JPH0138645Y2 - - Google Patents

Description

[Detailed explanation of the idea]

The present invention relates to a character selection device in a serial printer. A serial printer is a sequential printing device that prints desired characters, symbols, etc. on various types of slips, postage, etc. For example, it has achieved high enough print quality and copying ability to print on unified slips, which are now rapidly becoming popular, and perform OCR processing. In this case, printing speed is also an important factor that a serial printer should have. Traditionally, serial printers were type bar type and achieved printing speeds of around 20 characters/second. However, with the introduction of online processing using computers, the above printing speed was no longer sufficient, and in recent years, servo-type serial printers have been proposed and printing speeds of around 40 characters/second have been secured. However, the above 40
Although the printing speed (characters per second) is twice that of conventional printing speeds, it has not yet reached the maximum printing speed that can be theoretically secured in a servo-type serial printer. Now, in order to realize high-speed printing, it is necessary to comprehensively realize all kinds of improvements in the mechanisms and operations within the serial printer. The present invention seeks to improve the type selection device in the print head. That is, an object of the present invention is to provide a type selection device for a print head that can ensure a printing speed sufficiently higher than that which can be achieved with existing serial printers. In accordance with the above object, the present invention provides the following advantages: when selecting a desired type by rotating the print head by a certain rotation angle;
Accelerate with a certain positive torque +T until the intermediate time t/2 of the selection operation time t, and then accelerate with a certain negative torque -T
We focused on the fact that so-called bang-bang control, in which the speed is reduced by This has been changed to a smooth waveform, thereby reducing the linear movement along the step-like waveform that tends to occur especially in the latter half of the type selection process, thereby substantially shortening the type selection time. In addition, in the servo control according to the present invention, the gain K of the switching function defined by θ+Kθ=0 is changed from the rotation amount θ of the print head and its speed θ according to the magnitude of the designated amount of rotational displacement. The effect of shortening the selection time due to the smoothing of the rotational displacement designation signal can be obtained in the same way for any rotational displacement designation amount, thereby making it possible to further shorten the type selection time. That is. FIG. 1 is a schematic structural diagram showing a part of a serial printer to which the present invention is applied. In this figure, reference numeral 11 denotes a printing medium on which desired characters are printed across several lines in a horizontal row, such as roll paper or paper. The print medium 11 is intermittently fed while being supported by a mechanism (platen) 12 that supports the print medium. The printing unit 13 ink-pacts the desired type onto the print medium 11, and rotates the print head 13-1, which has a crown-shaped arrangement of 64 characters and a total of 128 characters in two upper and lower rows. a type selection motor to be driven (not shown);
It consists of a drive mechanism section 13-2 that houses a hammering motor (not shown) and the like for impacting one printed character toward the medium 11. 13-3 is a ribbon cartridge containing red and black ink ribbons. The printing unit 13 rides on the moving shaft 14 and moves in the direction of arrow A in the figure to perform a spacing operation. The moving shaft 14 is rotated by a space motor 15, and a helical groove is formed on the moving shaft 14, so that the printing unit 13 is moved by the feeding operation of the helical groove accompanying the rotation. Perform spacing throughout. Then, each time the printing of one horizontal line is completed, it returns to the direction of arrow A' in the figure. or,
A control circuit (not shown) cooperates with the type selection motor to constitute a first electric servo device, which functions as a type selection means. The space motor 15 is provided with a potentiometer 17 on its back surface that is interlocked with the rotating shaft of the space motor. These space motors 15 and potentiometers 17
constitutes a second electric servo device in cooperation with the control circuit. The present invention will be described with respect to the first electric servo device among the first and second electric servo devices. FIG. 2 is an enlarged perspective view showing only the portion of the configuration of FIG. 1 according to the present invention, which is driven by the first electric servo device. Second
In the figure, 21 is a DC motor, which rotates the printing head 13-1 in the direction of arrow R or R'.
