US3898671A

US3898671A - Ink jet recording

Info

Publication number: US3898671A
Application number: US424025A
Authority: US
Inventors: James M Berry; Anthony J Hauser; Gary B Ollendick
Original assignee: Teletype Corp
Current assignee: AT&T Teletype Corp
Priority date: 1973-12-12
Filing date: 1973-12-12
Publication date: 1975-08-05
Anticipated expiration: 1992-08-05
Also published as: IT1026879B; CA1015818A; DE2458216A1; JPS5092645A; GB1485637A

Abstract

An ink jet printing device wherein conductive ink, under pressure is emitted from a nozzle and breaks into droplets in the vicinity of a charging electrode such that the charge on each droplet is determined by the voltage on the charging electrode at the instant that the droplet separates from the solid stream of ink issuing from the nozzle. A mask or drain catcher is provided for blocking and catching uncharged droplets of ink, and an electrostatic field is provided for deflecting charged droplets out of a path impinging on the drain catcher and into a path impinging on a record medium for printing thereon. A vibratory transducer is connected to the nozzle structure and imparts undulations at a fixed, droplet-forming frequency to the ink issuing from the nozzle so as to induce the stream of ink to break into droplets of uniform size and spacing, at the applied droplet-forming frequency. A data signal source is connected to the charging electrode, and the phasing between the data signal and the droplet-forming signal is randomly varied to obviate distortions due to frequent, regular, or periodic formation or separation of droplets during transitions of the voltage of the charging signals.

Description

United States Patent [191 Berry et al.

[ INK JET RECORDING [75] Inventors: James M. Berry, Deerfield; Anthony J. Hauser, Stone Park, Gary B. Ollendick, Chicago, all of ill.

[73] Assignee: Teletype Corporation, Skokie, Ill.

[22] Filed: Dec. 12, 1973 [2|] Appl. No: 424,025

[52] [1.8. Ci. 346/75 [SI] Int. Cl. 601d 15/18 [58] Field of Search 346/75 [56] References Cited UNITED STATES PATENTS 3,298,030 l/l967 Lewis et al 346/75 3.588.906 6/l97l Brimer et al 346/75 X Primary Examiner-Joseph W. Hartary Attorney, Agent, or FirmW. G. Doss; J. L. Landis [57] ABSTRACT An ink jet printing device wherein conductive ink,

FAME/UM LEL Y AMP Aug. 5, 1975 under pressure is emitted from a nozzle and breaks into droplets in the vicinity of a charging electrode such that the charge on each droplet is determined by the voltage on the charging electrode at the instant that the droplet separates from the solid stream of ink issuing from the nozzle. A mask or drain catcher is provided for blocking and catching uncharged droplets of ink, and an electrostatic field is provided for deflecting charged droplets out of a path impinging on the drain catcher and into a path impinging on a record medium for printing thereon. A vibratory transducer is connected to the nozzle structure and imparts undulations at a fixed, droplet-forming frequency to the ink issuing from the nozzle so as to induce the stream of ink to break into droplets of uniform size and spacing, at the applied droplet-forming frequency A data signal source is connected to the charging electrode, and the phasing between the data signal and the droplet-forming signal is randomly varied to obviate distortions due to frequent, regular, or periodic formation or separation of droplets during transitions of the voltage of the charging signals.

10 Claims, 3 Drawing Figures PRESSURE INK SOURCE PATENTED 51975 SHEET 1 PRESSURE INK SOURCE PICK-UR RANDOM DELAY PATENTEU AUG 5 I975 SHEET mmhtw EOQZ K It; mmmJDm 3.4m

x00 6 POD hxl INK JET RECORDING FIELD OF THE INVENTION This invention relates to ink jet recording and more particularly to systems for controlling the relative tim ing of ink droplet formation and transitions in the applied data signal.

BACKGROUND OF THE INVENTION In prior devices such as is disclosed in US. Pat. No. 3.500436 granted on Mar. 10. 1970. to R. W. Nordin, ink under pressure is ejected from a nozzle in a steady, solid stream. Undulations are imposed on the solid stream of ink by shaking the nozzle with a transducer or by various other means known to the prior art. The physical parameters of the ink stream control the point at which the solid stream breaks into a stream of discrete ink droplets.

