Electrical Print Impression Control

Abstract

Impression control for an impact printer is provided by changing the width of the pulse applied to the print hammers in accordance with the thickness of the forms on which printing is being performed and in accordance with the voltage of the source energizing the print hammers, so as to maintain a constant impact force to provide uniform print density for different forms thicknesses. Misregistration of characters caused by a variation in the rate of movement of the print hammer is compensated for by changing the start time of the pulse energizing the hammer.

Description

BACKGROUND OF THE INVENTION

1. Description of the Prior Art

U.S. Pat. No. 3,172,353 which issued to C. J. Helms on Mar. 9, 1965 discloses varying the voltage applied to the print hammer operating winding to provide impression control.

U.S. Pat. No. 3,183,830 which issued to D. M. Fisher et al. on May 18, 1965 discloses the use of delay means which are preset to adjust for variations in manufacturing tolerances between different print hammers.

SUMMARY OF THE INVENTION

Generally stated it is an object of this invention to provide an improved high speed printer.

More specifically it is an object of the invention to provide more uniform print quality from a printer.

Another object of the invention is to provide for varying the print hammer impact energy by modifying the width of the pulse driving the print hammer.

Yet another object of the invention is to provide for compensating for differences in flight time caused by varying pulse width, by varying the start time of the modified pulse driving the print hammer.

It is also an object of the invention to provide for maintaining a constant impact force in a printer with varying thicknesses of the forms on which printing is being performed.

It is also another object of the invention to provide for maintaining a constant impact force in a printer with changing source voltage.

Yet another important object of the invention is to maintain the impact force and impact time of a print hammer with varying forms thicknesses and varying source voltage.

The foregoing and other objects, features and advantages of the invention will be apparent from the following more particular description of a preferred embodiment of the invention as illustrated in the accompanying drawing:

DESCRIPTION OF THE DRAWING

In the drawing

FIG. 1 is a schematic block diagram of a printer control system embodying the invention in one of its forms.

FIG. 2 shows a set of curves illustrating the variation in pulse width with varying forms thicknesses to maintain a 21 lb. impact force.

FIG. 3 shows a set of curves illustrating the relations between the pulse width, flight time and source voltage for different forms thicknesses.

FIG. 4 is a schematic block diagram of the impression control portion of the system of FIG. 3.

FIG. 5 is a schematic circuit diagram of the forms thickness potentiometer and analog function mixer of FIG. 4.

FIG. 6 is a schematic circuit diagram of an inverting amplifier used in the analog function mixer and amplifier of FIG. 4.

FIG. 7 is a circuit diagram of a divider and vernier adjustment used in the analog function amplifier and mixer of FIG. 4.

FIG. 8 is a schematic circuit diagram of a pulse width generator used in the circuit of FIG. 4.

FIG. 9 is a schematic circuit diagram of a pulse width delay generator which is a non-linear function of the source voltage in FIG. 4 and

FIG. 10 is a pulse width delay generator that is a non-linear function of the forms thickness voltage in FIG. 4.

DESCRIPTION OF A PREFERRED EMBODIMENT

Referring to FIG. 1 the reference numeral 10 denotes generally high-speed printer apparatus of the type described in U.S. Pat. No. 2,993,437 which issued to F. M. Demer et al. on July 25, 1961 and has been modified for use with the present invention. As shown, a type cartridge 11 comprising an endless band 12 having a plurality of type characters 13 thereon is mounted between spacedapart wheels 14 and 15 for movement past a plurality of print positions along a print line on a document or form 22 upon which a printing operation is to be performed. A plurality of print hammers 20 are positioned one in each position along the print line for impacting the document 22 and a ribbon 24 against selected type characters 13 as they pass the different print positions. Timing marks 16 are provided on the band 12 which are scanned by a transducer 18 and used for producing timing signals for use in the control system.

