Electroacoustic network

Abstract

Disclosed is a passive delay network for insertion in the high frequency signal path of a crossover network in a high fidelity loudspeaker. The delay network provides means to adjust the optimum listening window electrically rather than by changes in the physical construction of the speaker cabinet. Electroacoustic high frequency and low frequency drivers may be mounted in a planar baffle or mounted co-axially. The delay network retains a 'flat' amplitude-frequency characteristic and minimizes time dispersion in the acoustic signal received by the listener. Correction for time delay is most important near the transition frequency of the crossover network because near the transition frequency substantially equal sound energy is radiated by both the high frequency and low frequency drivers. Also disclosed is a novel construction for a co-axial driver combination and a technique for adjusting both the delay in the electric delay line network and the air path delay of electroacoustic drivers. The co-axial drivers include sound absorbent means preventing direct sound radiation between the low and high frequency drivers. The adjustment technique comprises a method of experimentally testing and adjusting the mutual inductances in the electric delay line and the air path delay. Applicant's method and apparatus empirically selects and adjusts the electroacoustic networks of the loudspeaker to effectively reduce or eliminate the cause of amplitude peaks, coloration and variation throughout substantially the entire 'hemispherical' volume in front of co-axial loudspeakers. The method and apparatus is also effective with non-co-axial loudspeakers. The loudspeaker electroacoustic network components are physically, electrically and magnetically (inductively) adjusted for flattened acoustic phase response beyond the amplitude pass band roll-off of each filter, driver and delay network both on and off axis. Then the components are further adjusted as a combination to minimize the on and off axis amplitude peaking throughout the hemispherical listening volume in front of the loudspeaker. This approach to equalizing the loudspeaker permits the final adjustments to quickly converge on the values necessary for flat amplitude and phase response throughout substantially the entire volume in front of the loudspeaker. Applicant's empirical approach eliminates the need for sophisticated computer analysis of oversimplified loudspeaker simulations.