EP0013296B1 - Multiple speed ink jet printer - Google Patents

Description

Technical area

The present invention relates to an inkjet printer capable of operating at different speeds, with concomitant modification of the resolution of the characters.

State of the art

The IBM 6640 inkjet printer uses the principles described in US-A-3,596,275. In this machine, a continuous jet of pressurized ink from a nozzle is disturbed and divides into droplets at a predetermined distance from the nozzle. These droplets receive a charge which is applied to them by a charging electrode and then pass between two deflection electrodes to which a high voltage is applied, the value of the charge thus applied to the ink droplets determining their vertical deviation with a view to formation. of characters or other symbols on paper or other similar support placed in the immediate vicinity of the deflection electrodes. If no character or symbol is to be formed on the support, the droplets receive no charge, or receive only a minimal charge; they are therefore not deflected when they pass between the deflection electrodes and are intercepted by a gutter, from where they are brought back to their starting point for reuse.

It appears from the above that the vertical deflection of the droplets makes it possible to form or trace a column of ink stains or dots on the plane of the paper. Each of the droplets of a given column receives a slightly greater charge than the droplet which precedes it, so that the column is formed, in the example cited, from bottom to top. The characters are therefore formed, in a given matrix, by means of a series of columns of ink stains. In principle, the various elements of the print head are arranged on a mobile carriage which carries out a predetermined path along the paper so as to allow the printing on the latter of characters or other symbols.

The IBM 6640 printer has a resolution of characters which makes it possible to obtain a print quality practically equal to that of a traditional typewriter, this resolution being of 94.5 pixels (hereinafter called »PELS ") / Cm. In the case of an inkjet printer, a peel is defined as being a point formed on the sheet of paper or other support by an ink droplet. For example, in the case of a resolution of 94.5 pels / cm, a character printing frame with a height of 4.23 mm (height of the character matrix), and printing performed in steps of ten (i.e. ten characters per inch or 3.94 characters per cm), we can form a matrix with a width of 94.5 / 3.94 = 24 pels and a height of 94.5 x 4.23 mm = 40 pels. In this respect, in the IBM 6640 printer, the tolerance relating to the positioning of the peels is only ± 0.033 mm, while the tolerance relating to the height and width of the characters is ± 0.076 mm, which is necessary. to obtain the required print quality. The height of the matrix requires a maximum deviation of the droplets of 0.51 cm since it takes about 0.09 cm to prevent them from being intercepted by the gutter.

The printing speed of the IBM 6640 is around 90 characters per second and the character resolution is close to that obtained with a classic typewriter of excellent quality. However, it often happens that such a high resolution is not essential, especially if the document that is printed is for internal use or, for example, an advertising circular, an address list or any other document of the same type. .

There are currently two U.S. patents in the prior art. which are analogous to the present invention. The first of these two patents, US-A-3 878 517, relates to an ink-jet printer of the type in which the magnitude of the charge is controlled and whose printing speed is either high or low, the circuit control located in the character generator generating a signal which controls the drive speed provided by a servo motor connected to the carriage. On the other hand, although the frequency of ink droplet formation remains constant, the control circuit, depending on whether the printer operates at high speed or at low speed, transmits a control signal to an adjustable frequency divider which changes the ratio of the number of droplets receiving a video signal to the total number of droplets formed, so as to compensate for variations in speed. In the printer of the present invention, the ratio of the number of droplets loaded to the number of droplets not loaded remains essentially the same whether the machine is operating in coarse printing modes or in fine printing mode. There is no change in the frequency of appearance of the number of charged droplets or in the ratio resulting from the number of charged droplets to the total number of droplets formed.

The second patent, US-A-3 938 641, describes a sequential dot matrix printer provided with a drive device of the type step by step. In this latter printer, the printing speed of the dot columns is controlled by modifying the printing speed signal. In either of the two operating modes described above, the frequency is not subject to any variation to modify the carriage speed or the printing speed.

The IBM Technical Disclosure Bulletin, Vol. 20, No. 12, May 1978, pages 5097-98, describes a print head with threads allowing after the printing of a series of dots, the printing, during the return of the carriage of a first series of dots additional in the height direction thanks to a slight tilting of the print head and a second series of additional points in the width direction thanks to a change in pitch of the carriage carrying said print head. However, this document does not describe the operating mode of the ink jet printer of the present invention.