A charging electrode is positioned in the region where droplet formation takes place. Data signals are applied to that charging electrode and are referenced with respect to the nozzle or the body ofthe conductive ink. Therefore. charges collect at the outer end of the solid stream of conductive ink which charges are of a magnitude proportional to the voltage difference between the solid body of ink and the charging electrode. When a droplet separates from the solid stream of ink as it progresses from the nozzle towards the paper or web on which the record is being printed, that droplet carries with it a charge which is proportional to the instantaneous voltage difference between the charging electrode and the solid stream of liquid ink at the instant that droplet formation or separation takes place. Substantially uncharged droplets of ink impinge upon a mask or drain catcher and are withdrawn to a sump and discarded. However, droplets having sufficient charge are deflected out of their initial path intercepting the drain catcher as they pass through an electrostatic or magnetic field, and are directed into a path whereby they impinge upon a web of paper or other record medium.

It is possible to form or separate a droplet during a transition of the data signal from a noncharging voltage to a charging voltage or vice versa. A droplet formed or separated during such a transition is neither fully charged nor fully uncharged but is charged to some intermediate and indeterminate voltage. Such an indeterminate charge level may cause the droplet to succeed in bypassing the drain catcher but strike the paper at some undesired location. This occurs because, a drop let having only a partial charge will not be fully deflected by the electrostatic or magnetic deflection field, but will be only partially deflected. Such spuriously charged droplets adversely affect the quality of the copy produced and are to be avoided.

The prior art contains several examples of systems designed to synchronize droplet formation with transi tions in the data signal such that each droplet is formed substantially in the center of each data signal pulse, so that each droplet will contain substantially the electrostatic charge quantity associated with discard via the drain catcher or only one other discrete electrostatic charge quantity associated with deflection to the proper location on the paper. Unfortunately, however. it has been found that through the unavoidable vagaries of a typical ink jet stream, such synchronization is very difficult to assure since the undulation of the transducer which produces a given droplet actually results in the production or separation of that droplet a substantial time after the occurrence of that transducer undulation. Consequently. any specific undulation that the transducer produces must be synchronized with a very much later data pulse applied to the charging electrode. During this rather extensive interval of time. considering the quantities involved, the ink jet may now be forming a droplet somewhat sooner or somewhat later than anticipated. It is difficult to monitor accurately an actual droplet formation without unreasonable, and commercially unacceptable circuit and apparatus complexity.

In another technique, the adverse effect of formation of successive droplets during successive transitions in the input data signal voltage in minimized by operating the droplet forming undulation-generating transducer at a frequency other than the characteristic bandwidth frequency of the data signal applied to the charging electrode or some multiple thereof. This, however, can result in a periodicity of the spuriously-deflected droplets. The eye is very good at discerning such periodic phenomena. Therefore, periodically-misdirected droplets are to be avoided.

SUMMARY OF THE INVENTION In accordance with the present invention, the adverse effect of either repeated or periodic formation of a droplet during a transition in the input data signal voltage is minimized by inducing a random variation in the phase relationship between droplet formation and the data signal.

BRIEF DESCRIPTION OF THE DRAWINGS A more complete understanding of the present invention may be had by reference to the following detailed description when considered in conjunction with the accompanying drawings wherein:

FIG. 1 is a schematic diagram of an ink jet recording system including the present invention;

FIG. 2 is a waveform timing diagram associated with the apparatus depicted in FIG. I; and

FIG. 3 is a schematic diagram of a portion of the circuit shown in FIG. I but with greater specificity as to the delay circuit.

DETAILED DESCRIPTION General Arrangement and Printing System Referring now to the drawings and more particularly to FIG. 1, there is shown an ink jet recording system in schematic form. A sheet of paper I] is positioned in front of a platen 13. A carriage 14, indicated by a dotted box, is mounted in front of the platen l3 fortransverse movement on a lead for transverse 15. The lead screw 15 is driven at a more-orless constant speed or with a stepped motion by a motor 16. A timing-andposition-encoding disc 17 is mounted at the end of the lead screw 15 for rotation therewith. The disc 17 indicates the advance of the carriage [4 across the paper 11 for synchronization of the printing of the dots which make up indicia to be printed on the paper 11.

