Piezo Microphone Frequency Response

Before you start

A piezo microphone works as an input device that converts vibration or pressure into an electrical signal. Its frequency response is not flat like many studio condenser or dynamic microphones: piezo elements are often more sensitive in a higher band and can roll off strongly at low frequencies. Because of that, the same input level reading can mean different loudness depending on the frequency of the sound source.

You can evaluate frequency response locally with a computer, audio interface, or recording app. No online service is required. The process is reversible and does not change the microphone itself. Before checking the response, confirm that the microphone is connected, selected as the active input device, and not muted in the operating system sound settings.

Step-by-step instructions

Use a recording or analysis tool that can show the spectrum of the incoming sound. Start with a controlled sound source, such as a tone generator played through a speaker, and keep the microphone at a fixed distance.

• Open the sound settings and set the piezo microphone as the input device. Set the input level or gain to a moderate value to avoid clipping during the loudest test tone.
• Confirm the sample rate for the recording. A higher sample rate, such as 48 kHz, can represent frequencies up to roughly half that rate; this is sufficient for most microphone response checks in the audible range.
• Place the microphone at a consistent distance from the speaker. Mark the position so repeated tests use the same placement.
• Play a sine tone at one frequency, such as 100 Hz, and note the recorded level shown by the analyzer.
• Repeat the same procedure with spaced test frequencies, for example 200 Hz, 400 Hz, 1 kHz, 2 kHz, 4 kHz, 8 kHz, and 10 kHz. Keep the playback level unchanged.
• Compare the recorded levels across the frequencies. The pattern shows where the piezo microphone is less or more sensitive.

If the recording tool supports a real-time spectrum display, use it while each tone is playing. Otherwise record each tone as a short clip and compare the levels afterward. Do not rely on the piezo element's published curve as an exact match for your mounting method: a piezo disc attached to a surface can respond differently than the same element used in free air.

Check the output

A useful check is that lower tones produce a clear drop in level while mid and upper tones remain audible. The exact roll-off point depends on the piezo element, its housing, and the connection impedance, so there is no single correct curve.

Watch for two limitations. First, a speaker and room can color the result, so this procedure shows a combined response rather than a laboratory measurement. Second, background noise can lift the measured level at frequencies where the microphone itself is weak, making the low-end roll-off appear less severe than it is.

Common problems and next steps

If the levels jump near the highest measured tones, the tone may be quieter in the room or outside the useful band of the speaker. Lower the measurement range or use a source with more even high-frequency output before concluding that the microphone has a sharp peak.

If low frequencies are missing, check that the microphone is not connected through a high-pass filter in the audio interface or recording software. Such a filter can remove the same low band that a piezo microphone already reproduces weakly.

For speech or instrument use, a piezo microphone often needs corrective equalization. Reduce prominent high-frequency peaks and add only as much low-frequency gain as the remaining background noise allows. Compare the equalized result with an output device such as headphones or monitor speakers, because the visible spectrum alone does not show how the recording will sound.

If the signal is too noisy, reduce the input level only as a temporary check and identify the noise source first. Raising gain later to recover weak low frequencies also raises the noise floor. When deciding whether the microphone is usable, judge the frequency response together with the noise at the intended recording volume, not from isolated numbers.