Statement of the Invention

One of the objects of the present invention is therefore to provide an ink-jet printer making it possible to obtain the printing of characters whose resolution (or definition) can vary as described in claim 1. In a first embodiment of the present invention, the printer operates in a first mode called coarse printing (compared to the fine or high quality printing previously mentioned), in which the signal for the start of column printing is regularly suppressed or periodic, so that if you delete, for example, one of these signals out of two, the machine only prints a column of dots out of two also, which makes it possible to approximately double the speed of movement of the cart. In this respect, it is possible to avoid having to modify the predetermined angle of the deflection electrodes, if the inclination of these electrodes is undesirable in the case where the machine operates in the coarse printing mode, by implementing the process described in the document FR-A-2 412 411 filed on November 23, 1978 by the applicant which makes it possible to automatically modify the inclination of the deflection electrodes as a function of the speed of movement of the carriage. In a second embodiment of the present invention, the printer operates in a second coarse printing mode, in which only one dot out of two is printed in each column, which has the effect of deleting a row of dots on of them. However, in the latter case, it is necessary to correct the aerodynamic effects specific to the droplets and to compensate for the interactions between the charged droplets. Thus, in the first embodiment of the invention, in which only one column of dots is printed out of two, although the width and the height of the frame in which each character is printed does not vary, the maximum number of peels (or dots) in each row, in the case of a printing carried out, for example, in the step of ten, is twelve (24/2), while the maximum number of pels that each column contains remains forty. In the second embodiment of the invention, in which only one row of pels is printed out of two, the maximum number of pels in each row of said frame is 24 (in the case, as before, of a printing in steps of ten), but only 20 pels in each column, this achievement, like the first, to accelerate the speed of movement of the carriage, while maintaining the predetermined ink flow. On the other hand, it is easy to combine the removal of one column out of two with the removal of a row of points out of two in order to increase the speed of the carriage, so that it is equal, for example, four times that obtained in the case where the printer operates in the so-called fine printing mode. In this regard, it should be noted that only minor modifications to the existing IBM 6640 printer and its electronic circuits are necessary for the machine to be able to operate at least in the coarse printing mode in which only one column out of two is printed.

It will also be noted that, in the case of the IBM 6640 printer, the average diameter of the ink droplets is 0.06 mm.

In fine print mode, the dots or pels of the print matrix are separated by a center-to-center distance of 0.1 mm. However, since the ink tends to spread on the printing medium (which is generally constituted by paper), there is an increase in the diameter of the dots or of the peels, the value of which then becomes approximately equal to 0, 15 mm, which results in an overlap of the dots or peels on the paper. In this context, in the first coarse printing mode of the present invention, which is the preferred mode, the dots are separated horizontally by a distance of 0.2 mm center-to-center, and by a vertical distance of 0, 1 mm center-to-center; in the second coarse printing mode (deleting every other row of dots), the dots are separated horizontally by a center-to-center distance of 0.1 mm and vertically by a center-to-center distance of 0 , 2 mm. In the case where the two modes are combined, the points are obviously separated, vertically and horizontally by a center-to-center distance of 0.2 mm. In summary, the resolution is sacrificed in favor of a corresponding increase in speed, without however the predetermined ink flow rate being modified.

Other objects, characteristics and advantages of the present invention will emerge more clearly from the following description, made with reference to the drawings appended to this text, which represent a preferred embodiment thereof.

Brief description of the Figures

• Figure 1 schematically shows an ink jet printer incorporating the present invention.
• Figure 2 shows an enlargement of a character printed using the machine of Figure 1 operating in the fine print mode and using a first resolution, appropriate network signals and a column start print signal being represented under said character.
• Figure 3 is similar to Figure 2 and shows the same character formed when the printer operates in the first coarse printing mode using a second resolution.
• Figure 4 is similar to Figure 3, but the character shown is that obtained when the printer is operating in the second coarse printing mode.
• Figure 5 schematically shows the electronic circuits of an inkjet printer incorporating the devices of the present invention.
• Figure 6 is a logic diagram illustrating a first logic pertaining to the first coarse printing mode.
• Figure 7 shows the circuits relating to the positioning of the ink droplets as well as a logic diagram relating to the logic of the first coarse printing mode.
• Figure 8 schematically shows the correction data memory used in the second coarse printing mode.
• Figure 9 is a time diagram relating to the various signals represented in mnemonic form in Figures 6 and 7, for the purposes of correct operation of the printer in the first coarse printing mode (which is the preferred operating mode ).

