In this low range, most of the prominent overtones have ratios to the fundamental greater than 15 (see Fig. The lowest bars, those that fall within about the first octave of the gamelan, display the highest total number of prominent overtones (between 16 and 26 per bar). Although there is little consistency from bar to bar, there are general trends that can be found across the range of the ensemble. The primary purpose of this study was to determine the important overtone ratios and modal shapes of gamelan metallophone bars and investigate their consistency across the ensemble. This usually yielded a clear modal shape, and it allowed compilation of a final list of the prominent overtones of each bar. In these cases, and in cases where the SLDV indicated an overtone but showed no clear modal shape, the bars were excited again with a sine wave at the expected frequency of the overtone. (In-plane bending and stretching modes were not taken into account.) In several cases, overtones that appeared prominently in the acoustical measurements were not present in the initial SLDV scans, probably because there was insufficient energy in the broadband acoustic signal to excite that mode. Each bar was excited with a pseudorandom signal from a Mackie HR-824 studio monitor, and the PSV was used to determine the prominent transverse modes of vibration.

One ugal, one jublag, one penyacah, and one pair each of jegogan, pemade, and kantilan (27 bars total) were examined with SLDV the bars that were scanned are highlighted in Fig. The ratio of each prominent overtone frequency to the fundamental frequency was then calculated for all bars.įor the SLDV measurements, a Polytec PSV-400 was used to scan 150–180 points per bar. All overtones with a sound pressure level ∼ 35 dB or higher were listed as prominent. Trials were made with different percussionists and striking techniques, and strike-to-strike differences were not significant for this particular study. The relatively long time window permitted the frequencies to be resolved using Fourier analysis to within 0.2 Hz. Because spectral analysis of the recordings showed the struck bars produced very stable, high-quality-factor resonances, the data were treated as stationary for the purposes of determining frequency. Data were acquired at a sampling frequency of 50 kHz. An acoustic recording was made of each of the 162 bars using a type-1 12.7-mm (0.5-in.) prepolarized microphone and a National Instruments USB-9233 24-bit data acquisition module attached to a laptop running LABVIEW. Figure 1 lists the metallophones of gamelan Bintang Wahyu and shows their ranges.

Gamelan wav samples