Guide to Transducer Sensitivity Testing for Medical Devices

A transducer can meet its nominal electrical specifications and still produce performance data that are unsuitable for a regulatory submission. Small changes in drive conditions, loading, couplant, alignment, temperature, or measurement bandwidth can materially affect the reported result. This guide to transducer sensitivity testing addresses the controls needed to generate repeatable, traceable data for medical device verification.

For ultrasound probes and other medical transducers, sensitivity is not a single universal value. The relevant measurement depends on the device function, intended use, operating mode, and applicable standard or test method. A meaningful protocol begins by defining the measurand before selecting the fixture or instrument.

Define What Sensitivity Means for the Device

Sensitivity describes the relationship between an applied stimulus and the transducer response. In transmit mode, it may be expressed as acoustic output for a defined electrical input. In receive mode, it may be expressed as electrical output voltage per unit of incident acoustic pressure. For an imaging system, the practical performance question may instead concern the minimum detectable target or the depth at which contrast and spatial resolution remain acceptable.

These distinctions matter because they require different test configurations. A hydrophone-based measurement can characterize an ultrasound transducer’s transmitted acoustic field. A calibrated acoustic source or reciprocal method may be used to assess receive sensitivity. System-level image quality testing evaluates the combined behavior of the probe, beamformer, processing chain, display settings, and operator-selected controls.

The test plan should state the following in clear, controlled terms: the device configuration; operating mode; frequency or frequency range; drive waveform; loading condition; measurement location; environmental conditions; and the sensitivity metric to be reported. If the objective is comparison against a prior device generation, the protocol should also identify which settings must remain equivalent and which changes are intentional.

Establish a Controlled Test Configuration

Sensitivity results are only as useful as the stability of the setup. For acoustic measurements, fixture design and alignment often contribute more variation than the analyzer itself. The transducer face, acoustic axis, target or hydrophone position, and water-path geometry should be controlled and documented.

Water is commonly used as the propagation medium for ultrasound testing, but its condition requires attention. Temperature affects sound speed, attenuation, and measurement geometry. Degassing helps limit bubbles that can scatter sound or create inconsistent coupling. The water temperature, degassing approach, and stabilization period should be defined in the procedure when they can influence the result.

Electrical drive conditions also require control. Record the source impedance, voltage definition, pulse repetition frequency, pulse shape, duty cycle, and cable configuration. A voltage measured at the generator may not equal the voltage present at the transducer terminals, particularly where cable losses, impedance mismatch, or high-frequency ringing are present. Measuring the electrical input at an appropriate reference plane reduces ambiguity.

For receive sensitivity, the incident field must be characterized at the measurement location. The source should provide a stable, known acoustic stimulus over the required frequency range. The receiver termination, preamplifier gain, filtering, and acquisition bandwidth must remain controlled because each can alter the apparent sensitivity.

Alignment and Spatial Sampling

An ultrasound field is rarely uniform. A small position shift can change the measured pressure or intensity, especially near the focal region or in a field with narrow lobes. Establish the acoustic axis through a scan rather than assuming that the mechanical centerline is the acoustic centerline.

The spatial sampling interval should be appropriate for the wavelength, beam width, and purpose of the measurement. Coarse scans can miss local maxima and understate peak output. Excessively dense scans may add time without improving the decision, particularly in routine production comparisons. The appropriate balance depends on whether the work is design characterization, verification, periodic surveillance, or investigation of a nonconforming result.

Select Calibrated Measurement Equipment

A traceable equipment chain supports confidence in both the result and the uncertainty statement. Depending on the method, the chain may include a calibrated hydrophone, acoustic power meter, reference transducer, oscilloscope, digitizer, signal generator, power amplifier, positioning system, thermometer, and environmental monitor.

Calibration status alone is not sufficient. Confirm that each instrument’s calibration range, frequency response, resolution, and uncertainty are suitable for the intended measurement. A hydrophone with a valid calibration certificate may still be unsuitable if the test frequency lies outside its characterized range or if its active element is too large to resolve the field being measured.

