EP0009033A1

EP0009033A1 - Printing device

Info

Publication number: EP0009033A1
Application number: EP79900244A
Authority: EP
Inventors: Nori Sinoto
Original assignee: Individual
Current assignee: SINOTO Nori
Priority date: 1978-02-15
Filing date: 1979-09-11
Publication date: 1980-03-19
Also published as: SE7908476L; FR2477071A1; GB2035905B; GB2035905A; IL56669A0; DE2936561A1; DK432979A; WO1979000627A1; BR7906616A

Abstract

Un dispositif d'impression a matrice comprend des elements d'impression adjacents (41A-41N) deportes par rapport a l'horizontal d'un angle ((Alpha)) et decales verticalement (V) d'une quantite suffisante pour permettre aux centres des lignes horizontales (42, 43) des points successifs d'etre disposes plus pres les uns des autres que cela ne serait possible si les centres des elements adjacents devaient etre espaces verticalement de distances egales aux diametres des elements adjacents.A matrix printing device includes adjacent printing elements (41A-41N) offset from the horizontal by an angle ((Alpha)) and offset vertically (V) by an amount sufficient to allow the centers horizontal lines (42, 43) of successive points to be arranged closer together than would be possible if the centers of adjacent elements were to be spaced vertically by distances equal to the diameters of adjacent elements.

Description

Printing Device

Technical Field

This invention relates to the field of matrix printers to print letters, numbers and other graphic symbols with higher resolution than has heretofore been possible in aymbols having a total height of less than four millimeters. In particular, the invention relates to printing heads having adjacent printing elements spaced center-to-center as required by the diameters of the elements but offset relative to each other both horizon¬ tally and vertically to permit visibly separate rows of dots to be printed closer together than would be possible based on the center-to-center spacing of the elements.

Background Art

Matrix printers are commonly used with calcu- lators and computers and other electrically operated devices to print alpha-numeric information. Standard Arabic numbers and standard Roman letters, both capital and lower case, can be legibly printed in a matrix of 7 x 9 dots or even 5 x 7 dots arranged in a checkerboard pattern. A 7 x 9 array permits a maximum of four spaced vertical lines and five horizontal lines to be printed, if the dots are a full space apart, but this small number of lines is not sufficient to form a clear or even a legible reproduction of the approximately 2000 Chinese characters most commonly used in Japanese printing. A minimum resolv¬ ing power of about 13 x 13 dots is required, and preferably an array having at least 17 x 17 dots should be provided.

A typical printing head does not have the total number of printing elements suggested by the multiplication of the two numbers that define the array. A 5 x 7 array does not have five columns of seven printing elements, nor does a 7 x 9 array have seven columns of nine elements. Instead, in a full-space matrix a checkerboard pattern of 35 dots is produced by providing only a single vertical column of seven printing elements and moving the printing head five horizontal incremental steps to print the 35-dot array and a 7 x 9 printer has a printing head with only one vertical column of nine printing elements. The printing head is moved seven horizontal incremental steps to print a 63-dot checkerboard array.

There are also printers that generate half-space matrices in which the head moves in increments half a dot wide. However, the speed of movement of the head, called the scanning speed, is too great to permit overlapping dots to be printed, because the printing elements cannot be operated at the necessary repetition rate. Dots can be printed in any non-overlapping positions. A disadvantage of half-space matrix operation is that a 9 x 7 half-space matrix can only form three separate vertical lines, not five as in full-space matrix operation.

The number of non-overlapping dots that can be formed in a given space is determined by the size of the printing of the printing elements. The dots are basically circular and the diameter of the printing elements is usually the main limitation in minimizing the spacing of the dots. A preferred type of impact matrix printing element is a ballistically operated wire, or needle, that is struck at one end by an armature of a solenoid to be projected forward like a spear to print a dot and then rebound with the aid of a spring. It is essential that the impact of the armature cause the needle to move axially and not simply to bend.

About the smallest diameter needle capable of operating for a commercially acceptable length of time has a diameter of about 0.35 mm, and it^' is common practice to support the front portions of the full complement of such needles, typically seven or nine, side by side in a bearing made of suitable material, such as synthetic ruby or alumina, with a corresponding number of highly polished holes closely spaced in a row. The diameter of each hole is about 0.37 mm and their centers are about 0.37 mm apart. Needles smaller than about 0.35 mm in diameter, down to about 0.2 mm diameter, have been used but require a support sleeve, which limits the closeness of spacing.

The center-to-center spacing of adjacent bearing holes determines the distance between the centers of adjacent horizontal rows of dots so that adjacent ver- tically spaced dots do not overlap substantially and this limits the number of such rows that can be formed in a given vertical space. The row of bearing holes is nor¬ mally perpendicular to the direction of incremental movement of the printing head and is therefore normally vertical.

