JPH087575B2 - Automatic piano - Google Patents

Description

Description: TECHNICAL FIELD The present invention relates to an automatic piano capable of half-pedal driving, and more specifically, recording of pedal performance data and automatic performance by the pedal is performed by individual piano pedals. The present invention relates to an automatic piano that can be performed uniformly without being affected by the characteristics of the system.

"Prior Art" Generally, in an automatic piano, performance data recorded on a floppy disk or the like is read out, and a key solenoid or a pedal solenoid is driven according to the performance data.

By the way, in the case of driving the pedal, if there are only two kinds of control states of "stepping on" and "release", the pedal solenoid may be controlled by either ON or OFF. Therefore, in the recording of performance data in such a mode (two-step mode), "stepping on" the pedal is performed.
It suffices to detect and record two types of performance states of "and release", and in reproducing, the pedal solenoid may be turned on / off based on the recorded data.

On the other hand, in order to further improve the reproducibility of the performance, it is necessary to reproduce the so-called half pedal, and in order to create such performance data, it is necessary to continuously detect and record the pedal depression amount. . Then, during the automatic performance, the pedal solenoid is driven by an amount corresponding to the recording data. Such driving of the pedal solenoid can be performed by stepwise feedback controlling the electric power supplied to the pedal solenoid. In addition, this control is
Generally, pulse width modulation control is often used, but it can also be performed by controlling voltage, current, or the like.

[Problems to be Solved by the Invention] By the way, the static characteristics and dynamic characteristics of the pedal system are characteristics unique to each piano, and the type of pedal (for example,
Loud pedal and shift pedal). Furthermore, it is difficult to make the characteristics of the solenoid that drives the pedal uniform, and the relationship between the solenoid drive signal and the actual displacement amount is not linear.

Therefore, there is a problem that the pedal operation at the time of recording cannot be faithfully reproduced even if the performance data recorded by using one automatic piano is reproduced by another automatic piano.

In order to solve such a problem, the present applicant previously proposed Japanese Patent Application No. 2-9551 "Automatic Piano".

In the "automatic piano" of this application, a conversion table between the pedal position and the drive signal is created by actual measurement, and a drive signal for controlling the position of the pedal is generated by referring to this conversion table.

According to this configuration, it is possible to accurately drive the pedal to a position according to the supplied target position data regardless of the individual difference of the piano. That is, the pedal position can be faithfully stored / reproduced. However, this configuration also has the following problems.

That is, the area from the pedal position when the driving force starts to be transmitted to the damper head when the loud pedal is pushed down to the pedal position when the damper head is completely separated from the string, that is, the half pedal area is There is a problem in that recording / reproduction cannot be performed faithfully to the sound actually played, because each pedal is different. In other words, even if the pedal position is the same, the damper head touches the strings in some pianos, but in some pianos, the damper heads move away from the strings, resulting in different sounds being produced by the piano. There was something that happened.

It should be noted that such differences in individual pianos are due to variations in damper weight and adjustment, and the magnitude of frictional force.

The present invention has been made in view of the above circumstances,
The half pedal area can be set appropriately for any piano with any characteristics, and the same half pedal effect can be obtained by playing back on any other piano using the data recorded on one piano. The purpose is to provide an automatic piano that can.

[Means for Solving the Problems] In order to solve the above problems, in the invention according to claim 1, a pedal drive solenoid that drives a pedal, a pedal position detection sensor that detects the position of the pedal, and the pedal drive. By at least changing the value of the drive signal for driving the solenoid and supplying it to the pedal drive solenoid and detecting the position of the pedal corresponding to each value of the drive signal by the pedal position detection sensor, at least half At least among the pedal characteristic detection means for determining the correlation between the position of the pedal including the pedal region and the drive signal value, and the correlation between the position of the pedal and the drive signal value determined by the pedal characteristic detection means. Correction means for arbitrarily correcting the correlation of the half pedal area and a target for generating target position data indicating the pedal target position Based on the position data generating means and the target position data generated by the target position data generating means, a drive signal is generated in accordance with the correlation between the pedal position corrected by the correcting means and the drive signal value. Drive signal supply means for supplying the drive signal to the pedal drive solenoid.

Further, the invention according to claim 2 is the automatic piano according to claim 1, further, the position of the pedal is pushed down to a predetermined position between the maximum depression position and the half pedal region, and then the amount of depression is gradually decreased. And a control means for generating and outputting a predetermined keystroke signal as well as outputting a drive signal for reducing

According to a third aspect of the invention, the pedal drive solenoid for driving the pedal, the pedal position detection sensor for detecting the position of the pedal, and the value of the drive signal transmitted to the pedal drive solenoid are sequentially changed and the drive is performed. A normalization table for normalizing the pedal position detected by the pedal position detection sensor, which is created by detecting the position of the pedal corresponding to each value of the signal by the pedal position detection sensor, and the normalization table. Correcting means for arbitrarily correcting at least half pedal area data of the normalization table, and referring to the normalization table based on the pedal position detected by the pedal position detection sensor, normalization data obtained by normalizing the pedal position is obtained. And a normalizing means for outputting.

According to a fourth aspect of the invention, in the automatic piano according to the third aspect, further, storage means for sequentially storing the normalized data output from the normalizing means, and the normalized data stored in the storage means. Reading means for sequentially reading the data, an inverse conversion means for converting the normalized data read by the reading means into data indicating a pedal position in accordance with the correction content of the data by the correction means, and conversion by the inverse conversion means. Drive signal supply means for supplying a drive signal generated based on the generated data to the pedal drive solenoid.

[Operation] Referring to FIG. 1, according to the above configuration, the pedal characteristic detecting means supplies the sequentially changing drive signal to the pedal drive solenoid to drive the pedal, and the pedal signal is set to each value of the drive signal. The corresponding pedal position is detected by the pedal position detection sensor. As a result, the pedal characteristic detecting means is
The correlation between the pedal position including at least the half pedal area and the drive signal value is determined. Further, the correction unit arbitrarily corrects at least the correlation in the half pedal region among the correlations between the position of the pedal determined by the pedal characteristic detection unit and the drive signal value. Then, the drive signal supply means generates a drive signal corresponding to the correlation between the corrected pedal position and the drive signal value based on the target position data generated by the target position data generation means, and generates a pedal drive solenoid. Supply and drive the pedal. That is, the control of the pedal position with reference to the target position data is performed according to the correlation between the pedal position and the drive signal value, which is determined and corrected based on the actual measurement value.

