JPH0789132A - Printer - Google Patents

Abstract

(57) [Abstract] [Objective] A printer with a maximum resolution of 1200 DPI and variable resolution is realized by using four sets of LED heads with a resolution of 300 DPI. [Structure] Using LED heads A to D having a resolution of 300 DPI, the heads A and B are arranged with a shift of 1/2 of the arrangement pitch of the light emitters and arranged, images are combined by a half mirror, and an image is formed by a monocular lens array. Similarly, the heads C and D are also arranged by shifting by 1/2 the arrangement pitch of the light emitters, combined by a half mirror, and formed by a monocular lens array. The light from the heads A and B and the heads C and D are combined by a half mirror, and the image is reduced by a half by a monocular lens array, and the resolution is quadrupled in total. When reducing the resolution to 1 / N,
The image data in the page memory 30 is read N times by the address generator 34 and expanded into N pieces of data.
The gate 38 and the data distributor 40 distribute to four heads.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a variable resolution printer, and more particularly to a variable resolution control of a printer using an LED head, an EL head, a plasma head, a thermal head or the like.

[0002]

[Terminology] In this specification, a printer includes a digital printer, a facsimile, and the like in addition to a printer in a narrow sense, and a printer is defined as a device that prints on paper.

[0003]

2. Description of the Related Art Printer improvements have been made primarily as resolution improvements. For example, in the case of a printer using an LED head, an attempt is made to improve the resolution from 300 DPI (dots / inch) to 600 DPI, and then the resolution 120.
Improvements to 0DPI are under consideration. When the resolution is improved in this way, it becomes necessary to make the resolution variable as the next problem. For example, if the resolution is 1200
Assuming you get a DPI LED head, this is always 1
It is not necessary to drive at 200 DPI. If the resolution of the image data is only 200 DPI, the head is also 200 DPI
It may be preferable to drive with. In order to rewrite 200 DPI data to 1200 DPI, it is not always preferable to interpolate the data and reduce the printing speed.

In this regard, Japanese Patent Laid-Open No. 5-138,99
No. 3 proposes to print image data of, for example, 600 DPI with a 400 DPI head using a look-up table. In this proposal, image data of 600 DPI is cut out every 3 dots, and this is converted into print data of 2 lines at 400 DPI. The data of (0,0,0) or (1,1,1) is (0,0) + (0,0) or (1,1) +
You can change it to (1,1), and the problematic (1,0,0) is
(1,0) + (0,0) and (1,1,0) is (1,0) +
Convert to (1,1). As is clear from this, there is no concept of image data enlargement in Japanese Patent Laid-Open No. 5-138,993. Further, in this method, print data is cut out by a predetermined number of dots, processed by a reference table, and the data of one line is converted into the data of two lines. This requires a lot of memory, the calculation is complicated,
The control algorithm also becomes complicated.

Increasing resolution is not a solved problem. In the case of a printer using a solid type head such as an LED head, an EL head, a plasma head, or a thermal head, it is necessary to reduce the arrangement pitch of the image elements in order to improve the resolution. This includes miniaturization of image elements, increase of the total number of image elements, reduction of line width of wiring to the image elements,
The size of the bonding pad for connection with the drive circuit is reduced, the total number of bonding pads and the number of connection points are increased, and the cost of the head is increased and the yield is sharply reduced. For example, if the yield at 300 DPI is 90%, the yield at 600 DPI will decrease to 81% as the total number of image elements doubles, and the yield decrease due to miniaturization will be added. Therefore, the overall yield is further reduced. Further, it is difficult to realize a resolution higher than 600 DPI because it approaches the limit of fine processing.
Summarizing these things, it can be said that the improvement in resolution due to the miniaturization of the image element has reached the limit in the vicinity of 600 DPI.

