Ultrasonic testing is the NDE method of choice for most CJP groove welds on structural steel — it covers the full weld volume faster than radiography, does not require radiation clearance, and can locate planar flaws that RT can miss. But UT is only as reliable as the calibration behind it. A technician with improperly calibrated equipment can sweep a weld full of lamellar tears and generate a clean report. AWS D1.1:2025 Annex I lays out the calibration block requirements and procedures that prevent that outcome. Here is what QC managers and CWIs need to understand about those requirements.
What Calibration Actually Does for UT
Unlike liquid penetrant or magnetic particle inspection, UT gives no directly observable indication — the technician interprets signal amplitude and time-of-flight on a screen. Calibration establishes three things before inspection begins:
- Horizontal linearity: Sound path distance shown on the screen accurately represents actual depth in the material.
- Screen height linearity: Echo amplitude is proportional to signal strength — a 100% FSH (full-screen height) indication actually represents the reference amplitude.
- Sensitivity reference: A known reflector at a known depth produces a known amplitude, so the technician has a baseline for evaluating indications.
Without verified calibration of all three, UT results are anecdotal, not inspected.
The IIW V1 Reference Block
AWS D1.1:2025 Annex I designates the IIW (International Institute of Welding) V1 reference block as the standard for initial calibration of angle-beam transducers. The V1 block has a well-defined geometry with:
- A 100 mm radius circular surface used to calibrate the transducer index point (the true acoustic exit point of the beam from the face of the wedge)
- A 1.5 mm diameter side-drilled hole (SDH) at 25 mm depth for beam angle verification
- Flat surfaces for calibrating horizontal sweep
The technician uses the V1 block to establish the beam exit point (index point), verify the actual refracted angle, and confirm horizontal sweep calibration. This work is done with each transducer and wedge combination — if the wedge wears or changes, the calibration must be repeated.
An alternative for smaller transducers is the miniature angle-beam calibration block (sometimes called the ASME reference block), which Annex I accepts as an equivalent when the V1 block is not practical for the transducer size in use.
The reference block must be made from material acoustically similar to the base metal being inspected — typically a low-alloy structural steel with equivalent sound velocity. Using a reference block from a dissimilar alloy introduces velocity and attenuation errors that corrupt the calibration.
Distance-Amplitude Correction (DAC) Curves
The V1 block calibration sets horizontal and screen height linearity, but it does not account for the fact that sound attenuation increases with distance — a flaw at 6 in depth produces a weaker echo than an identical flaw at 1 in depth, even if both are the same size. The DAC curve corrects for this.
To build a DAC curve, the technician maximizes the echo response from a series of side-drilled holes (SDHs) at different depths in a calibration block fabricated for this purpose. The SDHs are typically:
- 1/4 in (6 mm) diameter
- Located at depths that bracket the thickness of the base metal being inspected — for example, at 1/4T, 1/2T, and 3/4T for a given plate thickness
By connecting the peak responses from each SDH, the technician draws (or electronically establishes) the DAC curve on the UT screen. Any indication that breaks the DAC line by a defined ratio — Annex I specifies rejection at amplitudes meeting or exceeding the DAC level for the applicable weld classification — triggers further characterization.
AWS D1.1:2025 Annex I specifies the exact SDH diameters and depths for the two weld categories it recognizes:
- Statically loaded structures: Larger SDH reference reflectors and higher amplitude thresholds — acceptable for connections where fatigue or fracture is not the design concern.
- Cyclically loaded structures: Smaller SDH reference reflectors and tighter amplitude thresholds — required for structures designed under cyclic load provisions where smaller flaws are structurally significant.
The distinction matters on your project. A transit rail fabrication job falls under cyclic loading requirements; a building moment frame may fall under the static criteria unless the EOR specifies otherwise. Confirm which category applies before setting calibration sensitivity.
For a full comparison of UT and other NDE methods for structural welds, see NDE method selection for structural welds: RT, UT, MT, and PT.