Further, the control means gradually returns from the state where the pedal is pressed down, and outputs a keystroke signal for driving a predetermined key. Therefore, it is possible to correct the correlation between the pedal position and the drive signal value by using the key tap sound.

Further, the normalizing means refers to a normalization table for normalizing the pedal position detected by the pedal position detection sensor, which is created based on the actual measurement value and at least the data of the half pedal area is arbitrarily corrected, It outputs normalized data obtained by normalizing the pedal position based on the pedal position detected by the pedal position detection sensor.

Further, the storage unit sequentially stores the normalized data output from the normalization unit, the reading unit sequentially reads the normalized data stored in the storage unit, and the inverse conversion unit according to the correction content of the data by the correction unit. The normalized data read by the reading means is converted into data indicating the pedal position. The drive signal supply means in the invention according to claim 4 supplies the drive signal generated based on the data changed by the inverse conversion means to the pedal drive solenoid. Therefore, the normalized data is inversely converted into data indicating the pedal position in accordance with the correction content of the normalization table referred to when the normalizing means outputs the normalized data, and the data indicating the inversely converted pedal position is obtained. The position of the pedal can be controlled based on

[Examples] Examples of the present invention will be described below with reference to the drawings.

(1: Configuration of Embodiment) FIG. 2 is a block diagram showing the configuration of an embodiment of the present invention. In the figure, the automatic performance grand piano GP is
Under the control of the controller 6, an automatic performance is performed according to the performance data, and when the player performs a performance, the performance data corresponding to this performance is supplied to the controller 6. FIG. 3 is a side sectional view showing an external configuration of the automatic performance grand piano GP and peripheral devices. As shown in the figure, the controller 6 is built in the automatic performance grand piano GP, so that performance data and various signals can be exchanged with the wagon 8 having built-in peripheral devices via the cable 7. Has become. The controller 6 is provided separately in the key drive unit of the automatic performance grand piano GP into a control unit 6a and an I / O unit 6b. As shown in FIG. 2, the control unit 6a includes a CPU 9 that controls each unit of the automatic grand piano GP, a ROM 10 that stores programs used by the CPU 9, and a RAM 11 that temporarily stores various data and a conversion table described later. Has been done. This control unit 6a, via the I / O unit 6b,
It is connected to an automatic performance grand piano GP and a floppy disk driver (hereinafter abbreviated as FDD) 12, and records / reads performance data.

Next, the solenoid 20a shown in FIG.
1a is driven, and FIG. 4 is a rear view thereof.
As shown, the end of the loud pedal 21a is rotatably connected to the lower end of a vertically movable rod 22a via a universal joint, and the upper end of the rod 22a has a solenoid 20a.
Is rotatably connected to the lower end of the plunger 20ap.
On the other hand, the upper end of the plunger 20ap of the solenoid 20a is the rod 2
The rod 23a is connected to the damper driving mechanism in the piano body. The solenoid 20b is the shift pedal 21b.
Is a solenoid for driving the motors, and each driving force is transmitted by the rods 22b and 23b as in the case of the loud pedal 21a. Sensors 35a and 35b for detecting the displacement amounts of the plungers 20ap and 20bp are provided above the solenoids 20a and 20b, respectively. This sensor 35a, 35b
Consists of a gray scale that moves up and down integrally with the plungers 20ap and 20bp, a light emitting element that irradiates the gray scale with light from the fixed side, and a light receiving element that receives the transmitted light from this light emitting element. The displacement amount of the plunger 20ap, 20bp from the output signal of the light receiving element, that is,
The displacement amount of each pedal 21a, 21b is detected. Next, the sostenuto pedal 30 and the loud pedal 21a
And a shift pedal 21b, and is connected to a single rod 31 that is vertically movable. The sensor 32 is a sensor that detects a predetermined position of vertical movement of the rod 31, that is, the start and end of the operation of the function of the sostenuto pedal 30.

(2: Operation of Embodiment) Next, the operation of this embodiment with the above configuration will be described. Although there are various operation modes in this embodiment, the operation modes directly related to the invention will be described in detail, and the other operation modes will be described only in outline.

A: Piano characteristic measurement and control coefficient calculation operation First, the characteristic measurement principle will be described.

First, the PWM signal supplied to the solenoid 20a is sequentially increased (the pulse width is increased) to move the loud pedal 21a to the upper limit, and then the PWM signal is decreased to return to the original position. The fifth relationship with quantity
As shown in the figure. In this figure, the horizontal axis shows the control code for instructing the magnitude of the PWM signal. This control code is a value from [00] to [7F] in hexadecimal notation. It should be noted that the value of the control code is not limited to the above range, but is arbitrary. Further, the vertical axis of FIG. 5 shows the amount of displacement. This displacement amount is obtained by converting the output signal of the sensor 35a into a digital value of 128 steps by the A / D converter.

The PWM signal and the amount of displacement have the characteristics shown in the figure.
This is because of the elastic characteristics of the members that make up the pedal drive system and the associating between the members. The area where the curve becomes flat in the ascent process is the area of the half pedal. Here, the state of the half pedal will be described. FIG. 6 is a side view showing the outline of the drive system of the loud pedal. In the figure, when a current corresponding to the PWM signal is supplied to the solenoid 20a, the plunger 20ap rises according to the value. When the plunger 20ap rises, the lever 40 moves the fulcrum 41
And the rod 42 is pushed up. When the rod 42 is pushed up, the lever 43 rotates around the fulcrum 44 and pushes up the damper wire 45. When the damper wire 45 is pushed up, the damper head provided at the upper end of the damper wire 45
46 rises and moves away from string 47. And damper head 46
The state where is completely away from the string 47 is the damper-on area. On the other hand, the half pedal area is a period from when the driving force is first transmitted to the damper head 46 to when the damper head 46 is separated from the string 47. In this half pedal region, the increase rate of the plunger 20ap decreases even if the value of the PWM signal to the solenoid 20a is increased. Therefore, in the ascending process, the increase rate of the displacement amount decreases in the half pedal region as shown in FIG.