[0006]

SUMMARY OF THE INVENTION An object of the present invention is to make the resolution of a printer variable in accordance with image data, and in particular, the original data of one line does not extend over two lines. , It is to be able to change the resolution with simple processing. The subject of the invention of claim 2 is
Another object of the present invention is to make it possible to improve the resolution of an optical head by a method other than downsizing the light emitting body and to easily change the resolution.

[0007]

According to a first aspect of the present invention, there is provided a printer which supplies image data to an image element and prints line by line at a predetermined resolution, the image data being the resolution of the printer and the resolution of the image data. It is characterized in that a stretching means is provided according to the ratio.

According to a second aspect of the present invention, in a printer in which an optical image from a light emitting body is formed by using a lens array, there are two rows in which the light emitting bodies are arranged with a shift of 1/2 of the arrangement pitch. One set is provided, and for each set of two rows of light emitters, a half mirror for combining the light of the light emitters and a monocular lens array for reducing the combined light to form an image are provided, and Providing n sets of the two rows of light emitters, (n
Is a natural number of 2 or more), a new half mirror is provided to combine the light from the n sets of light emitters, and 1 dot of image data is converted into N / m dot image data (N is a natural number of 2 or more). , M is 1 or 2 and is a divisor of N), a means for distributing the image data stretched to N / m dots to the n sets of light emitters, and the value of N. Is provided to change the resolution of the printer to 1 / N of the maximum resolution.

[0009]

According to the first aspect of the invention, the image data of one dot is expanded into a plurality of dots. For example, resolution 1200
When printing 200 DPI data with a DPI digital printer, for example, one dot may be expanded to 6 dots. Similarly, in the case of 300 DPI data, in principle, it may be expanded to 4 dots, and in 600 DPI, it may be expanded to 2 dots. As a result, the resolution of the printer can be changed very easily.

The invention of claim 2 relates to a printer using an LED head, wherein 2n rows of light emitters are used to increase the resolution of the printer by 2n times. First, in order to double the resolution, two rows in which the light emitting elements are arranged with a shift of 1/2 of the array pitch are used, and one row is formed. In the case of two rows of light emitters, the light emitters of one row are inserted between the light emitters of the other row, and the resolution is doubled. Then, since the two rows of light emitters are processed as if they are arranged in one row, the light of the two rows of light emitters is combined using a half mirror. A monocular lens array is provided on the opposite side of the light emitter as viewed from the half mirror, and the light combined by the half mirror is reduced to form an image. Next, n sets of light emitters each including two rows of light emitters are provided (n is a natural number of 2 or more), and light from the n sets of light emitters is combined by a half mirror to form an image. If this is done, the resolution will be n
In the set of two rows of light emitters, the light emitters are arranged with a half pitch shift to increase the resolution twice.
The resolution can be increased 2n times in total. With the configuration of the present invention, the resolution can be increased without reducing the arrangement pitch of the light emitters. For example, resolution 300DP
If four rows of I light emitters are used, the resolution is 1200 DPI.
Printer is obtained, and the resolution is 1800DP if 6 sets are used.
I printer is obtained. If four rows of light emitters with a resolution of 400 DPI are used, the total resolution is 1600 DP.
I, the total resolution reaches 2400 DPI by using four rows of luminous bodies having a resolution of 600 DPI.

The resolution may be increased to 2n times, and may be increased to 6 times or 8 times in addition to 4 times. However, it is practically important to increase the resolution to 4 times. For example, even if an image of a black-and-white binary image is increased in resolution to 2000 DPI, it reaches almost the same image quality as that of a natural image of black and white 256 gradations, and it is almost impossible to discriminate between a binary image and a natural image of 256 gradations with the naked eye. It is possible. Therefore, the resolution is 2000
It is not necessary to make it higher than the DPI, and if an image with a resolution of 1200 DPI is obtained starting from a row of light emitters with a resolution of 300 DPI, an image close to a natural image will be obtained. Also, starting from a row of light emitters with a resolution of 400 DPI, an image quality close to a natural image with a resolution of 1600 DPI can be obtained. Furthermore, if four rows of luminous bodies with a resolution of 600 DPI are used, a resolution of 2400 DPI and a resolution higher than that required for expressing a natural image can be obtained.