Description of the invention

In the ink-jet printer schematically represented in FIG. 1, ink 1 maintained at a determined pressure is expelled from a nozzle 2a, by means, for example, of a droplet generator 2, under the form of a jet. In order to prevent the latter from dividing into droplets having quasi-random dimensions and spacing, the ink contained in the nozzle cavity is made to vibrate at a fixed ultrasonic frequency by means, for example, of a circuit 3 which excites a piezoelectric crystal placed inside the droplet generator 2. The pressure waves thus generated cause the jet 1 to be divided into a series of droplets having uniform dimensions and spacing, at a good distance defined from nozzle 2a. A typical droplet generator is described in the publication "IBM Technical Disclosure Bulletin", Vol. 21, No. 5, October 1978, pages 1949-1950.

Each droplet thus formed receives an electric charge of a predetermined size when a voltage is applied to a charging electrode 4 which surrounds the point where the ink jet divides into droplets. The droplet retains this charge for the duration of its flight to a support 5 consisting, for example, of a sheet of paper. The different droplets pass through an electro-static field obtained by applying a fixed high voltage across the terminals of a pair of deflection plates 5 arranged horizontally. Each droplet receiving a separate charge, it can be deflected vertically by a desired distance. In the case of the IBM 6640 printer, the droplets are deflected vertically from bottom to top, so as to form on the sheet of paper a column entirely or partially composed of dots or peels. The droplets which are not used for the purpose of printing a character and which therefore must not appear on the paper receive no charge and are not deflected. These latter droplets are intercepted by a gutter 6 and brought back to the droplet generator 2 for a new use, via an ink tank 6a, a filter 6b, an ink supply 6c and a pump 6d, which continuously supplies ink to the generator 2. A typical ink recirculation system is described in patent US-A-3,929,071. On the other hand, since the height at which a charged ink droplet is deflected and projected onto the sheet of paper 5a is a direct function of the time interval during which it remains between the deflection plates 5 (the longer this time interval, the higher the is projected), the speed of the droplets or inkjet must be precisely controlled. The U.S. Patent No. 3,787,882 describes several servo systems enabling this speed to be controlled by detecting the latter, then by controlling the pump so as to obtain the desired pressure in the droplet generator 2. Another method for determining the speed of the droplets is described in French patent application FR-A-2405754 filed on September 8, 1978 by the applicant.

The droplet generator 2, the charging electrode 4, the deflection plates 5 and the gutter 6 are mounted on a carriage 7 which is driven horizontally and follows a predetermined path which brings it towards or away from the plane of the paper of the Figure 1, at a relatively constant speed during the printing operation. The droplets are therefore deposited on the paper at suitable locations inside a character frame or a frame so as to form the desired character or symbol. The carriage 7 is coupled to a drive device 7a, comprising for example a motor operating in direct current, which is excited by signals delivered, for example, by an excitation device 7b (in the preferred embodiment) under the control of the electronic circuits 8 of the system.

When the printer is operating, ie in the continuous printing mode, it is essential that means are provided to indicate the position of the carriage at any given time so that the start of each column printing can be determined and that the charging electrode can receive the correct data or the appropriate voltage level to deflect the ink droplets as required. On the other hand, means must be provided to determine the direction in which the carriage moves. In this regard, the printer comprises a detector 9 comprising a matrix 9a which emits and which receives and a concave mirror 9b arranged on the carriage on either side of a fixed network 9c of markers (mounted on the chassis of the machine), which network makes it possible, in conjunction with the matrix and the mirror, to generate signals which are transmitted to the electronic circuits of the system, both for applying a voltage to the charging electrode and for controlling the drive device 7b and, consequently, the device 7a connected to the carriage 7. The detection method and device used are described in more detail in European patent application EP-A-0 006 441 filed on May 22, 1979 by the applicant. The detection circuits used are also described in European patent application EP-A-0 006 482 filed on June 1, 1979 by the applicant. The network 9c consists of two parts offset by 90 ° with respect to each other and includes opaque and transparent marks. The array 9a includes a pair of light sources and a pair of detectors; the light which crosses the network is reflected by the mirror 9b and strikes the detectors which, by means of the circuits which are associated with them, generate alternating signals, such as those represented under the letter "B" of Figures 2, 3 and 4 (for the sake of clarity, the signals generated by the two detectors are respectively referenced "Det. A" and "Det. B"). The output of one of the two detectors can be used in conjunction with a counter to count the transitions (opaque to transparent) and, the network comprising a predetermined number of marks, to indicate with precision the position occupied by the carriage 7 the along its course. In the example shown in Figures 2, 3 and 4, the output of detector B is used to indicate the position of the carriage and provides the information relating to the start of printing of each column.