Information to be printed is stored in a print line buffer 26 which is scanned by the usual X drivers 28 and Y drivers 30 under the control of an X ring 32 and a Y ring 34 as described in the Demer et al. patent. A subcycle gating ring 34 is provided so that the print positions are scanned in a predetermined spaced order as described in the Demer et al. patent. By using a three stage subcycle gating ring with stages 1, 2 and 3 scanning the buffer 26 in that order as described in the Demer et al. patent, stages 2, 3 and 1 of the subcycle gating ring may be used with the stages being used shifted one position to gate a pulse control system 40 for controlling the actual firing of selected ones of the print hammers 20 simultaneously for each subcycle, on the following subcycle. Instead of utilizing the hammer select circuit 175 of the Demer et al patent for directly firing the print hammers 20, the hammer select circuit 175 is used to set select ones of a plurality of latches 36 which are connected through AND circuit 38 to the different print hammers 20. The AND circuits 38 are gated in subcycle groups by the pulse control circuit 40 under the control of the subcycle gating ring 34 for actually firing the print hammers 20 which are selected during the previous subcycle. The type cartridge 11 is shown by the arrows as schematically movable toward and away from the print hammers 20 for accommodating different thicknesses of the forms 22. A potentiometer 42 is provided which is operatively connected to the cartridge mechanism so as to be operated when the cartridge is moved closer to or further away from the type hammers 20.

FIG. 2 shows the width of pulses applied to the print hammer magnets required to hold the print force constant at 21 lbs. for 1, 3 and 6 part forms at different values of voltage from the source applied to the print hammers 20. These conditions relate to print hammers having electromagnetic actuating means whose magnet cores do not saturate during the time that the drive pulse is applied. This condition allows a considerable range of control of the print hammer during the time it is energized, thus altering its output or operating characteristics. The main embodiment is the provision of a correct hammer kenetic energy to the paper forms to maintain essentially constant print density over the intended range of forms thicknesses. Other conditions relate to the reduction of (1) print density variations caused by the poorly regulated bulk power supply for the print hammers, and (2) print registration variations caused by dynamically altering the print hammer flight time to cause density control and (3) print registration variations caused by the same power supply variations as described.

FIG. 3 shows graphically what the circuit must accomplish. The pulse width curves F1, F3 and F6 are smoothed data representations of the curves shown in FIG. 2 and the F3.5 curve is calculated from the F1, F3 and F6 relationships: it represents the mid range for the forms control. Subscripts denote the relative number of forms. In reality the forms control is calibrated in thousandths of an inch and is continuously variable such that the pulse width generator circuitry of the pulse control system 40 must produce a pulse width representative of any point within the area A, B, C, D.

Observation of the pulse width function area A, B, C, D shows that for a particular forms thickness (any F curve) the required pulse width is a non-linear inverse function of the print hammer supply voltage. It also shows that (1) for any particular voltage the pulse width is a positive linear function of forms thickness but that (2) the discrete range of widths is an inverse function of the particular voltage.

Since the print hammer input energy and thus the hammer velocity is deliberately altered with forms thickness adjustment, and undesirably altered by supply voltage variations, the hammer flight time is being changed, and the result is a varying relationship between the hammer and type element at the time of impact. The flight time (FLTM) curves F1, F3 and F6 show the time between print hammer energization and hammer impact. (FLTM) is observed to be a non-linear function with respect to voltage and forms thickness in all respects.

It is desired that impact time be held constant relative to the type characters 13 of FIG. 1 to prevent printing misregistration; see the IMPACT TIME line. Since actual flight times under any set of conditions is described by the area A, E, G and H, the start of the pulse width generator circuit must be delayed by the amount from any point within the area up to the IMPACT TIME so that the registration will be held constant.

FIG. 4 shows a block diagram of the pulse control circuit 40 of FIG. 1 for delaying and controlling the width of the pulses used to fire the print hammers 20. As shown, a typical latch 44 is set by a pulse from the transducer 18 of FIG. 1 and a delay generator 46 is gated on by the latch 44 and its duration is a non-linear function of the 24 volt print hammer power supply. Also during its interval a delay generator 48 is reset. At the conclusion of the voltage controlled delay of delay generator 46, delay generator 48 is gated on and its duration is a non-linear function of forms thickness as determined by the setting of the potentiometer 42. When delay generator 48 times out, an output pulse causes the latch 44 to be reset and the delay printer cycle is complete and dormant until the next synchronization pulse is received from the transducer 18.