When the number of rows of the light emitting elements to be combined is increased from four rows, the number of half mirrors used is increased and the brightness is lowered. The increase in the number of half mirrors means that the physical structure of the printer is complicated and the assembly is difficult. For example, the number of half mirrors required to combine four rows of light emitters is three, and if eight rows of light emitters are used,
Since 3 half mirrors are required for 4 rows of light emitters, 6
One half mirror is needed, and then one half mirror is needed to combine the light of the four rows of light emitters with the light of the remaining four rows of light emitters. As a result, the number of half mirrors required is 7, and 2n-1 half mirrors are generally required.

When four rows of light emitters are used, the light passes through the half mirror twice until the image is formed, so the amount of light is reduced to 1/4. If eight rows of light emitters are used here, the number of half mirrors that pass through until image formation is three, so the amount of light is 1 /
Drop to 8. Since the amount of light decreases as the number of rows of light emitters used increases, four rows of light emitters are used and n is 2
It is preferable that That is, it is preferable to use two sets of light emitters in two rows and to increase the resolution four times in order to avoid complication of the structure of the printer and to prevent a decrease in light amount. Further, in general, if four rows of light emitters are used and the resolution is increased four times, the theoretically required resolution can be approximated and a sufficient image quality can be obtained.

In order to make the resolution variable, image data of 1 dot is expanded to image data of N / m dots (N is a natural number of 2 or more, m is 1 or 2 and a divisor of N). To stretch the image data, for example, as shown in FIG. 1, the address generator is operated to read one image data N / m times. Alternatively, as shown in FIG. 6, the image data read at a certain frequency is converted into N / m
Read at double frequency. The method itself for stretching the image data is arbitrary. Here, m is set to 1 when simultaneously driving two rows of effector arrays arranged with a half pitch shift, and m is set to 2 where two rows of light emitters are shifted by a half pitch. This is the case of driving only one side.

The stretched image data is distributed to a row of 2n sets of light emitters. For example, when m is 1, 2K pieces of image data are distributed to one set of light emitters, and K pieces of image data are supplied to each light emitter row. Then, the number of supplied image data is integrated by a counter or the like, and when 2K pieces of image data are supplied, they are distributed to other sets of light emitters. If m is 2,
Of the set of two rows of light emitters, only one row of light emitters is driven. Then, when K pieces of image data are supplied to a row of one light emitter, the next K pieces are supplied to one row of another set of light emitters.

For example, a printer having a total resolution of 1200 DPI is used by using four rows of light emitters arranged at a resolution of 300 DPI. In addition, each luminous body is divided into blocks for every 48 luminous bodies.
The blocks are time-divisionally driven. In this case, 1
When a 200 DPI printer is driven with m = 1 and N = 6, image data with a resolution of 200 DPI can be stretched 6 times and printed with a 1200 DPI printer. Then, three of the image data stretched 6 times are driven to the light emitters in the first row, and the next three are supplied to the light emitters in the second row. In this manner, 16 pieces of image data are read out and expanded by 6 times to form 48 × 2 96 pieces of image data, and the image data is supplied to the first column and the second column for each block. The next 16 pieces of image data are similarly stretched 6 times and supplied to the light emitters in the third and fourth rows. If this cycle is repeated, the image data of 200 DPI will be 1
This means that a 200 DPI printer can be driven, and the resolution of the printer has been reduced to 1/6. If m is 2,
One piece of image data is stretched three times, and the original 16 pieces of image data are supplied as 48 pieces of image data to the light emitters in the first row. Similarly, the next 16 pieces of image data are expanded to 48 pieces and supplied to the light emitters in the third row. If such a cycle is repeated, 200 DPI image data
It can drive a 200 DPI printer. The resolution reduction ratio N is variable and is determined according to the ratio of the resolution of the image data to be printed and the resolution of the printer. With these, the resolution of the printer can be changed to 1 / N of the maximum resolution, and image data of various resolutions can be printed by one printer.