Next, in the descending process, as shown in FIG. 5, the displacement amount is initially flat, but after that, the displacement amount is processed comparatively smoothly to return to the original position.

In the embodiment, the value of the PWM signal is increased step by step, and the drive amount in the rising process is measured. In addition, from the measured displacement amount, half point x _HP ([3
[0] vicinity), and further detects the play area of the loud pedal 21a.

On the other hand, when the characteristics of the automatic piano are as shown in Fig. 22 (b), the half point x _HP is not clear, so it is tentatively determined to a predetermined value and then adjusted while checking with the ear. .

The above is the principle of measurement and detection.
The measurement and detection of b is based on exactly the same principle.
However, as shown in FIG. 7, the characteristic of the shift pedal 21b is almost linear in the rising region,
It does not detect half points like the loud pedal.

Next, the conversion table used in the embodiment will be described.

The conversion tables used in the embodiment include the following, which are created by the processing described later.

The first conversion table is a position-PWM conversion table for converting the pedal position data xi into a control code of a PWM signal based on the above measurement result. This position-PWM conversion table is used when converting the position data read from the disk during automatic performance into the control code of the PWM signal. By using this position-PWM conversion table, the static characteristics of the pedal system that differ depending on the piano and pedal type are compensated. Therefore, the position-PWM conversion table is created for each piano and each pedal, and is appropriately updated to compensate for the change over time. In addition,
Actually, when the position-PWM conversion table is created, some data correction is performed to correct the nonlinear characteristics of the solenoid.

The second conversion table is a normalization conversion for converting the position data of about [128] stages detected by the sensors 35a and 35b into the data of [16] stages and correcting the pedal characteristics according to the conversion. It's a table.

For example, for the loud pedal 21a, as shown in FIG. 8, the position data xi of about [128] stages, the assobi region, the half pedal region, and the damper-on region are detected by the above-described measurement. The value is normalized and assigned to 16-stage position data that is further compressed as appropriate. That is, xa in the figure is the position of the pedal release point, xb is the position of the half pedal starting point, xc is the position of the starting point of the damper on, and xd is the position at the time of maximum depression. It is assigned to each point of Xa, Xb, Xc, Xd. In the normalized data Xi in this case, a portion corresponding to the half pedal area is finely divided, and a portion corresponding to another area is coarsely divided. This is because, in the half pedal region, a delicate nuance is expressed in the performance, and it is necessary to perform high-precision control. on the other hand,
The other areas are coarsely divided because it is not necessary to perform position control little by little.

As for the shift pedal 21b, the normalized data is linearly divided into 16 sections, as shown in FIG. This is because, as described above, the shift pedal 21b has a linear characteristic in the ascending process. Further, the pedal release point xa and the maximum depression point xd are made to correspond to the normalized data Xa and Xd, respectively, as in the case of the loud pedal described above. It corresponds to Xa. These normalized conversion tables are used when recording a performance.

By the way, the reason why the normalization conversion table is used to correct and compress the data into 16 stages is as follows.

First, if you try to record the amount of pedal displacement using 128-step digital data, the amount of pedal data in the recording medium will be extremely large, and for example, a disc that can normally record about 70 minutes cannot be recorded on a disc for about 15 minutes. The inconvenience occurs. Therefore, the data is compressed into 16 levels. However, if the pedal data is simply compressed and recorded, there arises a problem that the reproducibility of the pedal performance deteriorates due to the difference in the characteristics of the pedal system of each piano. Therefore, each area of the measured position data xi, which is a unique value in each piano and pedal, is converted and compressed so as to correspond to a predetermined area of the normalized data Xi. The half pedal area of the normalized data Xi is finely divided so that data can be recorded / reproduced satisfactorily.

Next, the third conversion table is an inverse normalization conversion table that performs the reverse conversion of the conversion tables shown in FIGS. 8 and 9. That is, it is a table for converting the value of the normalized conversion data Xi into xi which is a correction value peculiar to the piano. However,
The input of the denormalization conversion table is data of 128 steps, and the output is data of 128 steps or less. Therefore,
In the actual processing, the 16-step normalized data Xi read from the recording medium is converted into 128-step normalized data Xi by interpolation processing and then supplied to the denormalization conversion table.

Next, the calculation coefficient will be described.

Position data xi obtained by the above inverse conversion table
Is converted to a PWM signal control code (hereinafter referred to as control code PWMs) using the position-PWM conversion table, and a PWM signal according to this control code PWMs is supplied to the solenoids 20a and 20b. It is possible to perform the pedal drive with compensation. If the positions of the plungers 20ap, 20bp are feedback-controlled using the output signals of the sensors 35a, 35b during this driving, the pedal position control can be performed with high accuracy. However, when the pedal moves quickly, the feedback loop cannot follow and the pedal position control is disturbed. Therefore, it is conceivable to increase the gain of the feedback loop, but the position control of the plungers 20ap and 20bp is one-way control, so if the gain is increased, oscillation will occur. This is because although the protrusion amount of the plunger 20ap, 20bp can be controlled, the return is a return due to gravity etc., so if feedback control is performed with a high gain, it will not converge to the target position and will be in a hunting state. is there.

Therefore, in this embodiment, the velocity data i is calculated by differentiating the position data xi output from the inverse normalization conversion table. Then, the speed data i is multiplied by the coefficient K1, and the multiplication result is added to the control code PWMs as the control code PWM1 for correction.
The M signal is supplied to the solenoids 20a and 20b. That is, the feedforward control is performed using the speed data i, and the coefficient K1 becomes a control coefficient in the feedforward control. The coefficient K1 is the solenoid
A PWM signal having a constant increase rate is applied to 20a and 20b, and the speed is determined based on the speed data i obtained at this time.

