JPS5938645B2 - Saisei Souchi - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION It has been considered to record and reproduce audio signals using a magnetic recording and reproducing device having a rotating magnetic head, similar to a VTR for recording and reproducing video signals.

FIG. 1 shows an example of the recording system, in which an audio signal (frequency band, for example, 20Hz to 20kHz) is supplied from an input terminal 1 through a preamplifier 2 and a preemphasis circuit 3 to an FM modulation circuit 4, where it is converted into an FM signal ( For example, the carrier frequency is 600kHz2 and the occupied band is ±400kHz2)
This FM signal is supplied through the recording amplifier 5 to, for example, two rotating magnetic heads 6A and 6B.

The heads 6A, 6B are constructed similarly to those of a VTR, that is, the heads 6A, 6B are mounted with an angular spacing of 1800 degrees from each other with respect to the rotating shaft 11, and are driven by a motor 12. For example, it is rotated at a rate of 30 times per second.

The magnetic tape T is run obliquely at a constant speed over an angular range of approximately 1800 degrees with respect to the rotating peripheral surfaces of the heads 6A and 6B. Therefore, heads 6A and 6B
The FM signal supplied to the FM signal is sequentially recorded on the tape T as one diagonal magnetic track at a time of just over 1/60 seconds.
Further, at this time, for example, two pulse generating means 13A and 13B are provided for the rotating shaft 11, and these generating means 1
Pulses are taken out from 3A and 13B alternately every half rotation of the heads 6A and 6B, and these pulses are supplied to a mixing circuit 15 through pulse amplifiers 14A and 14B, and are converted into pulses every half rotation of the heads 6A and 6B. , this pulse is supplied to the magnetic head IT through the recording amplifier 16, and this pulse is recorded on the edge of the tape T as a magnetic track along its length.

On the other hand, FIG. 2 shows an example of the reproduction system, in which the heads 6A and 6B are rotated by the motor 12 at the same speed as during recording.

Pulses are extracted from the mixing circuit 15 every half rotation of the heads 6A and 6B, and this pulse is supplied to the phase comparator circuit 18, and the head 17 outputs pulses from the tape T.
The pulse recorded at the edge of be compared. This phase comparison output is then supplied to the motor 12 to drive the heads 6A and 6.
The rotational phase of B is controlled so that heads 6A, 6B correctly scan the magnetic track of the FM signal on tape 7. Therefore, from heads 6A and 6B, heads 6A and 6
FM signals are alternately reproduced every half rotation of B, and these reproduced FM signals are supplied to the switch circuit 22 through the reproduction amplifiers 21A and 21B, and the amplifier 14
Pulses are taken out from A and 14B alternately every half rotation of the heads 6A and 6B, and these pulses are supplied to the switching signal forming circuit 20 to form a rectangular wave signal that is inverted every half rotation of the heads 6A and 6B. This rectangular wave signal is supplied to the switch circuit 22 as its control signal.

Therefore, the FM signals reproduced by the heads 6A and 6B are continuously taken out from the switch circuit 22. The FM signal from this switch circuit 22 is transmitted to the limiter 23.
The demodulated audio signal is supplied to the FM demodulation circuit 24 through the FM demodulation circuit 24 to demodulate the original audio signal, and the demodulated audio signal is supplied to the low-pass filter 25 to remove the carrier component. It is taken out.

In this case, since the audio signal is recorded and reproduced in the form of an FM signal with a high carrier frequency, various characteristics such as the dynamic range, frequency characteristics, and distortion characteristics can be made excellent.

Furthermore, since recording and reproduction are performed using a rotating head, the recording density can be increased and recording and reproduction can be performed for a long time.
However, in such a rotary head type device, if there is a skew or the like, switching noise will be included in the demodulated audio signal when the switch circuit 22 is switched.

In other words, for the sake of simplicity, if we consider the FM signal in an unmodulated state (no audio signal), if there is no skew,
The FM signal from the switch circuit 22 has a continuous waveform even at Ta and tb when the switch circuit 22 is switched, as shown in FIG. 3A.

However, if there is a skew, the FM signal from the switch circuit 22 will be transmitted to the switch circuit 22 as shown in FIG. 3B.
When switching, Ta and tb become discontinuous. In this way, when the FM signal is discontinuous, its phase changes rapidly as shown in Figure 3C, but demodulating the FM signal means is to differentiate the waveform of .

Therefore, when the FM signal (Bf) in FIG. 3 is demodulated, the waveform of the demodulated signal (audio signal) becomes a differential waveform of the waveform in FIG. 3C, as shown in FIG. The signals include Ta and tb when the switch circuit 22 is switched.
Pulse switching noise Sd due to the switching
will be included. In such a case, generally, a gate circuit and a hold circuit are provided in the audio signal path after the demodulation circuit 24, and the audio signal path is cut off during the period of the switching noise Sd to remove the switching noise Sd. At this time, the original audio signal is missing, so during this missing period, the level of the audio signal is held at the previous level.

However, in this method, since the deleted audio signal is corrected by holding it at the level immediately before it, distortion inevitably occurs in the waveform, and the high-fidelity recording and reproduction becomes meaningless. In view of these points, the present invention attempts to prevent switching noise from being included without impairing the special range of the demodulated signal.