Desired type 22 is selected and moved to the front of platen 12. In this case, a position detector 23 made of a photocoupler or the like is attached to the rotating shaft of the DC motor 21, and a position detection signal of each printed character 22 is output. When the desired type is selected, the print head 13
-1 by hammer means 24, 24' with arrow S
direction, the desired type 22 is impacted onto the print media on the platen 12. A drive line 25 in FIG. 2 is connected to a control circuit board to drive and control the DC motor 21. On the other hand, the control line 26 is connected to the control circuit board and the position detector 2
3 is supplied to the control circuit board to control the DC motor 21. Here, the first electric servo device described above is formed. FIG. 3 is a block diagram showing a schematic circuit configuration of the first electric servo device based on the present invention. In this figure, 21 is the already mentioned DC motor,
The printing head 13-1 is rotated by a predetermined amount to guide desired printed characters to the front of the platen. Further, below the DC motor 21 is the position detector 23 mentioned above, the output of which becomes an input to a speed detection circuit, etc., which will be described later. First, it is assumed that desired printed characters are specified in an arithmetic circuit section (not shown). The arithmetic circuit section looks at the current rotational position of the print head 13-1 and determines how many steps the desired printed character should be rotated. This is the displacement designation signal in the figure. Assuming that 64 characters are arranged in print head 13-1,
The print head 13-1 can be rotated in either forward or reverse directions, and a maximum of 32 steps of displacement can be specified, so in this case, the displacement designation signal is composed of a 6-bit digital signal. The desired type should be guided to the platen in the shortest possible time, and the arithmetic circuit section
It also commands the rotation direction of 3-1. This is the direction control signal in the figure. To explain the general operation after the displacement designation signal and the direction control signal are given from the arithmetic circuit section, the displacement designation signal is sent to the speed designation circuit 31.
The signal is converted into a stepped analog signal through a smoother circuit 32, which will be described later, and then led to a gain controlled differential amplifier 33, which will be described later. The differential amplifier 33 inputs the analogized displacement designation signal, and also receives the output from the speed detection circuit 34, and calculates the difference between these inputs. The position detector 23 described above outputs a sine wave whose amplitude peak value appears at each character position as the print head 13-1 rotates, and the frequency of this sine wave, that is, the rotation speed of the print head is The speed detection circuit 34 converts it into a voltage. The output of the speed detection circuit 34 is a signal that increases as the print head 13-1 rotates, while the analog displacement designation signal decreases as the print head 13-1 rotates. Therefore, the output of the amplifier 33 representing the difference between these signals is initially positive and becomes negative after they become equal and become zero. Due to this positive → zero → negative sequence,
The DC motor 21 is controlled and coarsely controlled to a predetermined rotational position. In this case, the direction control circuit 35 specifies the rotation direction of the DC motor 21 . Further, the output of the amplifier 33 representing the difference in the signals, which has passed through the direction control circuit 35, passes through the contact a of the switch SW during the course control, and is applied to the DC motor 21 through the clamp circuit 36. The clamp circuit 36 is
The voltage level applied to the DC motor 21 is suppressed to a predetermined value. During the course control period, when the arithmetic circuit unit determines that the displacement has occurred by the specified displacement given by the displacement designation signal, the lock signal shown in the figure is supplied and the contact of the switch SW is switched to the b side. Guided by the sine wave position detection signal itself, the DC motor 21 very minutely reaches the target position. The timing at which the lock signal is sent is usually 1/4 cycle before the sine wave reaches the target position. This behavior is
This is called fine control as opposed to course control. The above-mentioned control consists of a dual mode of so-called course control and fine control, ensuring highly accurate type selection. Furthermore, the above-mentioned type selection is carried out using a method that is quite similar to the bang-bang control method, and this is also said to be a convenient method for controlling a certain displacement at an optimal time. As shown in the graph of FIG. 4A, this bang-bang control is such that when a certain type selection time is t, at the first time t/2, it becomes +T.
A certain torque is applied to accelerate the vehicle, and then a -T torque is applied to decelerate it to the target position.