The Reference Block for Production Thickness
The calibration reference block must be fabricated from material with the same nominal thickness as the production weld being inspected — or within a thickness range that keeps the calibrated sound path representative of actual inspection paths. If your production welding includes 1 in and 2 in plate in the same job, you may need separate calibration references or must verify that a single block adequately covers the sound path ranges.
Annex I also requires that reference blocks be maintained and inspected periodically. A block with nicks, tool marks near the SDHs, or surface corrosion affects calibration accuracy. The UT contractor should have a documented block maintenance and traceability program. Ask for it.
Calibration Frequency and Record Requirements
Initial calibration happens before the first weld is scanned each inspection period. AWS D1.1:2025 Annex I requires calibration checks:
- At start of each inspection period
- At intervals not exceeding 30 minutes during continuous UT
- When the technician suspects drift (hit to equipment, extreme temperature change in the inspection environment)
- At the end of the inspection period
If a calibration check reveals that amplitude has shifted by more than 2 dB or the sweep distance has shifted, all welds examined since the last confirmed valid calibration must be re-inspected. The UT technician must log calibration verification as part of the inspection record, not just the weld-scanning results.
The calibration log should document:
- Date, time, and technician ID
- Instrument serial number and calibration status (calibrated within 12-month laboratory interval)
- Transducer serial number, frequency, size, and angle
- Reference block ID and traceability number
- Sweep setting, range, and reference amplitude in dB
- DAC or TCG curve verification results
For a discussion of how UT inspection records feed into a complete NDE documentation package for audit purposes, see NDE documentation for an audit-ready WPS package.
PAUT vs. Conventional UT: Calibration Differences
Phased array UT (PAUT) is increasingly used on structural weld inspection because it can sweep multiple angles in a single scan and produce a permanent sectoral scan record. However, the AWS D1.1:2025 Annex I calibration requirements were written for conventional single-element angle-beam UT.
PAUT used under D1.1 must be qualified under the supplemental PAUT requirements in Annex I (or through a written procedure approved by the engineer) that addresses the additional variables: focal law verification, full-matrix capture calibration, and aperture/element verification for each probe used. The conventional V1 block and DAC curve requirements do not simply translate one-for-one to PAUT.
For a detailed look at phased array UT requirements under D1.1, see phased array UT (PAUT) for AWS D1.1 structural welds.
Technician Qualification: The Other Half of UT Reliability
Calibration block requirements address equipment setup. The other half of UT reliability is technician qualification. AWS D1.1:2025 requires that personnel performing UT on production welds hold current certification to ASNT SNT-TC-1A or an equivalent national standard at Level II (or Level III). Level I technicians may perform scanning under the direct supervision of a Level II. The employer must have a written practice that governs the qualification, training, and certification of their NDT personnel.
When qualifying a new UT subcontractor, request their SNT-TC-1A Level II certifications for each technician, their employer's written practice, and evidence that the practice has been reviewed and accepted by a Level III. A UT report signed by an uncertified or lapsed-certification technician is not compliant with AWS D1.1.
Rule library based on AWS D1.1:2025; verify against your governing edition — the AHJ or contract may specify the 2020 edition, where Annex I article references may differ.
Practical Takeaway for QC Managers
UT is the workhorse NDE method for structural CJP groove welds, but it is not a black box. The calibration procedure, reference block traceability, DAC curve setup, and calibration frequency checks are engineering requirements, not technician preferences. Before an NDE contractor begins production scanning:
- Verify they have the correct reference block for your base metal thickness and loading category (static vs. cyclic).
- Confirm calibration intervals are being logged and checked.
- Verify technician certification is current and covers the weld type and thickness range being inspected.
- Confirm whether Annex I conventional UT or PAUT supplemental requirements govern.
A clean UT report from a properly calibrated inspection gives your QC and the EOR genuine confidence in weld soundness. An unchecked calibration makes the report paperwork.
To learn more about how your WPS and NDE records integrate into a unified quality package, explore the welding software tools at wpswelding.com.