The A-scan tells you the flaw is there and how big the echo is. The B-scan tells you how deep it sits and what cross-sectional shape it casts. The C-scan tells you exactly where on the part surface it maps. Modern phased-array instruments deliver all three views simultaneously from a single scan. The skill is not knowing how to switch between modes — it is knowing which mode answers the question the acceptance standard asks.