Since the speed correction is performed as described above, even if the pedal speed is high, the control that follows this is performed. However, for example, considering the start of stepping on the loud pedal 21a and the shift pedal 21b, the initial speed is 0 but a driving force is required, and the acceleration thereafter becomes large. In this case, it is necessary to control the pedal position according to the rapid increase in speed, but the initial speed is 0
Therefore, the speed data cannot be corrected and the control cannot follow. Such a situation occurs not only at the beginning of stepping, but also when other accelerations increase.

Therefore, in this embodiment, the acceleration data i is calculated by differentiating the position data xi output from the inverse normalization conversion table color twice. Then, the acceleration data i is multiplied by the coefficient K2, and the multiplication result is the control code PW for correction.
M2 is further added to the above acceleration result (PWMs + PWM1), and the PWM signal according to this addition result is applied to the solenoid 20
It is supplied to a and 20b. That is, the feedforward control is performed using the acceleration data i. The coefficient K2 is a control coefficient in this case, and has a value corresponding to the mass of the pedal drive system. Further, the coefficient K2 is determined based on acceleration data i obtained when, for example, a PWM signal that increases at a constant acceleration is applied to the solenoids 20a and 20b, or a PWM signal of a constant magnitude is instantaneously applied. To be done.

Further, the deviation between the signals output from the sensors 35a and 35b and the position data xi is taken, this deviation is multiplied by a coefficient K3, and this result is used as a correction control code PWM3 to correct the above acceleration result. ing. The coefficient K3 corresponds to the gain of the feedback loop. The value of the coefficient K3 is previously determined to be a stable value that does not oscillate by experiments or the like.

Further, in this embodiment, in order to improve the responsiveness, the speed data is obtained from the rate of change of the position data x with respect to the position, and the acceleration data is obtained from the rate of change of the speed data. Then, the difference Δ from the speed data i is obtained, and this value is multiplied by the coefficient K4 to obtain the control code PWM4. Further, the difference Δ between the acceleration data i and the acceleration data is obtained, and this value is multiplied by the coefficient K5 to obtain the control code PWM5. And these control codes PWM4,
Accelerate PWM5 to (PWMs ＋ PWM1 ＋ PWM2), and the result of this addition is PWM ＝ PWMs ＋ PWM1 ＋ PWM2 ＋ PWM3 ＋ PWM4 ＋ PWM5 ＝ PWMS ＋ K1 ・ i ＋ K2 ・ i ＋ K3 ・ Δx ＋ K4 ・ Δ ＋ K5 ・ Δ (however, Δ is the deviation) …… (1) The control code PWM is calculated.

Next, the half point and half region measurement, conversion table creation and coefficient calculation operations will be described.

This operation is performed based on the flowchart shown in FIG. First, the type of pedal is determined in step SP1. That is, it is determined which of the loud pedal 21a and the shift pedal 21b is to be measured. Which measurement is performed is performed according to an instruction of the operator (an instruction by operating the operation unit) or a predetermined program.

When it is determined that the "loud pedal" is selected in step SP1, the process proceeds to step SP2. In this step SP2, first, the control code is sequentially increased from [00] to [7F]. As a result, the PWM control section provided in the I / O unit 6b outputs the PWM signal corresponding to the control code to the solenoid 20a. This allows the plunger 20a
As p increases, the amount of movement is detected by the sensor 35a. The position signal output from the sensor 35a is I / O.
It is converted into a digital signal by the A / D converter in the unit 6b and supplied to the CPU 9 as position data xi. Next, proceeding to step SP3, the CPU 9 creates a position-PWM variation table from the relationship between the value of the control code and the position data xi, stores it in the RAM 11, and proceeds to step SP4. Step SP
In 4, it is determined whether or not there is a region where the increase rate of the position data (that is, pedal stroke) xi becomes smaller than a predetermined constant α. This judgment is "YES"
In the case of, that is, when the piano has the characteristics shown in FIG. 22 (a) or FIG. 5 and the half point is clear,
Go to step SP5. In step SP5, from the value of the position data xi that satisfies the condition that the increase rate of the pedal stroke xi is smaller than α, the half pedal area (shown in FIG. 8 is shown.
xb to xc) are set. The half point x is detected by detecting the portion where the rate of increase in pedal stroke xi becomes extremely small.
Set _HP . Next, move to step SP6 and position data x
Based on the condition that the rate of increase of i is below a certain value, release point xa
And the maximum stepping point xd (see FIG. 8) are set, and the process proceeds to step SP7. In step SP7, an inverse normalization conversion table is created. That is, a table for converting the normalized data into the position data xi is created based on the relationship between the measured position data xi and the control code. For example, if the normalized data is as shown in FIG. 11C and the corresponding position data xi is as shown in FIG. 11D, the normalized data is shown in FIG. ) Create a table to convert to the value shown in. The points Xa to Xd of the normalized data are stored in advance. When the processing of this step SP7 ends,
Proceed to step SP7a and display that automatic measurement is completed. This display is performed on the display unit (not shown) of the controller 6.

On the other hand, if it is determined as "NO" in step SP4, that is, if the piano has the characteristics shown in Fig. 22 (b) and the half point is not clear, step SP20
Proceed to. In step SP20, the same processing as in step SP6 is performed, and then in step SP21, half points are provisionally set. This process is performed as follows. First, the half-point Xa− of the normalized data
The ratio between Xd is m ₂ / m _{1 as} shown in Fig. 11 (c). Then, temporarily set half point x _HP ′ xa−x
The position of the half point x _HP ′ is set so that m ₂ / m ₁ = n ₂ / n ₁ when the ratio between d is n ₂ / n ₁ . Similarly, the half pedal area xc′−xb ′ is provisionally set, and further, the denormalization conversion table is provisionally set (step SP
twenty two). Next, in step SP23, a display indicating that the temporary setting has been performed, that is, a display indicating that automatic measurement of the half points is impossible is performed. After this process, or after the process of step SP7a described above, the process proceeds to step SP8.

In step SP8, a PWM signal that increases in an accelerating manner is supplied to the solenoid 20a, or a PWM signal of a constant magnitude is instantaneously supplied to the solenoid 20a, and the resulting position data xi is differentiated twice. Acceleration data i
To create. Then, the coefficient K2 is determined from the value of the acceleration data i. Next, in step SP9, a PWM signal having a constant increase rate is supplied to the solenoid 20a, and the resulting position data xi is differentiated once to create velocity data i. Then, the coefficient K1 is determined from this speed data i. After this processing, the process returns to the main routine (not shown).