When the delay generator 48 turns off it also turns on one of three AND circuits 50, 52 and 54 which are gated by signals from the subcycle gating ring 34 to start the hammer fire pulses. For example if AND 50 is turned on, it turns on a pulse width latch 56, AND 52 likewise controls a pulse width latch 58, and AND 54 controls a pulse width latch 60. These latches provide energizing signals of controlled widths for their respective subcycle groups of print hammers 1, 4, 7 etc., 2, 5, 8 etc. or 3, 6, 9 etc. depending on which of the groups of latches 36 had been set. When the latch 56 turns on, it starts a pulse width generator 62 which determines the width and hence duration of the pulse applied to print hammers 1, 4, 7, etc., and the latches 58 and 60 start corresponding pulse width generators 64 and 66; the duration of pulse width generators 62, 64 and 66 is dependent upon two independent variables and is non-linear with respect to each. The pulse width generators 62, 64 and 66 are controlled by an analog function mixer and amplifier circuit 68 which is responsive to the voltage of the 24 volt source and also to the setting of the potentiometer 42. When the pulse width generator 62 turns off, the latch 56 is reset and the pulse applied to the print hammers 1, 4, 7, etc. is terminated.

In practice it is likely that the required pulse width is greater than the interval between synchronization pulses. This problem is alleviated by utilizing identical sets of circuits controlled by the generators 62, 64 and 66 and causing them to be gated on in a skewed fashion by the subscan gating ring 34. For example, subscan pulses 1, 2 and 3 are used in sequence to scan the print line buffer 26 while subscan pulses 2, 3, and 1 are utilized in that order to control the firing of the respective groups of print hammers 1, 4, 7; 2, 5, 8, and 3, 6, 9, etc.

Referring to FIG. 5, the forms thickness sensing potentiometer 42 is shown as connected to the base and collector of a transistor T1 which with an emitter resistor 70 comprises a common collector amplifier to buffer the potentiometer 42, and it drives each pulse width generator 62, 64 and 66 via terminals 18a, 18b and 18c. The forms sense voltage from the potentiometer 42 appears at mixer point A of the mixer 68 with less than unity gain while the magnet supply voltage +V and its variations appear there amplified and in phase through the common base amplifier consisting of resistor 72 transistor T2 and the parallel resistors 73, 74 and 75. Transistor T3 and resistor 76 form a common collector amplifier to buffer point A and provide drive to terminals 24a, 24b and 24c.

FIG. 6 shows an inverting amplifier comprising transistor T4 which provides a +V correction current to each pulse width generator via terminals 25a, 25b and 25c such that the detector is not a function of the generator capacitor charging current. The load for this amplifier is the collective input impedances of the driven circuits. Capacitors C1 and C2 in FIGS. 5, and 6 respectively limit the circuit band width to prevent pulse width generator variations for rapid excursions of the print hammer supply voltage. Rapid variations of the supply voltage do not cause appreciable reactions in the print hammer circuit because of magnetic and mechanical integration.

In FIG. 7 resistors 78 and 79 and potentiometers P1, P2 and P3 form a divider and vernier adjustment so that each of the pulse width generators may be set to the same time duration. These generators must track over a relatively wide range or they will inject print registration variations; print density would vary also but it would be relatively small. Terminals 28a, 28b and 28c feed the adjustable bias voltage to the individual pulse width generators.

FIG. 8 shows a pulse width generator, for example, the pulse width generator 62 of FIG. 4, each of the generators 62, 64 and 66 being identical. The adjusted bias voltage at terminal 28a establishes the level of current in the wire B (and thus the current through transistor T9). Its magnitude is the difference between the currents in resistors 80 and 81 which are functions of +V and the forms thickness respectively. The current at B is transmitted to a current in wire D which is less than the current in B because the transistors alphas are less than unity. The voltage at point E is related to the voltage at point J which is a composite function of the voltage +V and of the forms thickness.