[0017]

EXAMPLE An example is shown in FIGS. In this embodiment, a printer having a resolution of 1200 DPI is made by using an LED array having a resolution of 300 DPI, and the resolution is variable within the range of 200 DPI to 1200 DPI. An EL array, a plasma array, a thermal array or the like may be used instead of the LED array. FIG. 1 shows a driving circuit for varying the dot density and the resolution. FIG. 2 shows the data structure of image data stretched N times by the drive circuit of FIG. FIG. 3 shows the optical system of the printer, and FIG. 4 shows the arrangement of the light emitters in a set of light emitters each consisting of two rows of light emitters. FIG. 5 shows the image formation principle of the printer and the image allocation to each light emitter.

In FIG. 3, reference numeral 2 is an example of the image forming surface of the photosensitive drum. Reference numeral 4 is a housing, and a housing made of plastic or the like is used. Reference numerals 6, 8 and 10 are half mirrors. The half mirrors 6, 8 and 10 are used so that the transmittance and the reflectance are equal and the unevenness of the light amount on the photoconductor drum 2 does not occur. 1
Numerals 2 and 14 are 1/2 reduction monocular lens arrays, which are 1/3 reduction when 6 rows of light emitters are used and 1/4 reduction when 8 rows are used. In addition, monocular lens array 1
The reduction ratios of 2 and 14 also differ depending on whether the LED arrays 24 are arranged in close contact or with a gap. Reference numerals 16, 18, 20, 22 denote substrates, and the LED arrays 24 are arranged on these substrates in a single row and in a straight line with almost no space. In the embodiment, the number of columns of the LED array 24 is 4, the number of half mirrors 6, 8 and 10 is 3, and L
One less than the number of columns of the ED array 24. Further, since the monocular lens arrays 12 and 14 are provided for every two columns of the LED array 24, if the number of columns of the LED array 24 is 2n, n monocular lens arrays are required. The printer of the embodiment is the same as that using four sets of LED heads, and these LED heads are virtually connected to LED head A,
We will call them B, C, and D.

FIG. 4 shows the arrangement of the light emitters in the LED heads A and B. First LED array A-1 in head A
In the head B, the respective light emitters of the first LED array B-1 are arranged so as to be shifted by ½ of the arrangement pitch. Each LED array 24 has a light emitter, for example, 300D.
They are arranged at an arrangement pitch of PI, and 48 light emitters are arranged for each array. Each of the LED heads A, B, C, D has 40 LED arrays 24 mounted therein.
The array pitch of the light emitters in the LED array 24 is 84.7μ.
Therefore, in the LED head A and the LED head B, it is necessary to shift and arrange the light emitting bodies by 42 μm. Similarly, in the LED head C and the LED head D as well, it is necessary to dispose the light emitting bodies by shifting them by 42 μm.
Further, since the lights from the LED heads A, B, C, and D are combined by the last half mirror 6 to form an image on the photosensitive drum 2,
It is necessary to arrange each light emitter with an accuracy of about ± 5 μm. For this reason, the mounting markers for the LED arrays 24 are provided on the substrates 16, 18, 20, 22 and are mounted in alignment with the markers provided on each LED array 24. Also each board 1
6,18,20,22 are assembled to the housing 4 with high precision. By doing so, it becomes possible to mount each light emitting body with an accuracy of about ± 5 μm. In the following, the head number and the LED array 24 number are indicated by symbols such as A-2, and for example, A-2 represents the second LED array in the LED head A.