On the other hand, in step SP2, when it is determined that the "shift pedal", the processing of steps SP10 to SP16 is performed. These processes are the same as those in steps SP2 to SP9 described above.
However, since the half pedal area is not set for the shift pedal 21b, there is no processing corresponding to steps SP4 and SP5.

B: Half point adjustment process This process is a process for adjusting the half points set as described above. FIG. 12 is a flowchart showing the half point adjustment processing, which is provided in the controller 6. It is activated by pressing a switch (not shown).

First, in step SPd1, a control signal for driving the loud pedal 21a to the damper on position is output. That is, the control code corresponding to the damper on position is set to CPU9.
Is output, and this is converted into position data according to the currently set inverse normalization conversion table. Here, the damper-on position is a position at which the damper is completely separated from the string, and is, for example, an arbitrary position in the range from the maximum depression position xd to the damper-on region lower limit xc. Then, a drive signal corresponding to this position data is created using the position-PWM conversion table. As a result, the loud pedal 21a is driven to the damper on position. Here, if the normalized data of the damper on position is [70], step SP
The state of processing d1 is shown in FIG. The time t ₁ shown in this figure is the end of the process of step SPd ₁ . Next, when 200 ms has elapsed, control proceeds to step SPd2 and the pedal stroke is returned to a predetermined position. Pedal stroke control is
It is performed in the same manner as step SPd1. Now, if the predetermined position is [5E] of the normalized data, by the processing in step SPd2, stroke loud pedal 21a is lowered as shown at time t ₃ of FIG. 13, in step SPd1, SPd2, The reason why the loud pedal 21a is once driven to the damper on position and then lowered to a predetermined position is to stabilize the initial position of the damper when striking the strings. Next, the 12th
Proceeding to step SPd3 shown in the figure, a string striking command is given to the 88th key (highest tone) and a predetermined chord is sounded.
The specified chord is the key number 32 (E note), 38 (A #
Sound), 40 (C sound), 44 (E sound), 47 (G sound), 50 (A #
Sound) and 52 (C sound). The volume of the chord is mf
It is set to about (mesoforte) and is set to, for example, the MIDI standard [40]. Processing in step SPd3 is performed at the time t ₄ when shown in Figure 13.

Next, in step SPd4, the sound generation timer is reset. The sound generation timer is a timer for elapse of the sound generation interval of the 88th key, and a timer IC or a timer means by software processing is used. As will be described later, the 88th key is adapted to be repeatedly depressed every time a sound generation timer passes a predetermined time. In step SPd5, the stroke timer is reset. The stroke timer is a timer for determining the elapse of every 50 ms from the time t ₄ , and the number of times of elapse corresponds to the stroke of the loud pedal 21a. Whether or not 50 ms has elapsed is determined in step SPd6, and the same step SPd6 is cycled until this determination becomes “YES”. If the determination in step SPd6 is "YES", the flow proceeds to step SPd7 to reduce the pedal slotalk value by a predetermined amount, and then, in step SPd8, a drive signal that outputs the reduced current pedal slotalk is output. . Now, if the predetermined amount to be reduced is [01] of the normalized data, step SPd4
At time t ₅ the process is 50ms elapsed from the time t ₄ when took place, the pedal stroke is in a position corresponding to the normalized data [5D]. Then, at step SPd9, time t ₄
It is determined whether or not 1 s has elapsed. This determination is made based on the content of the sound generation timer, and in the case of "NO", returns to step SPd5. Then, thereafter, steps SPd4 to SPd9 are cycled until it is determined to be "YES" in step SPd9. By this circulation processing, the pedal is returned by a predetermined amount every 50 ms (see times t ₅ and t _{6 in} FIG. 13). Then, at time t ₇ when one second has elapsed from step SPd4, the determination at step SPd9 becomes “YES, and step SPd10
Go to. In step SPd10, it is depressed key number 88 again (see time t ₇ shown in FIG. 13). And step SP
The process proceeds to d11, where it is determined whether the stop signal is input.
This stop signal is input when a switch (not shown) provided in the controller 6 is pressed.
If the determination in step SPd6 is "YES", the process ends,
If "NO", the process proceeds to step SPd12, and it is determined whether or not the pedal stroke has become the normalized data "22" (because the half point is [46], the sound is surely stopped at the position [22]. Must have been). If the pedal stroke is not [22] in the normalized data, step
After returning to SPd4, steps SPd4 to SPd12 are cycled until the determination at step SPd12 becomes "YES". By this process, from the time t ₄ each time one second has elapsed, the key 88 is depressed. That is, the highest note is every 1 second (see times t ₇ , t ₉ , t ₁₀ )
Is pronounced by. During this time, the stroke of the loud pedal 21a is decreased by a predetermined amount every 50 ms, and the damper is sequentially lowered. Therefore, the chord sounded at time t ₄ is stopped by the damper that has descended. In this case, if the stroke of the loud pedal 21a decreases corresponding to [01] of the normalized data every 50 ms, if the half point x _HP of the denormalization conversion table is appropriate, a chord will be sounded and then stopped. The standard time to reach is 1.5 seconds. Then, since the position data xi for controlling the stroke of the loud pedal 21a is created by converting the normalized data by the current denormalization conversion table, the half point x _HP (or x of the denormalization conversion table is generated. If ′ _HP ) is not appropriate, the time until the sound stops will deviate from standard time. Here, the operator can know the time from the start of the sound of the chord until the sound is stopped from the number of times and the interval at which the sound of the key number 88 (the highest sound) is sounded, and whether this is longer or shorter than the standard. Can be judged.