In the equilibrium state terminal 29a is plus and transistor T5 is on, so that all the current in wire D is diverted to ground through T5, and point G is approximately zero volts. Transistor T7, diode 83 and resistors 84, 85 and 86 form a clamp amplifier with point K sufficiently positive for transistor T8 to be on and saturated. Terminal 35a is at +5 volts.

When pulse width latch 62 turns on, terminal 29a goes to ground and transistor T5 turns off. Current in D is now diverted through capacitor C3. The bulk of this current flows through tthe resistor 84, but the excess is sufficient to cause transistor T7 to saturate. Point K goes to zero volts, so diode 83 is reverse biased and transistor T8 is turned off. Point L rises to +5 volts and capacitor C4 is discharged to zero volts across it. These conditions persist until the ramp voltage at point G approaches the voltage at point E.

As transistor T10 saturates its emitter current is diverted to base current, the current at D falls such that it can no longer keep transistor T7 saturated. The voltage at point K rises, transistor T8 turns on, and the voltage at point K drops to zero volts. This 5 volts negative shift is coupled through capacitor C4 to terminal 35a.

The negative going pulse at terminal 35a resets the pulse width latch 56. Terminal 29a goes positive and transistor T5 turns on. Point G is forced to zero volts and current in D is again diverted through transistor T5 and capacitor C3 is discharged. Transistor T6 is wired as a clamp diode to protect transistor T7 and to reduce the circuit restoration time by allowing a high discharge current to flow through capacitor C3 and the transistors T5 and T6. When transistor T6 is no longer forward biased the remainder of the discharge current is supplied through diode 83 and resistor 85.

When the voltage +V is high, the currents in B and D are relatively high and the current in excess of that in resistor 84 is much larger than when +D is near its low limit. As a result for a given value of voltage at point J the voltage at point G would rise higher with a high +V than with a lower value before transistor T7 would turn off. The actual pulse width would then grow progressively longer than the predicted value. The voltage at terminal 25a drops as +V increases, so that as current at D increases, the amount of the increase is diverted through resistor 86. The base current of transistor T7 is now not a function of the +V so that the pulse width function is not affected.

The time required for transistor T10 to saturate is a complex function of (1) the rate at which capacitor C3 charges as controlled by current at D, which is a non-linear function of the two independent variables, and (2) the voltage at point J which controls point E, and which is a linear function of the same two independent variables.

The following is an analysis of pulse width T as a function of print hammer supply voltage V and forms thickness voltage F. All discussion is related to FIG. 8 but the data is obtained from FIG. 3.

______________________________________ DATA TABLE (Area ABCD) ______________________________________ No. of Forms 1 3.5 6 V .DELTA.V T.sub.1 T.sub.3.5 T.sub.6 21.6 .phi. 1.271* 1.438 1.603* 24.0 2.4 1.110* 1.250 1.388 26.4 4.8 1.007* 1.127 1.247* ______________________________________ *See text

This data set describes three data points on each of three curves in FIG. 3 which represent the minimum, mid-range and maximum forms thicknesses. Time is given in milliseconds.

The following definitive equations are stated:

(1) V = V.sub.o + .DELTA.V where V.sub.o = 21.6V

(2) f = f.sub.o + .DELTA.F where F.sub.o = 1V and .DELTA.F = 2.4V for each form in excess of 1

(3) M = a bias voltage to be determined; the voltage at terminal 28a will then be offset from M by the factor V.sub.BE for transistor T9

(4) j.sub.o = a voltage to be determined when V = V.sub.o and F = F.sub.o

(5) J = J.sub.o + a.DELTA.V + b.DELTA.F where (a) is the gain of the voltage .DELTA.V and (b) is the gain of the voltage .DELTA.F as seen at point J

(6) .DELTA.v.sub.c = J + N where N is a voltage constant that includes typical values of V.sub.BE and V.sub.CE SAT for T10, V.sub.CE SAT for T5 and .DELTA.V.sub.BE for T7; .DELTA.V.sub.c is the voltage change across capacitor C3 during the time the generator is gated ON.