The pulses indicated in FIGS. 2, 3 and 4 and designated SOS indicate the start of printing of a column by means of a series of droplets. For example, if the network associated with detector B has 23 divisions per cm, and if the resolution is 94.5 pels / cm, each of the divisions of the network must be divided by four to obtain the resolution required to form a character such as the letter "B" shown in Figure 2. (The output of the second detector, such detector A, is used for comparison with the output of detector B to inform electronic circuits of the direction in which is moving carriage.)

According to the present invention, the resolution of a character can be modified so that it becomes that of, for example, the character represented in FIG. 3, this latter resolution being defined below as corresponding to the first printing mode. coarse. This result is obtained, in the case of character "B" in FIG. 3, by increasing the speed of movement of the carriage and by simultaneously suppressing one column start printing signal (or SOS signal) out of two, this signal being used to apply a voltage to the charging electrode 4. The second coarse printing mode can be obtained by printing only one row of dots out of two when plotting a column. In this regard, account should be taken of the fact that normal aerodynamic repulsions should be corrected and the interactions between charged droplets should be compensated for, for example by means of a modified correction memory. We can also combine the first and second coarse printing modes, in which case we obtain a reduction in the resolution of the printed character, such as the character "B" shown in Figure 4. In the latter case, we can obtain an increase of the printing speed, an increase which can be equal to four times that used to print the character "B" shown in Figure 2 and which corresponds to the fine printing mode.

FIG. 5 shows the electronic circuits 8 of the printer incorporating the present invention. The circuits schematically represented in FIG. 5 are essentially the same as those used in the IBM 6640 printer, with the exception of block 30, representing the logic relating to the coarse printing mode, and some modifications made to block 19, relative to the positioning of the ink droplets, blocks which are respectively described below with regard to FIG. 6 and of FIG. 7.

In FIG. 5, a microprocessor or other similar device 10 comprising other input devices 10a, such as a typewriter, a disk drive device, etc., functions as the main system vis-à-vis screw of the inkjet printer schematically shown in Figure 5. An input / output channel consisting of 14 lines is connected to the printer, namely eight (in the example shown) data lines, four control lines, an interrupt line and a line for transmitting signals from a main clock. Interface 11 of the printer provides, in a conventional manner, the transfer, logic, and amplification functions and the signals supplied by the main clock are applied to the system clock 12, in which frequency division circuits divide the frequency received from the main clock, which is converted to four clock frequencies, T1 to T4, shown in Figure 9. The signals from the input / output channel can be amplified and appropriately stored in memory that serial instructions can be received from microprocessor 10.

The data received by the electronic circuits of the printer and relating to the printing instructions consist of three bytes, as explained below. As an example, if we assume that the printing line has a maximum length of 35.5 cm and has 23.6 divisions per cm, as was previously indicated in connection with the 9c network of Figure 1, a total of 840 divisions, a single byte of data will give only 2 ⁸ or 256 bits, which is insufficient. Since it takes at least 840 bits to give the position code to the printer, in the case of a 35.5 cm print line, and since 210 = 1024 bits is the value closest to the code effective position, two bytes are necessary for the position signals transmitted to the printer by the microprocessor. On the other hand, in the system actually used, an additional byte is necessary to identify the signal defining the character to be printed, or the function to be executed, and the interface 11 therefore includes a buffer memory associated with these three bytes coming from the microprocessor. . In the IBM 6640 printer, the first byte of a printing instruction is an address which is applied, via the interface 11, to the character generator 13, defining the character to be printed. (Assuming it is recognized as a print instruction.) The generator 13 transmits the full or partial address of this character, via the common line "address", to memory 14, and receives from the latter the data relating to said character by means of common data lines 15. The data thus extracted from the memory 14 relate to the printing of a single column of dots forming part of this character.

The second and third bytes of the signal received from the microprocessor are applied to a logic circuit 16 called mechanism control, to inform the latter of the position of the first column of dots of the character to be printed.

