Ultrasound Probe Termination: Define the Measurement Boundary

By ALEVA | California, USA

Engineering question: When a probe produces a longer or weaker echo on a different receiver, did the acoustic stack change—or did its electrical load?

The receiver is part of the experiment

A probe comparison can change when the cable, connector, protection network or receiver input changes. Analog Devices' 2010 receiver overview explains the practical conflict: cable loading can restrict received bandwidth, while a low-impedance termination can reduce signal amplitude. A larger displayed echo therefore does not, by itself, establish a better probe. Compare the complete transfer path and noise conditions, not brightness alone. [1]

Separate the reference plane from the application load

A 2025 peer-reviewed study by Leipner, Wolf and Kupsch uses calibrated vector-network-analyzer S-parameters and time/frequency analysis to separate electrical and acoustic responses in high-frequency scanning acoustic microscopy. Its useful lesson for an OEM evaluation is to state where calibration ends and which hardware remains inside the measurement. The demonstrated microscopy method is not evidence of clinical-array performance or an already-validated multichannel production test. [2]

For a proposed comparison, document two boundaries: the calibrated connector plane used to characterize the probe, and the actual receive-chain interface used in the intended system. Moving between them requires accounting for the intervening cable, adapter and circuitry. A connector-plane impedance curve and an installed pulse-echo measurement answer different questions; retain both rather than merging them into one score.

Termination is a trade-off, not a universal setting

Texas Instruments' AFE5808 datasheet, revision D from January 2014, documents high-impedance, passive and programmable active termination configurations. It also describes frequency-dependent input impedance and different noise behavior between these configurations. This is an established engineering example, not a new component announcement or a statement that ALEVA uses that device. An input-resistance register alone is not a complete description of the load over the probe bandwidth. [3]

The proposed decision rule is to select a termination against the intended receive task: usable echo spectrum, ringing, sensitivity and noise together. Do not automatically choose the configuration producing the highest peak voltage. Conversely, do not accept a shorter pulse if weak echoes become unusable.

A proposed OEM comparison record

Record probe identity and revision, acoustic loading and target geometry, cable and adapter configuration, calibration plane, receiver gain and input setting, transmit waveform, analog filtering and sampling configuration. Change one electrical condition at a time and retain the raw RF captures.

Compare echo amplitude and spectrum, pulse duration under a stated definition, noise over the same bandwidth, and repeatability. Keep gain and display mapping fixed; flag clipping or transmit-recovery contamination. Set acceptance limits from the intended system requirements before comparing candidates. Repeat the assessment with the production cable and connector configuration before freezing the interface.

Limitations and OEM/ODM decision: These steps are a proposed evaluation method, not reported test results. Electrical characterization cannot replace acoustic-field measurements, device safety assessment or application-specific verification. For a development handover, specify the probe, interconnect and receive configuration as one controlled interface; a passing probe on an unspecified bench receiver is not a complete integration acceptance criterion.

References

1. Analog Devices, Overview of Ultrasound Imaging Systems and the Electrical Components Required for Main Subfunctions, May 10, 2010; manufacturer application article, LNA section.

2. E. Leipner, M. Wolf and C. Kupsch, Characterization of high-frequency ultrasonic transducers using S-parameters: Separation of electrical and acoustic responses, Mechanical Systems and Signal Processing 238, 113172, September 1, 2025. DOI: 10.1016/j.ymssp.2025.113172; peer-reviewed research.

3. Texas Instruments, AFE5808 datasheet SLOS688D, September 2010, revised January 2014; LNA input impedance/termination, pp. 54–55. Historical component documentation, not a 2026 release.