The encoder disc 17 has divisions around its periph ery which are sensed by a pickup 19. The purpose of the encoder disc 17 and pickup 19 is to indicate to subsequent electronie circuitry those points in the motion of the carriage 14 at which dots of the indicia may be marked on the paper and to synchronize movement of the carriage 14 with the dot locations on the paper.

The output of the pickup I9 is amplified in an amplifier 2I which produces a clock pulse train, each pulse of which represents a possible dot location on the paper 15 as the carriage l4 sweeps across the front of the platen. This clock pulse train is referred to as the dot clock" and is illustrated in waveform A of FIG. 2.

The output of the amplifier 21 is sent via a random delay circuit 22 to a logic circuit 23. The logic circuit 23 generates a charging pulse in response to each dot clock pulse from the amplifier 21. Each charging pulse has a duration of zero or at least as long as the time interval between the generation or separation of successive ink droplets (referred to in the interval (Ta), assuming one ink droplet contains sufficient ink to print a dot on the paper 11. The duration of the charging pulse can preferably be controlled by a monostable multivibrator contained in the logic circuit 23. A typical charging pulse signal train is illustrated in waveform B of FIG. 2.

The signals illustrated in waveform B of FIG. 2 are shown as rectangular pulses with very sharp beginnings and endings. However, one skilled in the art well recognizes that a voltage signal such as that shown in waveform B and as applied to the charging electrode 28, cannot change magnitude instantaneously. Each pulse depicted in waveform B has some finite rise time and fall time dependent upon, among other factors, the output impedance of a charging amplifier 27.

The lower value of the signal of waveform B is a noncharging reference level. Droplets formed while the charging electrode 28 is at this lower. reference voltage level will strike a grounded mask or drain catcher 30. The upper value of the signal of waveform B is a charging level at which droplets are selected for printing on the paper 11.

In the ope ration of the system of FIG. I, the charging electrode 28 is first held at the lower reference voltage level of waveform B of FIG. 2. When a droplet is to strike the paper II, the logic circuit 23 and the amplifier 27 cause the voltage of the charging electrode 28 to rise within a finite period of time (the rise time) from the reference voltage level to the charging voltage level. The voltage of the charging electrode 28 then remains at that charging voltage level for an extended period of time as determined by whatever timing means is contained in the logic circuit 23. The duration of this higher voltage is long enough for at least one droplet to be charged and thus selected for impingement on the paper ll.

After an interval sufficient to charge the requisite droplet, the logic circuit 23 and the amplifier 27 may cause the voltage of the charging electrode 28 to fall within a finite period of time (the fall time) from the charging voltage level to the reference voltage level. Or, the logic circuit 23 may immediately call for an other droplet to darken the next dot location on the paper II.

The logic circuit 23 serves the purpose of determining whether each dot location on the paper 1] will re ceive a droplet of ink (a dot) or no ink (a blank spot). The logic circuit 23 could comprise any one of a number of possible sources ofdata signals which are known to one of ordinary skill in the art.

For example. assuming proper speed normalization, the output of a video tape recorder could be used as part of the logic circuit 23. The timing signal generated by the pickup 19 is then used to call from the video tape recorder the nature of the next dot location in the sweep of the carriage I4.

Information in the video tape indicating an unprinted dot location will result in the issuance to the charging electrode 28 of a voltage signal at a reference voltage very nearly the same as the voltage of the nozzle. thereby causing the droplet next generated to be substantially uncharged and thus undcflected. The undeflected droplets strike the drain catcher 45 and are discarded.

Information in the video tape indicating a printed dot location causes a charging voltage signal or pulse to be applied by the logic circuit 23 to the charging electrode 28, resulting in deflection toward the paper I] of the droplet next generated. The duration of the charging pulse can be determined by a monostable multivibrator within the logic circuit 23, as described previously. In one droplet per dot is inadequate several droplets may be used to darken each dot. The length of the charging pulse can be a multiple of the droplet interval T.,.