In the case of an instruction relating to the execution of a function, the first byte is applied via the interface 11 to the logic circuit 16 in order to be decoded and to trigger the execution of the function whose it's about; for example, bit configuration 11010101 is decoded as representing a tabulation function and configuration 10110101 is decoded as defining a backspace function. The second and third multiples can also be applied to logic circuit 16 to perform appropriate actions when the carriage occupies the position specified by these bytes; for example, if a tabulation instruction has been decoded, the second and third bytes specify the point where the carriage should stop after the tabulation function is executed.

The functional block 16 can be used to perform many functions. For example: (1) control and synchronize the operation of circuit 18 of French patent application FR-A-2 410 248 filed on September 26, 1978 by the plaintiff, which circuit carries out a periodic check of the series of droplets of ink to determine if the height at which these are deflected is within the prescribed tolerances; (2) serve as an instruction decoder for the synchronization and servo control operations of an ink pump such as that described in patent US-A-3,787,882; (3) controlling the operation of the drive device 7a of the carriage and consequently the horizontal speed (in the present example) of the carriage 7 in the case in particular of the use of the excitation device 7b; (4) serve as a motor drive control device and receive the network position signals emanating, for example, from the detector 9 shown in FIG. 1 (cf. European patent applications EP-A-0 006 441 and EP-A-0 006 482 cited above). The variable speed drive can obviously take the form described in the publication entitled "IBM Technical Disclosure Bulletin", Vol. 20, No. 10, March 1978, pages 3993-3994, but it is preferable that it be of the type described in the patent application US-A-954374 indicated above.

When the data extracted from memory 14 and relating to the printing of a column of dots has been loaded into the shift register 39 (see Figure 7) of the character generator, a signal is transmitted via the line control common 13a to circuit 16, which then puts the associated mechanism in a position in which it is ready to proceed with printing, that is to say in the printing position. When the data thus extracted are in the character generator 13 and a print signal has been applied to the circuit 16, a signal indicating that the printing of a column can begin (or PIC signal) is transmitted by the intermediary of the common line 13b to the logic circuits 19 for positioning the ink droplets. A second signal known as the start of column printing (MCSOS) emanating from the circuit 16 and coinciding with the movement of the carriage to a predetermined point of the network 9c (Figure 1), this displacement being carried out in accordance with the data contained in the second and the third byte of the initial signal received from the microprocessor 10, causes the generation by the circuits 19 of an output which is applied via a common line 19a to a digital / analog converter 20, so that a load voltage is applied to the charging electrode 4 (FIG. 1) of the printer, in order to charge the various ink droplets so that a column is printed in accordance with the data extracted from the memory 14 and transmitted to the character generator 13.

The generator 13 continues to receive data relating to the printing of a column of the character designated by the first byte and to transmit via the common line 13a a print signal to the circuit 16 so that the cycle printing of this character continues until a so-called end of character printing signal is transmitted by the generator 13 to the circuit 16 via the line 13a. Furthermore, the circuits 19 generate another signal which opposes the transmission of the signals until the said circuits are ready to accept from the character generator 13 new data which will be used to apply the required voltages to the electrode. charging the printer. On the other hand, if new data relating, for example, to another character, is to be printed immediately after those which precede them, the printing signal transmitted on line 13a takes precedence over the end of character signal, such that the column printing start signal continues to be generated, and data continues to be retrieved from memory 14, etc., so that the required voltages can be applied to the charging electrode. In this regard, it will be noted that the memory 14 can comprise several unalterable memories, the selection of a determined assortment of characters being a function of the input received from the microprocessor 10. A method making it possible to change the assortment of characters by selecting different parts of memory is described in the US patent No. 3,963,591.

In the so-called fine printing mode, which makes it possible to obtain characters such as the "B" represented in FIG. 2, the machine operates essentially in the manner described above, which is that used in the case of the printer 1 BM 6640.

However, it often happens that one can be satisfied with a coarser resolution in order to obtain, in return, an increase in the printing speed. In this case, the first coarse printing mode can be used to produce characters such as "B" represented in FIG. 3. To this end, logic circuits 30 known as coarse printing are inserted between circuits 16 and 19 (Figure 5). These circuits 30 only intervene in the case where it is desired to obtain a rough impression. When fine printing is required, by appropriate instruction received from microprocessor 10, an SOS signal, which is actually a converted MCSOS signal (see below), can directly access circuits 19 bypassing circuits 30 .