Next, when the pedal stroke normalized data becomes [22], the process proceeds to step SPd13, and the correction value of the half point x _HP (or x ′ _HP ) is input. This input is the second
This is done by pressing the switch L or R shown in the figure. When the switch L is pressed once, the half point value increases by [01], and when the switch R is pressed once, the half point value decreases by [01]. It is supposed to do. The operator
As described above, the time from the pronunciation of the chord to the stop can be grasped, so the switch L or R is operated so as to set this as the standard time. Increasing the value of the half point, the position data for the X _HP normalized data x _HP
As the value of increases, the time until the sound stops becomes longer, and conversely, when the value of the half point decreases, the time until the sound stops becomes shorter.

Next, in step SPd14, the value of x _HP (or x _'HP) increases or decreases according to the correction value of the half point, in step SPd15, normalized conversion table which is currently set in accordance with the correction value of the half point And the inverse normalization conversion table are corrected. The correction in this case corrects the entire half area according to the change of the half point. That is, the values of xb and xc shown in FIG. 11 (d) are also corrected according to the variation of x _HP . However, the interval between xb and xc is constant. The same applies to the case of provisional decision (see FIG. 11B). In addition, each tentatively determined point x'c, x'HP,
After correcting x'b, x is set as a normal setting.
Represented by c, x _HP , xb.

The above is the half point adjustment processing. This process is performed to make the value normal after making the tentative decision (see FIG. 11 (b) → (a)). Even when the half-point automatic measurement process is performed, the half-point automatic measurement process may be performed to make the measurement value more definite (11th
(See (d) → (e)).

C: Half Point Confirmation Processing This processing is processing for confirming the set half points by ear, and has two modes, a reproduction test mode and a trill performance mode. Note that this process is usually performed after the process shown in FIG. 10 or the process shown in FIG.

Reproduction Test Mode FIG. 14 is a flowchart showing the reproduction test mode, which is activated when a switch (not shown) in the controller 6 is pressed. First, in step SPe1, the control signal for driving to the pedal on position is output, and then in step SPe1.
At SPe2, a control signal to return the pedal to the pedal half point position is output. In this step SPe1 and 2, the normalized data Xd and X _HP are converted to position data xd and x _HP according to the currently set inverse normalization conversion table.
Convert to. Here, Xd is [70], X _HP is shown at time t ₁₁ ~t ₁₃ as the result of the FIG. 15 step SPe1,2 the case of [46]. Next, the process proceeds to step SPe3, it is determined whether or not a predetermined time has elapsed, and step SPe3 is determined until the predetermined time elapses.
Circulate. Then, after a predetermined time has elapsed, step SPe3
If the judgment is “YES”, then proceed to step SPe4 and select the number 8
Keying of the 8 keys is instructed, and a keying signal for keying a predetermined chord is output. The predetermined chord is the same as in the case of step SPd3 described above. FIG. 15 shows a case where 500 ms is set as the predetermined time (see times t _{13 to} t _{14 in the} same figure). Then, in step SPe5, the sound generation timer is reset, and in step SPe6, it is determined whether or not 1 second has elapsed. When 1 second has elapsed, the determination in step SPe6 becomes “YES”, and the number 8 is again set in step SPe7.
The 8 key is pressed (see time t15 in FIG. ₁₅ ). next,
In step SPe8, it is determined whether a stop signal has been input. The stop signal is input by the switch operation in the controller 6 as described above. After that, step SP
Steps SPe5 to SPe are cycled until the determination of e8 becomes “YES”. This circulation, key number 88 every second is keyed as shown in Figure 15 (see time t ₁₅ ~t _17). By the way, the chord sounded at the time t ₁₄ is rapidly attenuated because the piano is in the half pedal state. Here, the operator can confirm by ear how long the chord is silenced, and can know the quality of the half point.

Next, when the stop signal is input, the determination in step SPe8 becomes "YES", the pedal stroke is returned to 0, and the processing ends (step SPe9).

Trill Performance Mode FIG. 16 is a flowchart showing the processing contents of the trill performance mode. This subroutine is the controller 6
It is activated by pressing a predetermined switch inside.

First, in step SPf1, the registers t and S each have 2
00 and [10] are substituted. The registers t and S are set in the RAM 11 shown in FIG. 2, for example. Next, the stroke timer is reset (step SPf2), and it is determined whether t seconds have elapsed (step SPf3). Since t is set to 200 in the initial state, it is determined whether or not 200 ms has elapsed. This judgment of the elapsed time is made by reading the time measured by the stroke timer. Then, the process proceeds to step SPf4, and the content of the register S is added to the previous stroke command value (0 in the initial state) to obtain a new stroke command value. Then step SP
Go to f5, increase the actual stroke at a constant speed, and
A control signal is output after 0 ms so that it matches the current stroke command value. Further, the generation of the control signal in steps SPs4 and 5 is performed based on the currently set inverse conversion table. Next, at step SPf6, it is judged if the stroke is [00]. In this case, since the pedal drive has been started in step SPf5, this determination becomes "NO", and the process proceeds to step SPf7. Here, it is determined whether or not the stroke has reached [60]. Immediately after the start of driving the pedal, the stroke is less than [60], so this determination becomes "NO", and the routine proceeds to step SPf8. Step
In SPf8, it is determined whether or not the stroke is 20, but this is also "NO" for the same reason as above. Therefore, the process returns to step SPf2, and the stroke timer is reset again. And 200m at step SPf3
s is passed. As a result, the stroke of the loud pedal 21a at time t ₂₀ shown in FIG. 17 becomes [10]. Next, when moving to the stroke SPf4, the stroke command value becomes [1
0] is added to become [20], and the loud pedal 21a is driven upward toward the stroke [20] by the processing of step SP + 5. Then, the process goes to step SPf2 again via steps SPf6, 7, 8 and the stroke timer is reset. Next, 200 ms is elapsed in step SPf3, As a result, as shown at time t ₂₁ of FIG. 17, the stroke of the loud pedal 21a is [20].