FIG. 5 shows the principle of image formation in the embodiment. LE
The D head A and the LED head B are arranged so as to be shifted by 1/2 pitch of the light emitting body. Similarly, the LED head C and the LED head D are also arranged with a 1/2 pitch shift. On the other hand, the LED heads A and B and the LED heads C and D are
The LED arrays 24 are arranged so as to be shifted by ½ of the length thereof. Then, the lights from the LED heads A and B are combined by the half mirror 10 and reduced to 1/2 by the monocular lens array 14 to form an image of 1200 DPI. This is reflected by the half mirror 6 to form an image on the photosensitive drum 2. Similarly, the light from the LED head C and the LED head D is reflected by the half mirror 8
Are combined with each other, reduced to ½ by the monocular lens array 12, and imaged through the half mirror 6. The image formation area is as shown by the shaded area on the photosensitive drum 2 in FIG. 5, and the light from the monocular lens array 14 and the light from the monocular lens array 12 are alternately switched for each lens, and as a whole, a break is formed. An image with no resolution of 1200 DPI is obtained. One LED array 24 corresponds to each monocular lens in the monocular lens arrays 12 and 14 (two LED arrays 24 and 24 in total for one monocular lens).

FIG. 1 shows a drive circuit of the printer of the embodiment. The left side of the communication symbol in the center of the figure represents the processing in the printer body, and the right side represents the processing in the LED head side. In the figure, 30 is a page memory of the printer main body, 32 is a frequency dividing circuit, 34 is an address generator, and 36 is a frequency dividing ratio N.
Is a mode selector for determining. Then, this frequency division ratio N is the ratio that reduces the resolution, and in FIG.
The data flow is shown for the case. Reference numeral 38 denotes a gate circuit which alternately outputs the print data read from the page memory 30 dot by dot from two output terminals.
Reference numeral 40 denotes a distributor, which has a built-in counter and counts the clock signal CLK, and is used for 2 blocks (here, L
Each time a clock signal of 96 times, which is twice the number of light emitters of the ED array 24, is counted, the switch is switched and output. 42 is a first-in first-out FIFO memory, in which four memories each capable of accommodating one block (48 dots) of data are arranged in parallel. Then, the data of the first bank of the FIFO memory 42 (the portion indicated by A in the figure) is set to L
The data of the bank C is supplied to the LED head C, the data of the bank B is supplied to the LED head B, and the data of the bank D is supplied to the LED head D.

The operation of the drive circuit shown in FIG. 1 will be described. When driving at the maximum resolution of 1200 DPI, the data transfer speed from the printer main body is insufficient for one-line serial transfer.
Therefore, 4-bit parallel transfer is used, and LED head A,
A data transfer line is provided for each of B, C, and D, and data is transferred in a total of 4 bits in parallel for each 1 bit. The basic clock frequency of the printer is NXMH
If it is z, the frequency dividing circuit 32 divides this into 1 / N, and reads the data from the page memory 30 with the clock CLK of X MHz. This means that the number of data read from the page memory 30 is reduced to 1 / N according to the resolution. Then, accordingly, the number of printing lines is reduced to 1 / N, and the number of data per line becomes constant.

Assuming that the page memory 30 corresponds to a resolution of 1200 DPI, in the case of 200 DPI, the image data is only a part thereof, for example, the image data is only in a range of 1/6 along the main scanning direction. . Therefore, a counter is built in the address generator 34, and the address is shifted by one each time N clocks are counted. As a result, for example, in the case of a resolution of 200 DPI, the same data is read out 6 times, and 1-bit image data is expanded to 6 bits. For example, if the image data at the address (x, y) in the page memory 30 is 1, the address generator 34 outputs 6
Since the data of the same address is read out once, the print data read out to the gate 38 is (1, 1, 1, 1, 1, 1, 1). The gate circuit 38 has two gates, and switches the gate to be turned on every clock. As a result, the image data is sliced for each dot, and the (1, 1, 1, 1, 1, 1,
The data of 1) is divided into two data of (1,1,1) and (1,1,1).