For the purposes of implementing circuits 30 (see Figure 6), a signal (part of the three-byte signal from microprocessor 10) being decoded by circuit 16 and indicating that the speed of the machine must be increased and the resolution reduced so as to form characters of the type represented in FIG. 3, the decoding of, for example, 11110100 gives a first signal (COARSE PRINT or IG) whose first appearance indicates a change to the coarse print mode. The signal IG (see Figure 9) is applied to one of the inputs of a flip-flop of the master / slave type or of a pair of flip-flops 31 so as to produce several signals at different clock times, T1 to T4, as shown in Figure 9. In Figure 6, the flip-flop 31 consists of NI gates, designated 31A to 31 H in the particular embodiment shown, and a single inverter 31j, to obtain the locking signals "(INLK)," not locked "(INLK)," sequence "(SEQ)," not sequence "(SEQ), and" not rough printing "(IG). The signals "power-on restore" (POR) or "no power-on restore" (POR) are the conventional signals obtained from the logic used to start or initialize the machine. As shown in the figure, the input signals applied to the flip-flop 31 are the clock pulses T1 and T4 as well as the first signal IG coming from the decoding part of the circuit 16. These signals are applied to a second flip -flop or to a second pair of flip-flops 32 the output of which is constituted by a signal called "rough printing speed" (VIG) which is applied to the motor control device to increase the speed of the motor 7a associated with the carriage 7 of the printer.

As shown in the figure, the master / slave flip-flop or the pair of flip-flops 32 also consists of NI gates 32a at 32h and includes an inverter 32j at its output. Some of the inputs of the flip-flop 32 are generated by this pair of flip-flops, for example the signals known as "excitation of the high-speed master flip-flop" (HSPM) and "non-excitation of the high-speed master flip-flop" (HSPM), which has the effect of bringing the inputs known as "high speed slave flip-flop excitation" (HSPS) and "non-excitation of the high speed slave flip-flop" as inputs to the gates NI 32a and 32b. "(HSPS), as well as the clock pulse T4. The signals SEQ and INLK are obtained from flip-flop 31.

A third master / slave flip-flop or pair of flip-flops 33 receives an MCSOS signal from the carriage position logic that comprises the circuit 16. As previously explained, this MCSOS signal is obtained from the detector 9 of Figure 1 and the circuits which are associated. On the other hand, as will be seen below, the MCSOS signal and the SOS signal are one and the same signal when the printer is operating in the so-called fine print mode, which allows characters such as that shown in Figure 2. As previously indicated, the MCSOS signals generated are a multiple of the number of lines or divisions in the network. Thus, in the example cited, if the network has 23.6 divisions per cm, 94.5 MCSOS signals will be generated per cm of movement. When the speed of the carriage increases, the frequency of repetition of these signals obviously increases, but the distance traveled, which is representative of each of these signals, will always remain the same. In this respect, the master / slave flip-flop or the pair of flip-flops 33 functions as a divider by two circuit generating a column printing signal emanating from the "master" part of the flip-flop (SCNM) and an analogous signal. supplied by the "slave" part (SCNSM), as well as the logic inverses of these signals, all of this in relation to the clock instants T2 and T4 which constitute the inputs of flip-flop 33. The latter consists of NI gates 33a at 33h and an AND gate 34 whose first input receives the SCNS output from gate 33h. The AND gate 34 has a second input which completely avoids the flip-flop 33 and is constituted by the MCSOS signal. In this way, when the signals SCNS and MCSOS are both at the high level and coincide, the AND gate 34 generates an output, which is the SOS signal. Since the flip-flop 33 functions as a divider by two circuit, when the machine operates in the coarse printing mode, the SCNS signal is at the high level only half the time and, therefore, the signal SOS has only half the frequency of the MCSOS signal. The SOS signal shown in Figure 3 therefore has only half the frequency of the SOS signal shown in Figure 2, so that, in comparison with the fine print mode, every other column is not printed . On the other hand, in the fine printing mode, the signal SCNS is always at the high level, and consequently, when this signal coincides with the signal MCSOS, the signal SOS is generated by the AND gate 34 at a frequency which is necessarily the same as that of the MCSOS signal. The various inputs received by the flip-flop 33 are generated by the HSPS signal emanating from the flip-flop 32, and by the flip-flop 33 itself, as shown in the Figure.