Then, in step SPf4, the stroke command value is
Increased to 30 and stepped through the stroke SPf 5,6,7
Proceed to SPf8. The determination at step SPf8 is “YES”, and as a result, the content of the register S is −10 at step SPf9.
It is determined whether or not. At this point, the contents of register S are
Since it is 10, the judgment is “NO”, and the routine proceeds to step SPf10. In step SPf10, 500 is assigned to the register t. Next, when proceeding to step SPf11, the trill performance start processing is performed and the pedal rising speed is 500 m.
Switch to a speed that increases by [10] with s. The trill performance process is a process of tapping the keys of numbers 44 and 45 eight times per second. After this step SPf11, the process moves to step SPf2 again. After that, step SP until the stroke reaches [60]
The processing of f2 to SPf11 is cycled. As a result, time t in FIG.
_As shown from ₂₂ to t ₂₅ , the pedal stroke increases at a constant speed and the trill performance is performed. Then, when the stroke reaches [60], the determination at step SPf7 shown in FIG. 16 becomes "YES", the routine proceeds to step SPf12, and -10 is substituted into the register S. After that, it returns to step SPf2. As a result, the stroke command value is reduced to 50 in the process of step SPf4, and the stroke is controlled to be lowered by the process of step SPf5. After that, the processes of the strokes SPf2 to SPf8 are repeated until the stroke becomes [20]. As a result, the stroke decreases at a constant rate as shown from time t ₂₅ to t ₂₉ in FIG. During this time, the trill performance will continue. When the stroke reaches [20], the determination at step SPf8 becomes "YES", and the determination at step SPf9 is made. This judgment is "YE
S ”, the process moves to step SPf13. Here, the end processing of the trill performance is performed, and then step SP
At f14, 200 is substituted into the register t. In step SPf14, the stroke descending speed is 200 ms.
It is switched so that it decreases by [10] every time. After this process,
Return to step SPf12. Then, in step SPf3, 2
00ms has elapsed. As a result, the stroke becomes [20] at time t ₂₉ shown in FIG. After that, step SP
Return to step SPf2 via f4 to SPf8, and then step SPf2 to
Cycle the processing of SPf8. Then, when the stroke becomes 0, the determination of the stroke SPf6 becomes “YES”, and the processing ends (see time t _{31 in} FIG. 17).

According to the above process, since the half points are included in the ascending and descending strokes of the pedal, the operator can know the quality of the half pedal by listening to the sound of the trill performance.

D; Reproduction Process Next, the process of reproducing the performance data recorded on the floppy disk FD will be described with reference to FIGS. 18 and 19. Note that FIG. 18 is a flowchart showing the reproducing operation, and FIG. 19 is a control block diagram in the reproducing operation.

First, the floppy disk FD by FDD12 is shown in FIG.
Read the normalized position data Xi from the I / O unit 6
It is supplied to the CPU 9 via b (step SPc1). And
In order to calculate the feedback correction amount, the sensors 35a, 3a
Input the detected value of 5b (step SPc2). Then CPU9
Is the 128-position normalized position data Xi
Convert to the normalized position data Xi of the stage (step SPc3),
Moving to step SPc4, using the denormalization conversion table set in the RAM11, the normalized position data Xi is converted to position data xi that matches the peculiar characteristics of the piano, and further,
Position data in RAM11-Position data xi using PWM conversion table
Is converted to control code PWMs (step SPc5). Then, the process proceeds to step SPc6 and the hood back correction amount Δ, Δ
To calculate.

Next, the process of step SPc7 is performed. That is, CPU9
Generates velocity data i by differentiating the position data xi once and multiplies the velocity data i by a coefficient K1 to obtain a control code.
Create PWM1. Further, the position data xi is differentiated twice to create acceleration data i, and a coefficient is added to the acceleration data i.
Multiply K2 to create control code PWM2. In addition, the 19th
As shown in the figure, the deviation Δ between the position signal x input via the A / D converter in the I / O unit 6b and the position signal xi output from the denormalization conversion table is detected at the deviation detection point. The deviation Δ is multiplied by the coefficient K3 to create the control code PWM3. Also, by multiplying Δ by a coefficient K4, the control code PWM4
Is created, and Δ is multiplied by a coefficient K5 to create a control code PWM5. Then, the respective control codes described above are added to obtain a control code PWM.

Next, the control code PW calculated in step SPc7
M is supplied to the PWM control unit as shown in FIG. The PWM control unit is a circuit provided in the I / O unit 6b,
From here, the drive current corresponding to the control code PWM is applied to the solenoid 20a or 20b (step SPc8). After the processing of step SPc8, the process returns to the main routine.

By the above processing, the pedal position control is performed in which the static characteristic correction and the dynamic characteristic correction of the pedal system, that is, the pedal speed, the pedal acceleration, and the feedback signal are all taken into consideration.

The pedal driving in the half point adjustment processing and the half point confirmation processing described above is also performed by the control block shown in FIG.

E: Recording Operation Next, the recording operation of performance data will be described. Here, the processing in the recording operation is shown in FIG. 20, and the control block diagram in this case is shown in FIG.

In step SPb1 shown in FIG. 20, position data xi
Input processing is performed. In this processing, the position signal output from the sensor 35a or 35b is input / output according to the performance of the performer.
This is a process in which the position data xi is converted by the A / D converter in the unit 6b and the position data xi is sequentially captured. Next, in step SPb2, the area is corrected in accordance with the normalized conversion table created in the RAM 11, and the position data is converted into 16-step normalized position data Xi. Next, in step SPb3, the normalized position data Xi is supplied to the FDD 12 for recording. The FDD12 writes the supplied normalized position data Xi to the floppy disk FD.

The above is the recording operation, and by using the normalization conversion table, the normalized position data in which the characteristic peculiar to the piano is corrected is created. In this case, if the half point adjustment processing has been performed, the corrected conversion table used in step SPb2 is used. Therefore, the conversion is done by the correct half-point identified by the ear.

(3: Effect of Example) Even if the half-point is unclear, the provisional half-point x _HP 'by step SP21 shown in FIG.
Is set, and is further corrected to an appropriate half point by the half point adjustment processing. For this reason,
It is possible to set half points for pianos of any characteristics. Further, since the temporary setting of the half point is made according to the half point of the normalized data, there is no great deviation from the correct half point, and there is an advantage that the correction process becomes easy.

In the half point unclear Oh piano (piano can not automated measurement of half point), automatically sets the position data x _'HP corresponding to the [46] of the normalized data to the half point, while the half-point automatic Even the measured piano can be set to a more accurate half point by the half point adjustment process.