In some printing heads the row of bearing holes is slanted slightly so that the letters slope as hand¬ printed letters frequently do, but the center-to-center spacing of bearing holes in such heads is great enough so that the dots formed by adjacent needles do not overlap substantially. Such an arrangement is described in British patent specification 1,343,570.

One of the principal objects of this invention is to provide a printing head capable of printing the Chinese symbols used by the Japanese people and called kanji

Many kanji are quite complex, requiring as many as twenty or more strokes for a single kanji. Moreover, the Japanese government has indicated that, for standard

O PI telecommunication and computer usage, the kanji should not exceed 3.7 mm in height. Kanji normally fit within a square area, and the resolution needs to be about equal in the horizontal and vertical directions. The lower limit of legibility for the more complex kanji requires a dot matrix of about 13 x 13 full-spaced dots capable of making at least seven horizontal and seven vertical lines. Higher resolution matrices are desirable to increase legibility.

Disclosure of Invention

This invention arranges the printing elements- of a matrix printing head so that adjacent elements are offset both horizontally and vertically with respect to each other to cause the centers of the horizontal rows of dots to be spaced more closely than would be possible if the same elements were arranged in the most compact vertical column possible.

The fact that the horizontal rows of dots pro¬ duced by adjacent printing elements may overlap partially, but less than half way, makes the horizontal rows wider than the minimum width of white space between two hori¬ zontal rows, but it is not necessary that the white space be as wide as the printed rows, only that it be clearly visible as a separation between rows. In addition, there are many locations in even complex kanji where no dots are required.

The complexity of kanji makes it desirable to use half-space matrices.

Instead of arranging thirteen or more printing elements in one printing head to print each symbol fully in one sweep across the page, the head may have a standard seven or more elements to print either the upper or lower

^"BU

O

, Λ half of each symbol on a line in one sweep and to print the other half in the next sweep, which may be in the reverse direction. Printing in both directions is common¬ place, and the memory system used in such printers can delay the actuating signals to each element to compensate for the horizontal offset thereof.

In order to provide ample space for actuating means for a relatively large number of needles, separate bearings may be provided for more than one set of needles. The separate multi-needle bearings can be mounted in precise, offset relationship to each other to allow the full number of horizontal rows to be printed in one sweep of the printing head and without having to move the paper incrementally less than one line at a time.

Additional objects of the invention will be brought out in the following specification and the related drawings.

Brief Discription of Drawings

Pig. 1 is a perspective view of a matrix printer in which the present invention is incorporated.

Fig. 2 is a simplified drawing of a ballistic printing element as used prior to the present invention and as used in this invention.

Fig. 3 shows a magnified portion of the front end of the printing head in Fig. 4 and a line of dots illustrating the printing of a horizontal line.

Fig. 4 is an illustration of an ar¬ rangement of printing elements according to the prior art.

^'BUREAU

O PI

< *JV Iθ Fig. 5 is a geometrical construction of an arrangement of printing according to one embodiment of this invention.

Fig. 6 is a side view of an improved bal- listically operated printing needle constructed according to one embodiment of the invention.

Figs. 7A and 7B show alternative embodiments of a printing head of the invention with two sets of offset printing elements.

Best Mode for Carrying Out the Invention

Fig 1. shows a typical matrix printer 11 for printing graphic symbols including letters, numbers, punctuation marks, and other symbols on a recording medium of which paper is the most common example. In the printer 11, the paper 12 runs over a cylindrical platen 13 in front of a printing head 14. The head moves laterally on a pair of guide rods of which only one rod 16 appears in the drawing. The operation of the printing head in printing the symbols and in moving along the guide rods, and the operation of the platen to feed paper through the printer 11 are all electrically controlled, either by operation of a keyboard 17 or by electrical signals from other sources.

Unlike a typewriter, which prints each symbol in its entirety at one stroke, a matrix printer builds up each symbol by printing an array of dots, each of which is produced at a specific location relative to the other dots required for any given symbol. The dots that make up a symbol are not all printed simultaneously, except for the simplest symbols, such as a decimal point, but are printed in a specific sequence. Each dot is printed by a printing

-_0 P/ / element, and the sequence in which the dots are printed depends on the location of the printing elements relative to each other. The printing end, or front end, of each printing element is at a location identified by reference numeral 18.

Fig. 2 shows a common printing element arrange¬ ment. The element is a thin wire, or needle, 19 sup¬ ported in a front jewel bearing 21 and rear guide 22. The front end 23 of the needle 19 faces the paper 12 wrapped partly around the platen 13. The paper may be a type that darkens in response to pressure, alone, which is referred to as "no-carbon" paper, or, as illustrated in Fig. 2 , it may be regular paper that requires an inked ribbon 24, which is guided through the space between the paper 12 and the front end of the needle 19.

