Despite extensive searching (by me), a schematic for the DK100 has not been found. However, a "representative" schematic for a ULB is shown below.
The schematic is quite simple, and shows two timing generation functions. The bottom dashed box (28) labeled pulse generator produces timing pulses at a given duration and rate. In the case of this implementation the duration and rate are selectable by the switch settings shown in that part of the schematic. In the case of the DK100 the duration is fixed at nominally 10ms, and the rate is fixed at nominally once per second. The other timing function is contained in dashed box (24). This oscillator produces the "carrier" frequency of the radiated energy. In the case of the DK100, this frequency would be set at 37.5 kHz.
Gates (26) and (30) provide a train of 37.5 kHz pulses at the duration (envelop) and rate (pulse repetition frequency) to an impedance matching transformer which drives a piezo-ceramic crystal. The ceramic is bonded to the case of the ULB, and couples this mechanical energy to the ULB case which then imparts acoustic energy to the water. Both the piezo ceramic and the case are "tuned" to resonate near 37.5 kHz. "Tuned" in this context refers to the mechanical properties of the ceramic and case.
The case is designed to have a high 'Q' (mechanical resonance) near the frequency of the carrier at 37.5 kHz. This design is necessary to maximize the conversion of the 37.5 kHz pulses to mechanical energy radiated into the water. The 'Q' of aluminum enclosures tuned to resonate near the 37.5 kHz carrier is typically on the order of 50 which provides a "window" at 37.5 kHz +/- ~ 375 Hz.
The acoustic output looks similar to the recorded (and demodulated) signal below. The signal below was produced by Rodney Thomson while analyzing the pulses obtained in the SIO early in the search for MH370. These pulses turned out not to be sourced by 9M-MRO. Note below is copy pasted from the Thomson link.
Rodney Thomson
(NOTE: the 'pinger' is audible at about 3.4kHz however it obviously has been demodulated down from the raw signal observed on the TPL-25 to bring into the audible range)