Since the trill performance mode and the reproduction test mode are provided, the quality of the set half point can be more accurately confirmed by hearing. In particular, the playback test mode can continuously check the sounding state at the half point, and the trill performance mode can hear the effect of actually performing the trill at the half point and its vicinity. .

"Effect of the invention" As described above, according to the invention of claim 1,
Even if the half pedal area is unclear, the correlation between the pedal position created by actual measurement and the drive signal can be corrected arbitrarily, so that the half pedal area can be set appropriately and individual pedal systems can be set. The effect that the position of the pedal can be accurately controlled in accordance with the characteristics of the half pedal region of 1 is obtained.

According to the second aspect of the invention, since the pedal can be automatically returned gradually from the depressed state, and the key can be automatically driven during that time, the half pedal effect can be confirmed by the sense of hearing. Therefore, the correlation between the pedal position and the drive signal can be appropriately corrected.

According to the third aspect of the present invention, a normalization table for normalizing the pedal position detected by the pedal position detection sensor is created by actual measurement, and the data of at least the half pedal area of the normalization table is created. Can be arbitrarily corrected, and the pedal position can be output as normalized data irrespective of the characteristics of each piano by referring to the normalization table. Therefore, even if the piano has a characteristic that the half pedal area is unclear, the half pedal area can be set appropriately, and the same half pedal effect can be obtained even if the data in one piano is used in another piano. Normalized data that can be output can be output.

According to the invention described in claim 4, the normalized data output from the normalizing means is sequentially stored by the storage means, and the stored normalized data is sequentially read by the reading means and read out. The normalized data is inversely converted by the inverse conversion means into data indicating the pedal position according to the correction content of the data by the correction means. And
Since the drive signal is supplied to the pedal drive solenoid based on the inversely converted data, it is possible to obtain a highly reproducible half pedal effect that is not affected by the individual characteristics of the half pedal region in the automatic piano.

[Brief description of drawings]

FIG. 1 is a block diagram showing the basic structure of the present invention, FIG. 2 is a block diagram showing the structure of an embodiment of the present invention, and FIG. 3 is a side sectional view showing the outer appearance of the same embodiment. FIG. 5 is a rear view showing a positional relationship between the pedal and a solenoid for driving the pedal, and FIG. 5 is a loud pedal 21a in the same embodiment.
6 is a side view showing the schematic structure of the drive system and damper mechanism of the loud pedal 21a in the same embodiment, and FIG. 7 is a characteristic view of the drive system of the shift pedal in the same embodiment. FIG. 8 is a characteristic diagram showing the relationship between the actual position data xi of the loud pedal 21a and the normalized position data Xi in the same embodiment, and FIG. 9 is an actual position data of the shift pedal 21b in the same embodiment. FIG. 10 is an explanatory view showing the relationship between xi and the normalized position data Xi, FIG. 10 is a flow chart showing the operation of measurement processing and the like in the same embodiment
FIG. 11 is a conceptual diagram showing a mode of conversion and inverse conversion in the same embodiment, FIGS. 12 and 13 are a flow chart and a timing chart showing a half point adjusting process in each of the same embodiments, FIGS. 14 and 15 are 16 and 17 are flowcharts and timing charts showing a trill performance mode in the same embodiment, and FIG. 18 is a flowchart showing a reproduction process in the same embodiment. FIG. 19 is a control block diagram in the reproducing process of the embodiment, FIG. 20 is a flowchart showing an operation of the recording process in the embodiment, and FIG. 21 is a control block diagram in the recording operation in the embodiment. . 6 ... Controller (pedal characteristic detection means, correction means,
Target position data generation means, drive signal supply means, control means, normalization table, normalization means, read-out means, inverse conversion means), 12 ... FDD (storage means), 21a ... loud pedal (pedal), 20a ... Solenoid (pedal drive solenoid), 35a ... Sensor (pedal position detection sensor 35a).

Claims (4)

[Claims] 1. A pedal drive solenoid for driving a pedal, a pedal position detection sensor for detecting the position of the pedal, a value of a drive signal for driving the pedal drive solenoid, which is sequentially changed to the pedal drive solenoid. By detecting the position of the pedal corresponding to each value of the drive signal supplied with the pedal position detection sensor,
Of the pedal characteristic detection means for determining the correlation between the position of the pedal and the drive signal value including at least the half pedal area, and the correlation between the position of the pedal and the drive signal value determined by the pedal characteristic detection means A correction means for arbitrarily correcting the correlation of at least the half pedal area, a target position data generating means for generating target position data indicating a pedal target position, and a target position data generated by the target position data generating means. And a drive signal supply means for generating a drive signal according to the correlation between the pedal position corrected by the correction means and the drive signal value, and supplying the drive signal to the pedal drive solenoid. Characteristic automatic piano. 2. The automatic piano according to claim 1, further comprising: a drive signal for pressing down the position of the pedal to a predetermined position between the maximum depressed position and the half pedal region, and then gradually decreasing the pressing amount. The automatic piano is characterized by comprising a control means for outputting and outputting a predetermined keystroke signal. 3. A pedal drive solenoid for driving a pedal, a pedal position detection sensor for detecting a position of the pedal, a value of a drive signal transmitted to the pedal drive solenoid is changed sequentially, and each value of the drive signal is changed. A normalization table for normalizing the pedal position detected by the pedal position detection sensor, which is created by detecting the corresponding position of the pedal by the pedal position detection sensor, and at least half of the normalization table. A correction unit that arbitrarily corrects data in the pedal area, and a normalization unit that refers to the normalization table based on the pedal position detected by the pedal position detection sensor and outputs normalized data that normalizes the pedal position. An automatic piano characterized by comprising: 4. The automatic piano according to claim 3, further comprising storage means for sequentially storing the normalized data output from the normalization means, and reading means for sequentially reading the normalized data stored in the storage means. , Inverse conversion means for converting the normalized data read by the reading means into data indicating the pedal position according to the correction content of the data by the correction means, and generated based on the data converted by the inverse conversion means A driving signal supplying means for supplying the generated driving signal to the pedal driving solenoid.