Hydrophone spatial averaging is a common consideration. When the sensitive element has a dimension comparable to the acoustic field structure, it can average localized pressure peaks. Corrections may be required, or a smaller-aperture hydrophone may be more appropriate. Similarly, cable loading and digitizer bandwidth can distort short pulses if the measurement system cannot capture the relevant frequency content.

For medical device programs, equipment records should connect directly to the reported data. Identification numbers, calibration due dates, instrument settings, software versions, and any applied corrections should be retained with the test record. This level of traceability is particularly important when results support design verification, a change assessment, or a regulatory submission.

Apply a Method That Matches the Intended Claim

The appropriate method depends on what the manufacturer needs to demonstrate. Transmit sensitivity characterization may require peak pressure, pulse intensity integral, acoustic power, beam profile, center frequency, bandwidth, or output stability across operating modes. Receive sensitivity may require frequency response, noise floor, dynamic range, and response linearity over a specified incident pressure range.

A single measurement point can be sufficient for a controlled production check, but it is generally not enough for design verification of a focused or electronically steered array. Array transducers introduce additional variables, including channel-to-channel variation, aperture selection, steering angle, focusing law, and element failures. The protocol should identify whether the reported result represents an individual element, a subaperture, a programmed transmit event, or a complete imaging mode.

Acceptance criteria should be technically justified. A tolerance copied from a predecessor device may be inappropriate if the new device has a different frequency, aperture, intended depth range, or signal-processing architecture. Where a requirement is tied to a recognized standard, the test report should identify the applicable edition, clause, and conditions. Where the requirement is internally derived, document the engineering rationale and its connection to user needs, risk controls, or performance specifications.

Account for Measurement Uncertainty

A result near the acceptance limit requires more than a pass or fail statement. Measurement uncertainty affects the confidence that can be placed in the conformity decision. Significant contributors may include hydrophone calibration uncertainty, positioning accuracy, temperature effects, electrical measurement accuracy, repeatability, operator alignment, and signal processing.

The uncertainty budget should reflect the actual method rather than a generic template. Some terms may be negligible for one configuration and dominant for another. For example, positioning uncertainty can be critical in a tightly focused field, while electrical drive variation may dominate in a pulsed system with changing load impedance.

Define the decision rule before testing. The rule should state how uncertainty is considered when the measured value approaches the specification limit. This prevents inconsistent disposition decisions and gives quality and regulatory teams a documented basis for assessing marginal results.

Build Records That Support Review and Reproduction

A complete sensitivity test report should allow a qualified reviewer to understand what was tested, how it was configured, and how the reported value was obtained. Include the device identification, firmware or software version where relevant, test procedure revision, applicable requirement, raw-data references, setup photographs or diagrams when useful, instrument identification, environmental conditions, calculations, corrections, uncertainty, and final disposition.

Raw data retention is especially valuable when a result is later questioned or a design change is introduced. Retaining waveform files, scan maps, exported images, and analysis settings enables comparison without reconstructing the original test from memory. It also helps distinguish a true device shift from a change in test setup or data processing.

When independent laboratory testing is required, an ISO/IEC 17025-accredited laboratory with relevant medical device and acoustic testing capability can provide a controlled method, traceable instrumentation, and documentation structured for compliance review. Sigma Scientific Services supports standards-based medical device performance testing where acoustic output, image quality, and related transducer characteristics require documented evaluation.

Common Causes of Misleading Results

Several recurring issues can compromise sensitivity data: measuring generator voltage instead of transducer-terminal voltage; using an unverified acoustic axis; allowing water temperature to drift; changing signal-processing filters between samples; applying a calibration correction outside its valid frequency range; and comparing results from different fixture geometries as though they were equivalent.

These errors do not always produce obviously abnormal data. They can create plausible results that shift enough to trigger unnecessary design investigations or mask a meaningful performance change. Method discipline is therefore as important as instrument capability.

Sensitivity testing should be treated as a controlled performance measurement, not simply a laboratory reading. When the measurand, setup, calibration chain, uncertainty, and acceptance rule are defined before data collection, the result becomes evidence that engineering, quality, and regulatory teams can use with confidence.