The SOS signal thus generated is applied to the logic circuit 19 for positioning the droplets which, as previously indicated, is excited when it receives this signal from circuit 16 together with a signal "ready to print column “(PIC) emanating from the character generator 13 and generated as described below. As can be seen in FIG. 7, the logic circuit 19 comprises an input flip-flop or flip-flop of the RS 35 type, which is coupled to a second flip-flop or to a second flip-flop RS 38, and a third flip-flop or a third flip-flop 37, appropriate input signals being applied to a pair of flip-flops and to a flip-flop of the master / slave type 38. Flip-flops 35 to 37 consist respectively of Ni gates 35a to 35d, 36a at 36d, 37a to 37c, while the pair of flip-flops 38 consists of NI doors 38a at 38h. A shift register 40 which is controlled by a control circuit 41 makes it possible to transfer the data relating to the printing of the columns, as described in more detail below, to the digital / analog converter 20 (FIG. 5), via the common line 19a, and to the charging electrode 4 associated with a nozzle 2a which comprises the droplet generator 2.

Logic circuit 19 operates as follows. When the column start printing signal (SOS) is applied to the input flip-flop 35, if the rest of the circuit composed of flip-flops 38, flip-flops 37 and 36 has completed the operations relating to the printing of the column of previous points, the flip-flops generate a loading signal of the shift register, or LDSR signal (see output from gate NI 37b) allowing the transfer of the data relating to the printing of a column, from the forty-bit shift register 39 contained in the character generator 13, at the forty-seven bit shift register 40. As shown in the figure, the shift register 39 has a serial data input SDI which receives the data relating to printing from memory 14 of a column. Once the register 39 has been loaded, a signal is generated and applied to the circuit 39a for decoding the PIC signals (which circuit is an AND gate with multiple inputs) which, when it simultaneously receives the LDSR signal emanating from the gate 37b, generates a PIC signal to indicate that the printing of a column is ready to be carried out. When it receives this same LDSR signal as well as the "column pulse" (SCN) signal emanating from the pair of flip-flops 38 (NI gate 38h), the control circuit 41 authorizes the transfer of the data relating to the printing of a column, via the common line 39d, to the shift register 40, and subsequently via the common line 19a, to the digital / analog converter 20.

The flip-flop or flip-flop 35 allows you to store the SOS signal until the printing of the previous column is complete (note that the NI 35a gate receives as inputs a delayed SOS signal (SOSD) and a signal T2 clock however that the NI gate 35b receives in particular a signal SCNST, meaning "non-stored SCN", and a clock signal T4). The signal “stored SCN (SCNST) is obtained at the output of the pair of flip-flops 36. In this way, the circuit shown in FIG. 7 makes it possible to process column printing data presented in series and of exceptional length. that can overlap the signal Next SOS, a catch-up can take place if these data are longer than those relating to the printing of the previous column and the next column.

In the printer'lBm 6640, it is necessary to compensate for the aerodynamic effects specific to the ink droplets, as well as the interactions which can occur between charged droplets, so that the droplets strike the sheet of paper at the relative positions desired. The method used in the IBM 6640 printer is described in US-A-4,086,601. According to said method, the least significant bits emanating from the shift register 40, for example the last seven bits or more, are used as the address of the main memory 42 of a correction data memory, as shown in FIG. 8. The output of the main memory 42 is applied to a common line called memory data and transmitted to the digital / analog converter 20 by via the common line 19a so that the appropriate voltage can be applied to the charging electrode for each bit representation of a droplet forming part of the printing of a column. The correction data memory uses a threshold voltage to prevent the droplets intended to be printed from being intercepted by the gutter. Further details will be found in the U.S. patent. cited above.

The present invention also makes it possible, instead of printing only one column out of two, to skip a row of printing points out of two or to print only one point out of two when printing a column (although this printing mode is not the preferred mode, the visual impression obtained may be more pleasant in the case of certain types of character). For example, omitting one out of two droplets increases the printing speed so that it is approximately twice as much as omitting one column out of two. that obtained in fine print mode. In this regard, the pair of flip-flops 33 in Figure 6 is unnecessary and the MCSOS signal can be directly applied to the flip-flop RS 35 in Figure 4, the latter signal then being equivalent to the SOS signal. However, the rest of the circuit of Figure 6 can be used to obtain the so-called coarse printing speed (VIG) signal which is applied to the motor control device associated with the carriage in order to increase the movement speed of the latter. .

