| 摘要 |
In a loudspeaker of the kind in which negative feedback derived by monitoring actual displacement of the loudspeaker's electroacoustic transducer or cone is used to compensate for transducer non-linearities, overload protection is provided by two variable gain amplifiers between the audio input and the power amplifier. One variable gain amplifier is within the feedback loop and the other variable gain amplifier precedes the feedback loop. A detection circuit has two thresholds close to but below the level at which clipping will begin. A first threshold level is about 1.0 dB below and the second threshold level is about 0.5 dB below. A first control signal to reduce the gain of the variable gain amplifier preceding the loop is generated when the first, lower threshold level is exceeded and causes the gain of the amplifier to decrease at a relatively slow rate. A second control signal to reduce the gain of the amplifier within the loop is generated when the higher threshold level is exceeded and reduces the gain of the amplifier within the loop very rapidly so that clipping is avoided. The first and second control signals may each be derived by a threshold amplifier and a charging circuit, the first charging circuit having a longer time constant than the time constant of the second charging circuit. Once the overload situation is under control, i.e. the signal is below the second threshold level, the second control signal increases the gain of the amplifier inside the loop, restoring full feedback. The feedback path may derive a first feedback signal component substantially proportional to the position of the electroacoustic transducer, a second feedback signal component substantially proportional to the velocity of the electroacoustic transducer and a third feedback signal component substantially proportional to acceleration of the electroacoustic transducer.
|