The data distributor 40 has a counter and four gates. The first gate is in the bank A of the FIFO memory 42, the second gate is in the bank C of the FIFO memory 42, and the third gate is in the FIFO memory 42. 42 bank B and the fourth gate to bank D. The operation of the gate is inverted every time the counter counts 96 clocks. As a result, the first 96 pieces of data (96 clocks) are stored in the bank A and the bank B, respectively.
The next 96 pieces of data are input to the bank C and the bank D, 48 pieces each. The frequency divider circuit 32 further divides the clock signal into quarters to obtain a frequency of X / 4 MHz.
To generate a clock signal. With this clock signal, the data in the FIFO memory 42 is read out, and the head A,
Transfer to B, C, D. In addition, each head operates the LED arrays 24 independently by this clock.

FIG. 2 shows the flow of data from the page memory 30 to the distributor 40. Now resolution 2
If the printer is driven at 00 DPI, one data in the page memory 30 is expanded to 6 dots data, and the gate circuit 38 inputs the odd number data out of 6 dots to the bank A of the FIFO memory 42. , The even-numbered data is input to the bank B. When 48 pieces of data are input to the banks A and B, the switch of the distributor 40 is switched, and 48 pieces of data are input to the banks C and D, respectively. Thereafter, this step is repeated to alternately use the banks A and B and the banks C and D. This is the same when driving at another resolution such as 300 DPI. When driving at 300 DPI, 1 dot data in page memory 30 is expanded to 4 dots
The odd-numbered data in the dots is input to the bank A and the even-numbered data is input to the bank B.

When driving at 400 DPI, the data in the page memory 30 is tripled. For example, if the data in the page memory 30 is (1,0), (1,1,1)
After reading the data of, the data of (0,0,0) is read. Since the gate circuit 38 slices this bit by bit, the data of (1, 1, 0) is stored in the bank A in the bank B.
Data of (1, 0, 0) is input to. Since the head A and the head B are arranged so as to be shifted by 1/2 pitch of the light emitting body, when printing is performed with the data of (1,1,0) and (1,0,0), the photosensitive drum is (1,1,1). , 1,0,0,0). When the resolution is 600 DPI, the data in the page memory 30 is doubled. As a result, the data 1 is expanded to (1, 1), and the data 1 is input to the bank A and the bank B, respectively. When the resolution is 1200 DPI, for example, when 2-bit data (1,0) is read from the page memory 30, the first data 1 is input to the bank A and the next data 0 is input to the bank B.

The distribution of the data from the distributor 40 to the FIFO memory 42 is switched every time data for two LED arrays 24 is supplied, and this is changed for each L.
The ED heads A to D are adapted to drive the LED arrays 24 one by one in a time division manner. The data in the FIFO memory 42 is a clock (1/1) that is obtained by further dividing the read frequency of the data in the page memory 30 into 1/4.
4) Read out at * CLK and supply to each LED head A to D via four signal transfer lines, and each LED head A to D is driven at this frequency. As a result, gate 3
The data sent with the frequency CLK to 8 are the four heads A
To D at a frequency of 1/4 CLK, and the number of data inputs and the number of outputs to the FIFO memory 42 match.

In the embodiment, the LED head A and the LED head B are continuously driven simultaneously, and the LED head C and the LED head C
We decided to drive head D constantly at the same time. However, if the resolution is 200 DPI or 300 DPI,
It is not necessary to drive the LED heads A and B at the same time, but it is sufficient to drive only one of them. Similarly, when the resolution is 200 DPI or 300 DPI, only one of the LED heads C and D may be driven. This corresponds to the case where m is set to 2. In this case, the dividing circuit 32 divides the basic clock into N / m. For example, in the case of a resolution of 200 DPI, one dot of image data in the page memory 30 is divided into three. The dot is expanded and input to the gate 38. Next, the gate 38 stops slicing the data for each bit and supplies 48 blocks of image data for one block to, for example, only bank A or only bank B. Similarly, in the case of a resolution of 300 DPI, the basic clock signal is not divided into ¼ but divided into ½ and supplied to only the bank A or only the bank B, for example. The operation in other cases is the same as that of the embodiment shown in FIGS. Although the method using the address generator 34 for expanding the image data of 1 dot to N / m dots has been described, for example, as shown in FIGS. 6 and 7,
The image data read to the latch 50 may be supplied to the LED head N times.