The second modification required to print only one row of dots out of two in each column consists in ensuring that the signal VIG is applied to the control circuit 41 of the shift register (this signal being designated VIG ^* so that only one data out of two relating to the printing of a column is loaded into the shift register 40 via the common line 39b, for example under the control of the loading pulse generated by the control circuit 41. In this way, the shift register receives only one binary data out of two, which indicates that one ink droplet out of two is not intended for printing and must be intercepted by the The third necessary modification, once again assuming that the process used to compensate for the interactions between charged droplets and the aerodynamic effects is that described in the aforementioned patent U $ -A-4,086,601, consists in selecting the memory 43, related e in the coarse printing mode, so that the required corrections can be made to the droplets via the common line "memory data" from this memory, instead of being from the main memory 42.

In a variant of the present invention, the techniques for printing a column of dots out of two and a row of dots out of two described above can be combined so as to obtain a character similar to that shown in the Figure 4 in the coarse printing mode, In this case, the MCSOS signal should be at the high level to allow printing of one column out of two and the VIG ^* signal should be applied to both control circuit 41 of the register shift 40 and memory 43 for the purpose of selecting the correction data memory. On the other hand, the microprocessor 10 should obligatorily inform the control circuits of the machine that, in the coarse printing mode requested, the operating speed should be approximately equal to four times that achieved in the fine printing mode, and this by means of a signal obtained from the signal VIG of FIG. 6, and would further transmit a second signal of which one would perform, for example, the AND function to indicate the increase in speed required by the microprocessor.

The present invention therefore provides an ink-jet printer capable of varying the resolution of the characters it prints while making a corresponding variation in the printing speed, this result being obtainable by making minor modifications to the existing equipment. .

Although the essential characteristics of the invention applied to a preferred embodiment of the invention have been described in the foregoing and represented in the drawings, it is obvious that a person skilled in the art can provide all of them. modifications of form or detail which it judges useful, by substituting means by equivalent means within the framework of protection as defined by the claims.

Claims (11)

1. A continous ink jet printer of the type in which a carrier (7) carries a nozzle (20) for emitting a stream of ink drops at a predetermined rate, an ink drop charging electrode (4) and ink drop deflection electrodes (5), to effect printing of characters on a medium by forming thereon columns of ink dots, a first predetermined character resolution being defined by a predetermined maximum number of picture elements per unit of width and beight in a predetermined size character box, said printer being characterized in that it includes:

a character generator means (13) for generating data representative of the charge to be applied to selected drops being a portion of the printing of a single columns of characters and detection means (19) to indicate the position and direction of movement of said carrier at any point in time and for generating a first so called »start of columns printing« signal indicative of the time of the. start of the printing of each column of picture elements forming a portion of a given character, an alterable speed carrier drive means (7a, 7b), an ink drop placement circuit means (19) for receiving said data and said first signal and compled to said charge electrode (4) for applying a charge to said ink drops in accordance with said data to form a single character at a first resolution and at a first speed of the carrier, the following means being used increasing the print speed on said media while maintaining the ink rate at a predetermined value,

a first circuit (31, 32) for receiving a second signal indicating a change in speed of formation and resolution of a character, said first circuit including means (32) for generating a third signal on receipt of said second signal for changing the speed of said drive means so that it drives said carrier at a second speed greater than said first speed, and

a second circuit (33) responsive to said second signal for forming a given character at a second resolution and at a second speed of carrier movement greater than said first speed.

2. An ink jet printer in accordance with claim 1, characterized in that said second circuit includes means (33h) for rendering selected ones of said start of column printing signals on inoperable.

3. An ink jet printer in accordance with claim 1 characterized in that said second circuit includes means (33h) responsive to said second signal and said first start of column printing signal, for regulary or periodically eliminating the start of column printing signal.

4. An ink jet printer in accordance with claim 2 characterized in that it includes means (34) for by-passing said second circuit, so that said first signal can be applied to said ink drop placement circuit (19) in the absence of said second signal.

5. An ink jet printer in accordance with any one of claims 1 -4 characterized in that it includes means (35) to permit application of said third signal to said ink drop placement circuit (19), and means for eliminating the data representative of the charge to be applied on alternate drops of the printing of a single column.

Images