The recording head or carriage 14, in traversing from left to right, traces an imaginary horizontal line across the paper 11. The stream of ink droplets marks selective dots along the line of relative motion between the carriage and platen to form indicia on the paper 11. within the carriage 14 there is a nozzle mechanism 35 which comprises a nozzle 37 and a transducer 39. Ink is supplied to the nozzle mechanism 35 by a pressure ink source 41 which provides ink through a tube 43 at a substantial gauge pressure. The pressurized ink is emitted by the nozzle 37 along the axis of the nozzle in a path designated by a dotted line extending directly towards the drain catcher 30. If nothing more were done, the ink emitted by the nozzle 37 from the pressure ink source 41 would progress from the nozzle 37 in a solid stream and would break into droplets at some intermediate point. These droplets would continue on substantially the same path along the axis of the nozzle 37 and would be captured by the drain catcher 30. The ink captured by the drain catcher 30 progresses down a tube 47 to a drain sump 49.

If permitted merely to be emitted by the nozzle 37, the ink would break into droplets at some intermediate point as mentioned previously, but would break into droplets rather indeterminate size and spacing in accordance with the vagaries of surface tension and viscosity of the ink as well as random movements of air, machine vibration, and other various random phenomena. In order to provide better control of these ink droplets for the recording of intelligence, it is desirable that these droplets to be of rather uniform size and spacing. This is accomplished with the use of the transducer 39 that is energized by an oscillator 51 operating through an amplifier 53. The energized transducer 39 shakes the nozzle mechanism 35 or otherwise directly imparts undulations or varicositics to the solid stream of ink issuing from the nozzle 37. The frequency of the oscillator is preferably chosen to be the approximate frequency of natural droplet formation at the end of the solid stream of ink. Consequently. these undulations and varicositics in the ink induce the ink to break up into relatively evenly spaced and evenly sized droplets.

The tunnel-shaped charging electrode or charging tunnel 28 is positioned around the stream of ink issuing from the nozzle 37. The charging electrode 28 is long enough to be adjacent to the end of the solid stream of ink and also adjacent to a position where the ink is sure to have broken into droplets, Therefore, the charging electrode 28 is positioned in the region where ink droplet formation is sure to take place. The output of the charging amplifier 27 is connected to the charging electrode 28 and maintains the charging electrode at a predetermined voltage with respect to the voltage of the conductive ink issuing from the nozzle 37. This causes charges to collect at the outer end of the solid stream of conductive ink in proximity to the charging electrode 28. These charges are isolated by the air capaci tance between the conductive ink and the charging electrode 28.

The charges on the end of the solid stream of conductive ink are proportional to the magnitude of the volt age difference between the voltage of the solid stream of ink and the voltage of the charging electrode 28. As a droplet of conductive ink separates from the solid stream of ink, it carries with it a charge which is then trapped on that droplet. That charge on the newly formed ink droplet has a magnitude that represents and is also proportional to the voltage difference between the charging electrode 28 and the solid stream of ink at the instant of droplet separation or formation.

If a droplet is charged to a suitable level or magnitude, it will be deflected as it passes between a pair of

deflection electrodes

57 and 59 from its initial path that extends towards the drain catcher 30. The deflection electrodes are maintained at a constant voltage difference represented schematically by the battery 61; and they are maintained at some voltage reference with respect to the nozzle 37, for example, by grounding the electrode 59. A

As a suitably charged droplet passes through the gap between the

deflection electrodes

57 and 59, it is de flected by the electric field between the electrodes. The speed of the droplet passing between these two deflection electrodes, coupled with its charge-to-mass ratio and the voltage gradient betweenthe deflection electrodes is suitable to deflect a properly charged droplet away from the path intersecting the drain catcher 30, but that speed is not slow enough so as to cause the properly charged droplet to impinge on either of the

deflection electrodes

57 or 59. And, a properly charged droplet continues, as it exits the gap between the deflection electrodes, on a path towards impingement upon the paper I] on the platen l3.