This corresponds to the above-described DC motor 21 being controlled in the sequence of positive → zero → negative. By applying torques +T and -T as shown in FIG. 4A, the rotational speed V of the DC motor 21 changes along a substantially triangular route as shown in the graph of FIG. Choice is secured. However, this method alone often operates in a nonlinear region and is vulnerable to external load fluctuations. A conceivable control method to solve this problem is to perform non-linear operation as much as possible in the course control section, and to perform linear operation as the fine control section approaches. When this is viewed from the phase plane, it becomes as shown in FIG. 5A. That is, the operation at the time point P when the torque of +T is switched to the torque of -T and at the point Q near the stop is performed in a linear manner. However, the horizontal axis in FIG. 5A represents the rotational displacement θ, and the vertical axis represents the speed θ〓. Further, the straight line Z is a switching function and is defined by θ+Kθ=0. By applying the above-described control method to the character selection operation, it is possible to achieve higher speed. For this purpose, the following is necessary. 1. Change the current step-like speed setting (output of the speed designation circuit 31 mentioned above) to a smooth (not step-like) speed profile as shown in FIG. 5B. 2. K in the above equation θ+Kθ = 0 is made variable depending on the number of type selection steps (corresponding to θ above). By satisfying the above two items, speeding up becomes possible. Specifically, the smoother circuit 32 in FIG. 2 operates to satisfy the above item 1), and the gain controlled differential amplifier 33 operates to satisfy the above item 2). Note that in the conventional circuit, there is no smoother circuit at all, and no gain-controlled differential amplifier is also found. Conventionally, the smoother circuit 32 shown in FIG. 3 was not provided, and the speed profile based on the displacement designation signal changed stepwise with time. Therefore, the DC motor will respond to this staircase in a certain region. When the DC motor is in course control, it cannot respond to the stairs because of its non-linear operation. However, as the target position approaches, the speed detection circuit output approaches the displacement designation signal value and is drawn into linear operation. As a result, during linear operation, the output voltage of the differential amplifier 33 becomes stepwise. If the output voltage of the differential amplifier 33 becomes step-like, the motor current also becomes step-like, and the braking torque also becomes step-like. As a result, the braking is insufficient, the rotation of the motor becomes vibratory, and the positioning time becomes long. Therefore, if the smoother circuit 32 is introduced to smooth the stepped waveform, the stepped displacement signal will be converted into a smooth waveform as shown in FIG. 5B.
It is possible to approximate the speed change by the ideal bang-bang control shown in FIG. 4B, and it is possible to shorten the type selection time. Details of the smoother circuit 32 will be described later. Conventionally, the gain K at the switching function θ+Kθ=0 (straight line Z) shown in FIG. It was designed to be optimal for. Therefore, as the number of steps during character selection becomes smaller, the control deviates from optimal control. For this reason, the type selection time becomes as shown by the curve shown by the solid line in FIG. 6, and when the number of steps is small, it takes an unnecessarily long time. In addition, in Fig. 6, the vertical axis indicates the commanded rotational displacement amount (number of steps).
