Components | Sound Testing

Test Conducted by The Archery Sound Lab

Sound testing for the 2026 study was performed by The Archery Sound Lab, an external acoustics-testing facility specialized in archery noise measurement. PCA mailed the full set of fletched test arrows and broadheads to the lab; all shooting, recording, and audio processing described below was handled at their facility using their own bow, shooter, and instrumentation.

Because of the external arrangement, the bow, shooter, and shot environment for sound testing are not the same as the equipment used on the drag and restorative-lift pages. Specifics of the 2026 sound-test setup:

• Bow: Mathews V3X33 (75 lbs, 28.5″ draw).
• Launch mechanism: Purpose-built shooting machine.
• Bow tune: Tuned to produce a bullet hole through paper with a bareshaft and fletched arrow.

Overview and Test Facility

This page details the approach used by the Archery Sound Lab to measure the in-flight noise generated by various arrow configurations. Accurate measurement of arrow fly-by noise requires strict control of environmental variables. Testing was conducted in the Arrow Sound Testing Chamber (ASTC), a custom shoot-through facility designed to isolate the arrow's in-flight noise from external sounds (wind, wildlife, traffic, aircraft) and from noises generated by the bow or arrow impact.

The ASTC achieves an average noise floor of 22 dB(Z) across 500 Hz–31,500 Hz, with validated free-field acoustic conditions down to 250 Hz (per ISO 26101:2017). Reverberation times are low (0.03 s RT60, T20) from 500 Hz to 8 kHz (ISO 3382-1).

To confirm the chamber does not affect measurement accuracy, a control test was run both outdoors and within the ASTC. After adjusting for ground reflections, measurements matched within ±0.5 dB, confirming the reliability of the facility.

Arrow Fly-By Measurement Setup

Arrows were shot through the ASTC using a purpose-built shooting machine so they passed a calibrated measurement microphone at a fixed distance of 1 meter. Stringlines, forming a 50 mm-wide window, ensured a consistent flight path. Any contact with the stringlines was visually detected and that shot was repeated, resulting in all valid measurements occurring between 975 mm and 1025 mm from the microphone. This constrains theoretical sound level variation to ±0.2 dB (inverse-square law).

Equipment

• Microphone: Earthworks M30 Measurement Microphone (Class 1)
• Audio Interface: RME Babyface Pro FS
• Software: Audio and Acoustical Response Analysis Environment (AARAE) in MATLAB
• Calibration: ¼″ microphone calibrator (94 dB @ 1 kHz)
• Velocity: Garmin Xero C1 Pro Doppler chronograph

Arrow Testing Procedure

• 24 broadheads and 24 fletching arrow configurations tested.
• Each fletching configuration was shot using a shooting machine.
• Each configuration was shot repeatedly until the target number of valid, stringline-cleared measurements was obtained: 10 samples per vane configuration and 5 samples per broadhead configuration.
• All arrows were shot from a Mathews V3X 33 (75 lbs, 28.5″ draw).
• Average test velocity: 284 fps across all configurations.
• The bow was tuned to produce a bullet hole through paper with a bareshaft and fletched arrow.

Audio Processing and Loudness Calculation

1. Calibrate microphone input using the ¼″ calibrator.
2. Trim audio to a 0.2 second window to exclude bow noise and target impact.
3. Apply a Tukey window (tapered cosine, ratio 0.2) for smooth fade-in and fade-out.
4. Screen each recording by ear and omit any with audible background noise.
5. Apply 1/3-octave band FFT with IEC Class 0 filter slopes.
6. Compute L(Z)max values (Fast, 125 ms time weighting) from 500 Hz–31,500 Hz.
7. Log-average valid shots per arrow configuration in each 1/3-octave band.
8. Apply Deer-Weighting and Human-Weighting: each band is adjusted per the white-tailed deer auditory threshold (Heffner, 2010) to reflect perceived loudness for deer, whose hearing is centered roughly one octave higher than humans. In addition, A-Weighting is applied to each band to reflect perceived loudness for humans.
9. 95% confidence intervals are calculated for each band's SPL for each configuration using the t-distribution with degrees of freedom equal to the sample size minus one, to appropriately reflect the increased uncertainty associated with small sample sizes.
10. For bands where the lower 95% confidence interval is not defined (i.e., negative in the linear domain), a value of −120 dB SPL is substituted for the purposes of weighted summation. These cases are marked as ‘N/A’ in band-specific CI reporting.
11. Sum the weighted bands logarithmically to yield a single-number perceived loudness metric.