(7) i.sub.b = ((v-m) .div. r.sub.80) - (m-f) .div. r.sub.81

(8) c = c3 .div. .alpha..sup.2 where .alpha. is the typical common base current gain of transistor T9 or T10; this is equivalent to using I.sub.D in the subsequent equations rather than I.sub.B

(9) t = c.DELTA.v.sub.c .div. I.sub.B

Substitution of equations (1) through (8) into (9) yields:

(10) T = (R.sub.80 R.sub.81 C (J.sub.o + a.DELTA.V + b.DELTA.F + N)) .div. (R.sub.81 (V.sub.o -M + .DELTA.V) - R.sub.80 (M -F.sub.o - .DELTA.F))

This can be put in the form

(11) K.sub.o T.DELTA.V + K.sub.1 T + K.sub.2 .DELTA.V + K.sub.3 + K.sub.4 .DELTA.F + K.sub.5 T.DELTA.F = (0) where

(12) K.sub.o = R.sub.81

(13) k.sub.1 = (r.sub.81 (v.sub.o - M) - R.sub.80 (M-F.sub.o)) .div. R.sub.81

(14) k.sub.2 = -aR.sub.80 C

(15) k.sub.3 = -r.sub.80 c (j.sub.o +N)

(16) k.sub.4 = -bR.sub.80 C

(17) k.sub.5 = r.sub.80 .div. r.sub.81

by successive substitution of the (*) values from the data table into equation 11 the K factors can be evaluated but six unknowns are present. A final solution is found by iteration with assumed values of J.sub.0 until the maximum value of J does not exceed (M-1.1D) to guarantee base drive to both transistors T9, T10.

FIG. 9 shows a pulse width delay generator that is a function of a print hammer voltage + V and corresponds to delay generator number 46 of FIG. 4.

The dash-enclosed portion 88 performs for this delay generator the same function that the circuit of FIG. 5 performs for the delay generator in FIG. 8. The generator 46 is gated on at terminal 89 by the delay fire latch 44 of FIG. 4 and its output at terminal 90 is used to control the next generator 48 of FIG. 4. The analysis and operation of this generator are identical in principle to those described for the generator 62 of FIG. 8.

FIG. 10 shows a pulse width delay generator 48 of FIG. 4 that is a function of the forms thickness voltage F from potentiometer 42 of FIG. 4. The dash-enclosed portion 92 of the circuit performs for this generator the same function that the circuit of FIG. 6 performs for the circuit of FIGS. 8 and FIG. 9. The forms control voltage is used directly at terminal 18m. This generator is reset when terminal 90 is positive and the voltage function generator in FIG. 9 is on.

When terminal 90 goes negative again the generator 48 is turned on, and at the conclusion of the delay a negative pulse occurs at terminal 94. As shown in FIG. 4 this pulse then resets the delay fire latch 44, and sets a pulse width latch 50, 52 or 54 as selected by the ring counter 34. The analysis and operation of this generator are identical in principle to those for the generator in FIG. 8. The reset of the generator in FIG. 10 occurs immediately before it is turned on, and transistor T13 must be held off during the transition between reset (terminal 90 positive) and turn on (terminal 90 negative). Capacitor 96 and resistor 97 insure that transistor T11 turns off, thus allowing transistor T12 to saturate before resistor 98 allows transistor T14 to turn off. This action keeps the base of transistor T13 cut off during that interval; the normal action of transistor T12 turning off and thus allowing transistor T13 to turn on, can then occur at the end of the delay because the transistor T14 is off. A minus OR function is performed by resistors 99 and 100 at their junction.

The cumulative delays of the circuits of FIGS. 9 and 10 effectively cause any point within the area AEGH in FIG. 3 to be delayed so that any print hammer will cause printing to occur at the same time relative to the synchronizing pulse that relates electrical timing back to a mechanical position of the type characters. In the actual circuits the voltage and forms delay generators insert fixed minimum delays of 50 and 57.5 microseconds, respectively, so that the common impact time is 107.5 microseconds later than shown in FIG. 3. The voltage delay generator 46 provides the function necessary to move the curve EJG up to the Impact Time Line. The forms delay generator 48 provides the function necessary to move any curve that parallels EJG up to coincidence with curve EJG. This delay characteristic is represented by curve JKL (as it would appear when plotted against a linear forms thickness scale).