Droplet Synchronization A sufficient gap is maintained between the drain catcher 30 and the deflection electrode 57 such that a suitably charged droplet can readily pass between them without impinging upon either one. This implies sufficient margin or tolerance such that ifa droplet is somewhat less than suitably charged, it might still be deflected by a sufficient magnitude to avoid capture by the drain catcher 30. If a droplet is formed during a transition in the voltage applied to the charging elec trode 28, it may carry only a fraction of the full charge level. A droplet charged sufficiently to avoid capture but not fully charged would proceed to impinge upon the paper II, but not at the desired location.

Waveform C of FIG. 2 represents in pulse form the ideal droplet formation syehronizing or phasing with respect to the data signals shown in waveform B. It is evident that it would be most desirable if all droplets would be formed in the interval midway between possi ble transitions in the charging signal of waveform B of FIG. 2. Assurance of such phasing is very difficult to attain.

The danger of attempting to maintain this desired phase relationship is that a drift or error of one-halfdrop-period places the system in exactly the worst phase relationship. With no further adjustments, the system will tend to stay in this adverse-phase state indefinitely. Waveform D of FIG. 2 depicts this possible, worstphase situation in which each droplet is generated or separates at the instant of a possible transition in the charging data signal waveform B.

Periodic misdirected droplets can also be objectionable, if the droplet frequency and data frequency are not quite identical, especially if many droplets are used to print each dot; because, the misdirected droplets tend to form regular patterns on the paper 11 which are far more discernable to the eye than just tiny dark dots randomly distributed as a background to the indicia.

It is evident that the problem of partially charged droplets is at least as severe if charged droplets are dis carded and uncharged droplets impinge on the paper ll.

Random Component The random delay circuit 22 of FIG. 1 is intended to prevent both consecutive misdirected droplets and regular or periodically misdirected droplets. Referring again to the circuit of FIG. 1, the output of the amplifier 21 is representative of the initiation of a dot on the paper as shown in waveform A of FIG. 2. The random delay 22 has a negligible effect on the location of the copy on the paper 11, but it causes the leading edges of the pulses of waveform A of FIG. 2 to jitter in time in a random manner, thereby preventing any regularity to the formation of a droplet during a voltage transition. This jitter is illustrated by waveform E of FIG. 2 which is waveform B with random variations in the starting transitions of the charging pulses. This introduction of randomness to the generation of droplets during the voltage transitions effectively prevents the spurious droplets from being consecutive or from causing recognizable, regular patterns in copy printed on the paper 11.

The random delay 22 could represent nearly random variation resulting from mechanical imprecision resulting in random variations in the output from the pickup 19 or electrical imprecision resulting in random variations in the output from the oscillator 51, or some combination of both. The random delay 22 could also comprise any one of several circuits common in the prior art. A voltage-controlled delay could even be used in conjunction with a circuit for producing signals of random voltage level. The difference between the maximum and minimum delay intervals need not exceed the magnitude of the interval between the formation of successive droplets.

FIG. 3 represents a specific embodiment of the random-delay 22 of FIG. I. The random-delay 22 is shown in dotted lines in FIG. 3 and includes a bistable multivi brator or flip-flop 66. The dot signal from the amplifier 2] is delivered to the set input of the flip-flop 66. The normal output of the flip-flop 66 is delivered to one of the inputs ofa two-input AND gate 67. The other input ofthe AND-gate 67 is derived from the output of a random noise pulse generator 68. Consequently, when the lead screw 15 (FIG. I) is in the appropriate position, the amplifier 21 provides a trigger pulse to the set input of the flip-flop 66 (FIG. 3). When the flip-flop 66 is switched to the set condition, it provides one input of the AND-gate 67. The next random noise pulse from the generator 68 provides the second input for the AND-gate 67. The AND-gate 67 then provides an output pulse signal to the logic circuit 23, causing the logic circuit 23 to initiate the appropriate charging pulse for the charging electrode 2! of FIG. I. The same random noise pulse that passes through the AND-gate 67 is also sent to the reset input of the flip-flop 66 and switches the flip-fiop 66 to its reset condition so that future random noise pulses from the generator 68 cannot pass through the AND-gate 67 until after the next timing pulse from the pick-up l9.