θ is plotted, and the horizontal axis plots the character selection time t. Therefore, in the present invention, the characteristic curve is modified as indicated by the dotted line in accordance with the number of type selection steps, that is, the amount of commanded rotational displacement, in order to shorten the type selection time by Δt in the figure. That is,
In the present invention, K is made variable and the switching function is determined so that optimal control can be performed for each character selection step. The gain controlled differential amplifier 33 shown in FIG. 3 performs this. By making K variable according to the number of steps (.theta.), the speed of character selection can be increased by the time .DELTA.t shown in FIG. Let's analyze this in a little more detail. In order to determine the time for character selection, since a bang-bang operation is performed, it is sufficient to approximately determine the time t in the area shown in FIG. 5B. Here, if K is a/b, the switching function can be rewritten as aθ+bθ〓 (1). Here, the function of θ in the region B of Figure 5 is given by θ=B(θ)−1/2At ² (2) θ〓=−At (3) since the function is a square curve. Therefore, by substituting the above equations (2) and (3) into the above equation (1), a(B(θ)−1/2At ² )−bAt=0 (4) is obtained. Solving this quadratic function of t for t, we get

[Formula], and since t is positive, is obtained. In this case, conventionally, a constant B(θ)=θ was given for B(θ), and no correction was made in accordance with the number of steps. Therefore, in the present invention, B(θ) is set to B(θ)=K ₂ (θ)/θ+K ₁ and includes correction according to the number of steps. In other words, the above equation (5) is given as. In the above description, A is treated as a constant, but if A is also determined as a function proportional to (θ), the time for character selection can be further reduced. Therefore, by changing the above equation (3) to θ〓=A(θ)t=(−K ₃ θ+K ₄ )t (7) and substituting it into the above equation (6), we get becomes. It can be seen that t in the above equation (8) is shorter than t in the above equation (6). That is, if the denominator of the second term in the above equation (6) can be made into a higher-order function, t will be shorter.
However, it is inevitable that the circuit configuration will become somewhat complicated. Based on the above idea, the gain controlled differential amplifier 33 of the present invention shown in FIG.
It is configured as shown in the figure. In this figure, the input terminal 71 inputs the velocity component (θ〓) from the speed detection circuit 34 (Fig. 3), and for θ〓, the coefficient (K ₄ - K ₃ θ). On the other hand, the input terminal 72 inputs the position component (θ) from the speed specifying circuit 31 via the smoother circuit 32, and further multiplies θ by the coefficient K ₂ /θ + K ₁ included in the above equation (6). . The difference between these input signals is taken by a subtraction circuit 73, resulting in an error signal ε. The error signal ε is multiplied by a necessary coefficient (for example, 1) to obtain the drive signal F(θ) to be applied to the DC motor. F(θ) is F(θ)=B(θ)θ+A(θ)
θ〓. The specific configuration of FIG. 7 is as shown in FIG. 8, and the resistors R ₁ and R ₂ are variable according to θ.
In other words, A(θ)=R ₃ /R ₂ (θ), B(θ)=R ₃ /R ₁ (θ
). The resistors R ₁ and R ₂ are each formed as an analog switch that receives θ as an input, and a plurality of resistor groups connected to the analog switch. FIG. 9 shows one example. Since the gain controlled differential amplifier 33 shown in FIG. 3 has been described above, the smoother circuit 32 shown in FIG. 3 will now be described. FIG. 10 is a block diagram showing an embodiment of the smoother circuit 32 based on the present invention, and FIG. 11 is a waveform diagram for explaining its operation. The smoother circuit 32 in FIG. 10 changes the stepped displacement designation signal outputted from the speed designation circuit 31 in FIG. 3 into a smooth profile as shown in FIG. Avoid following the specified signal. The smoother circuit in FIG. 10 converts the actual speed signal (output of the speed detection circuit 34 in FIG. 3) into a displacement signal by integrating it, and converts this integrated displacement signal into a step-like This is fed back to the displacement designation signal to smooth the displacement signal. In other words, the actual velocity signal is integrated until a position pulse (pulse emitted for each step (see Figure 11 C)) is obtained, and the amount of change in displacement θ _ij (see Figure 11 B) in this section is calculated. demand. Since this amount of change θ _ij is the amount of displacement in the section, it is calculated from the output of the speed designation circuit 31 (see waveform R at A in FIG. 11) to the integral value S _ij (first
(see Figure 1D), the step-like waveform can be converted into a square curve proportional to the displacement. This is the smoothed displacement designation signal S (dotted line) shown in FIG. 11A. Here, since the amount of displacement is obtained by integrating until the position pulse is output, it is necessary to reset at the timing when the position pulse is output. This reset signal is the 10th