In operation the print line buffer 26 of FIG. 1 is scanned by the X drivers and Y drivers under the control of the subcycle gating ring 34 to provide 3 subscans of all possible print positions. The hammer select circuit 175 sets the different hammer latches 36 in response to a compare between a character contained in the print line buffer for a given print position and the character on the type chain 12 in that position. The pulse control circuit 40 operates to gate the output of the selected latches 36 through the ANDS 38 to fire the selected print hammers 20. The pulse duration of the gating pulses applied to the ANDS 38 by the pulse control circuit 40 will be a function of the pulse width generator outputs and the delay of the pulses will be a function of the delay generators 46 and 48 in response to the source voltage and forms thickness respectively.

From the above description and the accompanying drawing it will be apparent that a new novel printer control circuit is provided utilizing a pulse control circuit which controls the timing and the duration of the pulses applied to the print hammers to not only maintain a constant print force but to also provide good registration regardless of the number of forms (up to six) or of variations in the source voltage.

While the invention has been particularly shown and described with reference to a preferred embodiment thereof, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention.

Claims

1. In a control system for a printer having a type carrier bearing a plurality of type characters and movable along a print line past a plurality of print positions for impacting one or more forms having a different thickness to print thereon,

a plurality of print hammers aligned with said print positions each having unsaturated electromagnetic actuating means to impart print energy to said hammers, said electromagnetic actuating means being selectively operable in response to electrical pulses applied thereto from a source of electrical energy to actuate said print hammers to impact said forms and type characters, said print energy varying with changes in the voltage of said source and the width of said pulses,

circuit means connected to effect selective energization of said print hammer electromagnetic actuating means from said source of electrical energy in accordance with the presence of particular type characters at each of said print positions, and

control means including means responsive to at least one of the variable operating conditions comprising variation of the voltage of the source and variation in forms thickness which affect the print hammer impact force connected to said circuit means and to said print hammer electromagnetic actuating means to selectively apply energizing pulses of variable width and hence duration to said print hammer electromagnetic actuating means in accordance with said variable operating conditions to correct for changes in print impact force caused by said variable operating conditions so as

2. The invention as defined in claim 1 characterized by said control means operating to dynamically vary the width and duration of the energizing pulses of said print hammers and effect a variable delay in the time of

3. The invention as defined in claim 2 characterized by said control means being operable to vary the delay in initiating said energizing pulses of said print hammer in the same sense as a variation in voltage of the source of energization and vary the width and hence duration of the

4. The invention as defined in claim 2 characterized by said control means being operable to reduce the width and hence duration of the energizing pulse of said print hammers with a reduction in the thickness of the forms on which printing is being done and reduce the delay in initiating said

5. The invention as defined in claim 1 characterized by said circuit means including a plurality of latches which are selectively set in accordance with the particular print hammers to be fired, and said control means being connected to gate means connected between said latches and said

6. The invention as defined in claim 5 characterized by said control means including a pulse width latch, connected to said gate means and a pulse width generator selectively producing pulse of variable width and hence duration responsive to but in non-linear relation with respect to source voltage and forms thickness connected to said pulse width latch to reset said pulse width latch at different times in a non-linear relation to said source voltage and forms thickness, so as to determine the duration of the energization of said print hammer electromagnetic actuating means by a

7. The invention as defined in claim 5 characterized by said control means including a fire delay latch connected by gate means to said pulse width latch and controlled by a delay generator which is non-linear with respect to source voltage and a delay generator which is non-linear with respect to forms thickness connected in cascade with said fire delay latch to

8. The invention as defined in claim 5 characterized by said circuit means including a subcycle gating ring connected to said control means to effect selective energization of said latches in groups in a predetermined order, and said subcycle gating ring being also connected to said control means in a different order so as to gate said latches on the following subcycle in which they are selected.