Although only one specific embodiment of the invention is shown in the drawing and described in the foregoing specification. it will be understood that invention is not limited to the specific embodiment described. but is capable of modification and rearrangement and substitution of parts and elements without departing from the spirit of the invention.

What is claimed is:

1. An improved ink jet printing system of the type wherein ink is delivered to a nozzle, a transducer vibrates the delivered ink in response to drive signals applied thereto, the ink breaks into droplets after issuing from the nozzle, a charging electrode is positioned in the region in which the ink jet breaks into droplets. the charging electrode is adapted to respond to data signals selectively to charge the droplets as they are formed. the improvement comprising:

data signal source means for applying data signals to the charging electrode at a first predetermined frequency;

vibratory drive means for applying drive signals to the transducer at a predetermined frequency which may differ from the first predetermined frequency, whereby a phase relationship exists between the formation of droplets and transitions in the data signals; and

means for randomly varying the phasing relationship between the data signals and the drive signals from data signal to data signal. thereby reducing the statistical likelihood that droplets will be formed on successive transitions in the data signals.

2. A system according to claim I wherein the predetermined frequency of the drive signals is the same as the first predetermined Frequency of the data signals.

3. A system according to claim I wherein the predetermined frequency of the drive signals is a function of the first predetermined frequency of the data signals.

4. A signal according to claim I wherein the predetermined frequency of the drive signals is a function of the first predetermined frequency and the rise time of the data signals.

5. A system according to claim 1 wherein the randomly varying means comprises a means for delaying by a random duration the application of data signals to the charging electrode.

6. A system according to claim 5 having a timing pulse generator means with an output for initiating application of the data signals to the charging electrode and wherein the random delaying means comprises means for imparting a random delay to the output of the timing pulse generator means.

7. A system according to claim 6 further comprising a random pulse generator having an output and wherein the means for imparting a random delay to the output of the timing pulse generator means comprises means for gating the output of the timing pulse generator means with the output of the random pulse generator.

8. A system according to claim 7 comprising means for storing the output of the timing pulse generator means for later gating with subsequent output of the random pulse generator.

9. A system according to claim I wherein the first predetermined frequency of the data signals is a function of the frequency of the drive signals applied to the transducer.

10. A system according to claim 1 wherein the predetermined frequency of the data signals is a function of the predetermined frequency of the drive signals and the rise time of the data signals.

i 0 i I i

Claims

1. An improved ink jet printing system of the type wherein ink is delivered to a nozzle, a transducer vibrates the delivered ink in response to drive signals applied thereto, the ink breaks into droplets after issuing from the nozzle, a charging electrode is positioned in the region in which the ink jet breaks into droplets, the charging electrode is adapted to respond to data signals selectively to charge the droplets as they are formed, the improvement comprising: data signal source means for applying data signals to the charging electrode at a first predetermined frequency; vibratory drive means for applying drive signals to the transDucer at a predetermined frequency which may differ from the first predetermined frequency, whereby a phase relationship exists between the formation of droplets and transitions in the data signals; and means for randomly varying the phasing relationship between the data signals and the drive signals from data signal to data signal, thereby reducing the statistical likelihood that droplets will be formed on successive transitions in the data signals.

2. A system according to claim 1 wherein the predetermined frequency of the drive signals is the same as the first predetermined frequency of the data signals.

3. A system according to claim 1 wherein the predetermined frequency of the drive signals is a function of the first predetermined frequency of the data signals.

4. A signal according to claim 1 wherein the predetermined frequency of the drive signals is a function of the first predetermined frequency and the rise time of the data signals.

5. A system according to claim 1 wherein the randomly varying means comprises a means for delaying by a random duration the application of data signals to the charging electrode. 6. A system according to claim 5 having a timing pulse generator means with an output for initiating application of the data signals to the charging electrode and wherein the random delaying means comprises means for imparting a random delay to the output of the timing pulse generator means.

9. A system according to claim 1 wherein the first predetermined frequency of the data signals is a function of the frequency of the drive signals applied to the transducer.