Therefore, in the present invention, the harmonic components of the switching noise Sd are extracted, the harmonic components are shaped to form a cancel signal having the same waveform as the switching noise Sd, and this cancel signal is used to eliminate the switching noise S.
d is removed. An example of this will be explained below.

In FIG. 4, the audio signal from demodulation circuit 24 is supplied to bandpass filter 31. In FIG.

This filter 31 has a passband of, for example, 20kHz to 1
00 kHz, that is, it has a characteristic that it does not pass the audio signal and the carrier component, but allows the harmonic component of the switching noise Sd to pass. Therefore, the harmonic component Se of the noise Sd is extracted from the filter 31, but since this harmonic component Se is extracted by the high-pass characteristic of the filter 31, the harmonic component S
The waveform of e is a waveform in which the noise Sd is differentiated by its high-pass characteristic, that is, a waveform as shown in FIG. 3E. Then, this harmonic component Se is supplied to the switching circuit 32, and the rectangular wave signal from the forming circuit 20 is also supplied to the switching signal forming circuit 30. As shown in FIG. Harmonic component S
A signal Sf that is rising for a half cycle period of e is formed, and this signal Sf is supplied to the switch circuit 32 as its control signal, and the switch circuit 32 is turned on only during the period that the signal Sf is rising. .

Therefore, as shown in FIG. 3G, the first half cycle portion Sg of the harmonic component Se is taken out from the switch circuit 32 at each time point Ta and tb. This extracted signal Sg is then supplied to the hold circuit 33 and its level is held, that is, as shown in FIG. 3H, for example, it becomes the same level as the signal Sg at time points Ta and tb,
Due to the time constant, the signal Sh becomes approximately O level at the next time Tb, ta, and this signal Sh is supplied to the differentiating circuit 34 to produce a differentiated pulse S1 at the time Ta, tb as shown in FIG. 3. It is said that

This pulse Si is then supplied to the subtraction circuit 35, and the audio signal from the filter 25 (containing the switching noise Sd) is supplied to the subtraction circuit 35, where the pulse Pi is subtracted from the audio signal.

In this case, since the signal Sg is a harmonic component of the noise Sd, the level of the signal Sg corresponds to the level of the noise Sd, while the level of the signal Sh at times Ta and tb is equal to the level of the signal Sg.

The level of the signal Sh at times Ta, tb therefore corresponds to the level of the noise Sd. Also at this time, the signal Sg
Holding also means integrating the signal Sg, and therefore the waveform of the signal Sh at times Ta and tb becomes a waveform that approximates the phase characteristic shown in FIG. 3C. Since the pulse Pi is a pulse obtained by differentiating such a signal Sh, it is a pulse having a waveform similar to that of the switching noise Sd.

Since this pulse Pi is subtracted from the audio signal in the subtraction circuit 35, the switching noise Sd included in the audio signal is canceled out by the pulse Pi. An audio signal free of noise Sd is taken out, and this is taken out to a terminal 28 through a de-emphasis circuit 26 and an amplifier 27. In this way, it is possible to reproduce an audio signal without switching noise Sd. In this case, according to the present invention, a pulse P having a waveform similar to that of switching noise Sd is reproduced.
Since the switching noise Sd contained in the audio signal is canceled by the pulse Pi, the audio signal does not suffer from waveform distortion.

Furthermore, in the filter 31, the harmonic component Se of the switching noise Sd is taken out, and this harmonic component Se is shaped to form the canceling pulse Pl, so that the canceling pulse Pi can be reliably formed. , no pulse Pi is formed from the audio signal.

Therefore, only the switching noise Sd can be reliably removed. In the above description, the audio signal from the filter 25 is supplied to the high-pass filter to remove the harmonic component Se.
You can also take it out.

In addition, if the input signal is an FM signal, and this FM signal is recorded on a magnetic sheet or the like as a separate track for each unit period, and continuous playback is performed by switching the playback output of the track, this is true. The invention can be applied.

[Brief explanation of drawings]

Fig. 1 is a system diagram of an example of a recording device, Fig. 2 is a system diagram of an example of a reproducing device, Fig. 3 is a waveform diagram for explanation, and Fig. 4 is a system diagram of an example of a reproducing device.
The figure is a system diagram of an example of the present invention. 24 is an FM demodulation circuit, 31 is a bandpass filter, 33
34 is a hold circuit, and 34 is a differential circuit.

Claims (1)

[Claims] 1. An FM demodulation circuit that demodulates the reproduced FM signal in a reproduction device that converts an input signal into an FM signal and reproduces the input signal from the FM signal recorded in a different track for each unit period. a filter with at least a high-pass characteristic that extracts signal components outside the band of the input signal from the demodulated signal of this FM demodulation circuit; a hold circuit that holds the signal components extracted by this filter; and an output signal of this hold circuit. and a differentiating circuit for differentiating the signal, the reproducing device is configured to cancel switching noise contained in the demodulated signal by subtracting an output signal of the differentiator from the demodulated signal.