Shown as Reset in the figure. In Figure 10,
Resistor R _i and capacitor C _f determine the time constant of the integration described above. These R _i , C _f and operational amplifier OP
The actual speed signal input from the input terminal 101 (output of the speed detection circuit 34 (FIG. 3)) is integrated, and the step-like signal input from the input terminal 102 is calculated using the ratio of resistors R ₁ and R ₂ . Apply feedback to the speed specification circuit output. Here, the smoothed displacement designation signal S obtained from the output terminal 103 (Fig. 11A
(see S). Note that at the beginning of the operation, an accurate speed cannot be obtained and regular integral operation cannot be performed, so an initial value (Int.Val in FIG. 10) is input to the input terminal 104. Input terminals 105, 10
6, 107 is a terminal for inputting a timing control signal, and controls opening/closing of analog switches AS ₁ , AS ₂ , AS ₃ and AS ₄ . R _f and R _s are resistors, OP' is an operational amplifier, and INV is an inverter. The operating procedure is to first set AS ₁ on, AS ₂ off, AS ₃ off, AS ₄ off in the initial value setting sequence for integral operation, and then during actual integral calculation, AS ₁ off, AS ₂ off, AS ₃ on, AS ₄ on, etc. Integration S _ij in Figure 11D is performed to obtain the desired output S. When the integration at each step is completed, a reset sequence is entered: AS ₁ off, AS ₂ on, AS ₃ on, AS ₄ off. Then we start the next integral calculation. in this case,
The initial value setting sequence described above is unnecessary. As explained above, according to the present invention, since the smoother circuit 32 is introduced, the linear operation along the step-like waveform that conventionally tends to occur especially in the latter half of the type selection process is reduced, and the type is substantially Selection time can be shortened. Furthermore, by introducing the gain controlled differential amplifier 33,
In general, the amount of rotation θ of the print head and its speed θ = θ
The gain K of the switching function defined by +Kθ = 0 can be changed according to the magnitude of the specified amount of rotational displacement, and the effect of shortening the selection time due to smoothing by the smoother circuit 32 can be applied to any rotational displacement specified. The same can be said for quantity. This makes it possible to further reduce type selection time and ensure a sufficiently high printing speed compared to conventional serial printers.

[Brief explanation of drawings]

FIG. 1 is a schematic structural diagram showing a part of a serial printer to which the present invention is applied; FIG. 2 is an enlarged perspective view showing only the portion of the structure of FIG. 1 according to the present invention; FIG. 3 is a block diagram showing a schematic circuit configuration of an electric servo device to which the present invention is applied, FIGS. 4A and B are graphs for explaining a general bang-bang operation, and FIGS. 5A and B are Graphs for explaining the basic operation of the present invention; Figure 6 is a curve showing the relationship between the number of character selection steps (θ) and the required time;
Gain controlled differential amplifier 33 shown in the figure
FIG. 8 is a block diagram showing the basic configuration of FIG. 7, FIG. 9 is a diagram showing an embodiment of the circuit shown in FIG. 8, and FIG.
FIG. 0 is a block diagram showing one embodiment of the smoother circuit based on the present invention, and FIGS. 11A, B, C, and D are waveform diagrams for explaining the operation of the circuit of FIG. 10. In the figure, 13-1 is a print head, 21 is a DC motor, 23 is a position detector, 31 is a speed designation circuit, 32 is a smoother circuit, 33 is a gain controlled differential amplifier, and 34 is a speed detection circuit.

Claims (1)

[Scope of claim for utility model registration] A motor that rotates the print head to select type, and the deviation between the commanded movement amount and the actual movement amount,
means for generating a step-like position deviation signal; means for converting the step-like position deviation signal into a smooth signal that follows an envelope of the step-like waveform; and generating a signal corresponding to the actual speed of the motor. and an amplification means for amplifying a difference signal between the position deviation signal and the actual speed, and the motor is always driven by the output signal from the amplification means, and the type selection is The van is driven with positive torque (+T) until half the time (t/2) of the time (t), and thereafter is driven with negative torque (-T).
A type selection device for a serial printer that selects a predetermined type under bang control, comprising means for changing the gain of the amplifying means in accordance with the magnitude of the commanded movement amount. Selection device.