[0029]

[Modification] In FIGS. 6 and 7, the data read from the page memory 30 is latched in the latch 50, and this data is stored in N
A modification in which the data of one dot is read and the data of one dot is expanded to N dots will be described.

Four heads A to D are provided as LED heads, and the number of dots of each head is 2560 each.
A total of 10240 dots is used as an A4-1200 DPI head. The arrangement of the heads A to D is changed from the above-mentioned embodiment,
The first 1/4 portion of one line is printed by the head A, the next 1/4 is printed by the head B, the next 1/4 is printed by the head C, and the last 1/4 is printed by the head D. One line is simply divided into four equal parts and 1/4 is allocated to each head. Each of the LED heads A to D is an array of LED arrays in which a predetermined number of LEDs are arranged.

Considering the data transfer from the printer main body, the data transfer rate is 2 with a 300 DPI head.
Since the data transfer amount is in proportion to the square of the resolution, the data transfer frequency is 32 MHz in the case of 1-bit serial transfer in the head of 1200 DPI. This exceeds the limit of serial transfer. Therefore, data is transferred in parallel in 4 bits and the data is input to the heads A, B, C and D at the same time. As a result, the maximum data transfer frequency is 8M
It becomes Hz. Further, the four heads A to D are driven simultaneously with the data being transferred to the heads A to D at the same time. As a result, in the operation of the head, one line is divided into four parts to the left and right, and the four heads A to D are assigned 1/4 each.
The heads A to D operate simultaneously. Each of the heads A to D may be time-division driven or static drive.

In FIG. 6, reference numeral 35 is an address generator, which is a frequency dividing circuit 32, which is 1 / N ² of the fundamental frequency CLK.
The address of the page memory 30 is generated at the frequency divided by. Data expansion is performed by the latch 50 described later, and the address generator 35 changes the address of the page memory 30 by one dot for each clock. Basic frequency CLK is maximum resolution and page memory 3
The frequency for writing the image data to 0, and the resolution to 1
/ N, the amount of data to write is 1 / N ² 1
As the clock signal of the read address also becomes 1
The frequency will be reduced to 1 of / N ² .

In order to drive the heads A to D at the same time, the address generator 35 reads the image data supplied to the head A for a predetermined number of dots and then reads the image data for the head B. A, head B,
Image data is read alternately by a predetermined number of dots in the order of the head C and the head D so that the image data of each head is not interrupted.

Reference numeral 50 is a 1-bit latch, and if the page memory 30 has a 1-bit latch, a built-in one may be used. Reference numeral 40 is the data distributor, and 42 is the FIFO memory. 52
Is a head control circuit for performing time division control of the heads A to D, and 53 to 56 are shift registers of the heads A to D. FIFO memory 42 and shift registers 53 to 56
Are connected by a 4-bit parallel transfer line, and the head control circuit 52 operates at a frequency (1 / (4N) * CLK) obtained by dividing the basic frequency CLK into 1 / 4N. Page memory 3
0 to the FIFO memory 42 are the printer main body, and the head control circuit 52 and the subsequent are the print head. The head control circuit 52 controls the time division drive of the heads A to D, latches the data of the shift registers 53 to 56 to the head, generates a strobe signal to the head, and controls a switch for switching the LED array. .

The operation of the modified example will be described. First, the operation at the highest resolution of 1200 DPI is shown. In this case, N is 1
Therefore, the address generator 35 and the data distributor 40 operate at a frequency of 32 MHz, and the LED head operates at a frequency of 8 MHz with the 4-bit parallel transfer. The address generator 35 reads the data of the page memory 30 dot by dot at the frequency of 32 MHz, and the data distributor 40 also distributes the data to the FIFO memory 42 at the same frequency.
The 1200 DPI data of 0 is only divided into 4 and input to the FIFO memory 42.

