IllustrationSpatial aliasing (grating lobes) appears when microphone spacing exceeds half the wavelength: d = λ/2 = v/(2f), with the paper taking sound at 343 m/s. The Infineon IM69D130 datasheet adds a ±2° phase match and a 4 x 3 x 1.2 mm package. This page lists what to measure and what to ask.
A voice wearable has little room for microphones. This page puts the sourced spacing facts side by side, lists what to measure on your own board, and ends with questions to settle before you pick a spacing. It has no Duobay price and no measured result. Where a device records people, plan for consent and a visible indicator.
What do the sources say about spacing?
| Fact | Source |
|---|---|
| Spatial aliasing (grating lobes) appears when element spacing exceeds half the wavelength. The paper puts the limit at half the wavelength. Since λ = v/f, that gives d = λ/2 = v/(2f). | arXiv 2406.07663 (ultrasound array paper), as of October 31, 2026 |
| The paper takes the speed of sound as 343 m/s. | arXiv 2406.07663, as of October 31, 2026 |
| The IM69D130 phase response is matched to ± 2° between microphones, to support beamforming. | Infineon IM69D130 datasheet, as of October 31, 2026 |
| The IM69D130 package is 4 × 3 × 1.2 mm, and the datasheet recommends a PCB sound port of 0.8 mm diameter. | Infineon IM69D130 datasheet, as of October 31, 2026 |
d = λ / 2 = v / (2 × f)
d spacing between neighboring mic ports
v speed of sound (the paper uses 343 m/s)
f frequencyTwo of these facts bound the design from different sides. The aliasing relation limits the spacing from above for a given frequency. The package outline and sound port size set how close two mic ports can sit on a PCB. The matching figure is the real-world phase error to compare with the phase difference that spacing produces. It belongs to this mic model only, so another mic needs its own datasheet value.
What should you measure on your own board?
The sources do not give beam patterns, noise suppression or matching for your assembled enclosure, and we have not measured them. Measure these before you fix the layout.
| What to measure | How to capture it | Raw data to record |
|---|---|---|
| Phase and level match between mics | Play a known tone or sweep from the same distance to all mics in a quiet space. Capture all channels in one synchronized recording. | Raw per-channel recordings, sample rate, source distance, temperature, mic part numbers, assembly lot, and the phase and level difference per frequency. |
| Real port pitch on the assembled board | Measure the center-to-center distance between the sound ports in the finished housing. | Measured pitch for each mic pair, measurement method, and the gasket or mesh in place. |
| Aliasing onset | Sweep a tone from a fixed angle off the beam axis, with the beam steered to each direction you plan to use. Record the beamformer output over frequency. | Steering angle, source angle, frequency sweep, output level per frequency, and the algorithm settings. |
| Beam pattern worn on the body | Rotate the worn device, or the source around it, in a quiet space. Record output level against angle at several frequencies. | Angle, frequency, output level, how the device is worn, the enclosure revision, and the room description. |
| Effect of temperature | Repeat the phase and aliasing tests at the coldest and warmest conditions the product is meant to see. | Chamber or room temperature, soak time, and the same raw data as the first test. |
| Speech result in a real scenario | Record the wearer and a second talker in a space that matches your use case. Run your own processing chain with the array on and with a single mic. | Raw multichannel audio, processing settings, how you will score it (written before scoring), and the scores. |
What should you ask before you pick a spacing?
Take this list to your acoustic and mechanical leads, and to any supplier who proposes a layout.
- What is the highest frequency our algorithm actually needs to beamform?
- Do we steer the beam in more than one direction, and which ones?
- What is the real port pitch in the enclosure, and what does the package outline allow?
- Does the datasheet’s matching figure apply to the mic we plan to use, and under which conditions?
- What temperature range will the device see, and how does it change the speed of sound we assume?
- How long is the aperture for each mic count we are considering, and does the housing have room for it?
- Is there a visible indicator and a consent step wherever the device captures voice?
- Which pass lines did we set, and where did we write down the reason for each?
Questions about mic array spacing
What is the maximum mic spacing for a given frequency?
Half a wavelength: d = λ/2 = v/(2f), where v is the speed of sound and f is the frequency. The cited paper uses 343 m/s (source, as of October 31, 2026).
Why does the half-wavelength rule come from an ultrasound paper?
The paper is about an ultrasound array, but the relation is frequency-agnostic. We cite it because it states the limit and the 343 m/s value in plain words.
Does a smaller pitch always beat a larger one?
No. A larger pitch narrows the beam at low frequencies, but aliasing starts at a lower frequency (f = c / (2d)). A smaller pitch pushes aliasing to higher frequencies, but it also shrinks the phase difference between mics, so a fixed matching error matters more. Your target band and enclosure set the tradeoff.
Does this page predict how my product will perform?
No. It lists sourced facts, a measurement checklist and questions to ask. It contains no measured results for a worn enclosure.
Can Duobay help choose spacing and mic count?
We do not publish prices. Tell us about your device and we will explain how it works. Whatever the layout, a product that captures voice needs a consent step and a visible indicator.
If you are planning a voice wearable, tell us about your device. You can also read about our audio and voice hardware development.
Sources
All pages accessed October 31, 2026. Re-read the documents before you decide, because datasheets get revised.
- arXiv 2406.07663, ultrasound array paper (grating lobes, spacing limit, 343 m/s): https://ar5iv.labs.arxiv.org/html/2406.07663 (accessed 2026-10-31)
- Infineon, IM69D130 datasheet (±2° phase match, package 4 × 3 × 1.2 mm, 0.8 mm sound port): Infineon IM69D130 datasheet (accessed 2026-10-31)
Planning a voice or audio device? Tell us what it needs to do. A Duobay engineer replies within 1 business day. Send a project brief →