Next, when the resolution is lowered, for example, N is 4
When the data is reduced to 300 DPI, the data in the page memory 30 is read at a frequency of 2 MHz, and the data distributor 40 distributes the data in the latch 50 at a frequency of 8 MHz. The dots are stretched to 4 dots. The counter built in the data distributor 40 counts the number of distributed data, and every time a predetermined number of data is distributed, the input destination in the FIFO memory 42 is changed to
Data is sequentially input to the banks A to D in the memory 42 by a predetermined number of dots.

The data input to the FIFO memory 42 is input to the shift registers 53 to 56 on the head side by 4-bit parallel transfer. Therefore, the transfer frequency is 1/4
To 2MHz. If the head control circuit 52 is operated with the clock of this frequency, the resolution of 30
The data of 0DPI is expanded to 4 times and 1200DPI.
You can print with the head.

When N is not 4, for example, N is 2 and the data is doubled, N is 3 and is tripled, and N is 6 and the data is 6 times. The expansion of these data is realized by reading the data of the latch circuit 50 into the data distributor 40 N times.

Although an LED head is used as an example here, an EL head, a plasma head, a thermal head or the like may be used.

[0040]

According to the first aspect of the invention, the image data of one dot is expanded into a plurality of dots to make the resolution of the printer variable. Therefore, the resolution can be changed by extremely simple processing. In particular, since the data in one line is processed in that line, it is not necessary to store the data for a plurality of lines in the middle, and the waste of memory can be suppressed. Also, since the image data is stretched to match the resolution of the print head, no complicated control algorithm is required and no reference table for data conversion is required.

According to the second aspect of the present invention, the resolution can be increased by 2n times and the resolution can be made variable by a method other than downsizing the light emitting body.

[Brief description of drawings]

FIG. 1 is a drive block diagram of a printer according to an embodiment.

FIG. 2 is a diagram showing a data structure in the printer of the embodiment.

FIG. 3 is a sectional view of the printer of the embodiment.

FIG. 4 is a view in which light emitters are arranged with a half-pitch shift and arranged.
Diagram showing a row light emitting array

FIG. 5 is a diagram showing a principle of image formation in the printer of the embodiment.

FIG. 6 is a block diagram of a modified printer.

FIG. 7 is a waveform diagram showing reading of data from a DRAM type page memory.

[Explanation of symbols]

 2 Photosensitive drum 4 Housing 6,8,10 Half mirror 12,14 1/2 reduction monocular lens array 16,18 Substrate 20,22 Substrate 24 LED array A, B, C, D LED head 30 Page memory 32 Dividing circuit 34, 35 Address Generator 36 Mode Selector 38 Gate Circuit 40 Data Distributor 42 FIFO Memory 50 Latch 52 Head Control Circuit 53-56 Shift Register

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁶ Identification code Internal reference number FI technical display location B41J 2/505

Claims (2)

[Claims] 1. A printer which supplies image data to an image element and prints one line at a time with a predetermined resolution, and a means for expanding the image data according to the ratio of the resolution of the printer and the resolution of the image data. A printer provided with. 2. A printer in which an optical image from a light-emitting body is formed by using a lens array, and two rows in which the light-emitting bodies are arranged with a shift of 1/2 of the arrangement pitch are provided as one set. , A half mirror for combining the light of the light emitters, and a monocular lens array for reducing the light of the combined light to form an image for each pair of the light emitters of the two lines, and the light emission of the two lines There are n sets of bodies (n is a natural number of 2 or more), a new half mirror for combining light from the n sets of light emitters is provided, and image data of 1 dot is N / m dots. Image data (N
Is a natural number of 2 or more, m is 1 or 2 and is a divisor of N, and means for distributing the image data stretched to N / m dots to the n sets of light emitters. A printer for changing the value of N so that the resolution of the printer can be changed to 1 / N of the maximum resolution.