The needle 19 moves to the left in the arrange¬ ment as illustrated in Fig. 2 when struck by the armature 26 of a solenoid 27. The solenoid includes a U-shaped core 28 of low-carbon steel or other ferromagnetically soft material with an energizing coil 29 around one leg of the core. The armature 26 is pivotally supported at the end of the other leg and is biased clockwise by a spring 31. The needle 19 is resiϊiently biased to the right by a spring 32 compressed between the plate 22 and a collar 33 affixed to the needle- near its back end 34.

The slight curvature of the needle 19 is due to the fact that this needle is only orie of a group of such needles, usually seven or nine in number, arranged so that their front ends are in a straight line and are parallel and as close together as possible. The respective sole¬ noids for each of the needles require much more space, and so the back ends of the needles are fanned out in a circular arc arranged to minimize the curvature of the needles.

^"BU E^ _O P_I The needle arrangement in Fig. 2 is referred to as a ballistic system because each needle, like the needle 19, receives a sharp impact force from the armature 26 when the solenoid 27 associated with that needle is energized. The counterclockwise movement of the armature 26 is limited, but the force of the impact is great enough to cause the needle to fly to the left, in spite of the concommitant, increased compression of the spring 32, until the front end 23 strikes the ribbon 24 and forces the ribbon against the paper 12. The spring 31 retracts the armature 26 by pivoting it clockwise immediately after the impact against the back end 34. of the needle 19. After the needle has flown ballistically forward in response to this impact and has caused a dot to be printed on the paper 12, the needle rebounds to the right, aided by energy stored in the spring 32.

There are other ways of obtaining, dot matrices, including non-ballistic impact, ink squirting, electric discharge, and others, but ballistic systems are widely used in practice and will be used in the description of this invention. The invention is not limited to a bal¬ listic system.

Fig. 4 is an enlarged drawing of only the jewel bearing 21 and the front ends 23a-23i of nine needles 19a-19i. The bearing 21 may be synthetic ruby and, in accordance with common practice, is divided into two mirror image parts 21a and 21b. The guiding surfaces for the needles 19a-19i are arcuate recesses 36a-36i in one edge of the bearing part 21a and corresponding arcuate recesses 36a'-36i' in the facing edge of the bearing part 21b.

The diameters of the openings defined by each pair of recesses is slightly greater than the diameter of the needle guided by that pair. For example, a standard

^'BΌ

_0 needle diameter is approximately 0.35 mm and the diameter of the bearing opening to accomodate needles of that size is 0.37 mm The center-to-center distance between each adjacent pair of the openings is also 0.37 mm, which is, therefore, also the center-to-center spacing between adjacent needles. It would be theoretically possible for each of the recesses 36a-36i and 36a'-36i' to be a semi¬ circular arc, but this would mean that the bearing ma¬ terial would taper down to zero width at the facing edges of the parts 21a and 21b. The material at such points would wear excessively, and so the arcs that define the recesses have been truncated to less than 180°. However, the arcuate surfaces that remain confine the needles 19a-19i sufficiently to prevent adjacent pairs of needles from rubbing against each other.

The size of each of the dots is determined approximately by the diameter of the needles, but the texture of the surface of the paper on which the dots are printed makes the shape and, to some extent, the size of the dots inexact. The contour of a dot as seen under a magnifying glass is likely to be rough, which makes it impossible to measure the diameter precisely. The size of each dot is also affected by the type, hardness, and thickness of the paper and the number of copies; the environmental conditions; the type of ribbon used and the ink on the ribbon; and other factors. These latter factors do not affect the dot size equally, and the most important factor is the diameter of the needle.

When the prior art structure shown in Fig. 4 is caused to scan across the paper 12 along a line perpen¬ dicular to the row of needles 19a-19i, as shown in Fig. 3, and two needles 19a and 19c spaced apart by the needle 19b are repeatedly actuated, two horizontal rows 37 and 38 of dots with a space 39 between them will be produced. The dots are approximately 0.35 mm In diameter, corre¬ sponding to the size of the needles, and their centers ar spaced apart 0.74 mm. To the naked eye, the dots in a ro tend to blend together, which means that each of the rows 37 and 38 appears to be a line about 0.35 mm wide, and the space between the two rows is about 0.39 mm The maximum height of a symbol printed by the 9-needle structure in Fig. 4 is, therefore, 0.37(N-l)mm + 0.36 mm, or about 2.94 mm but it contains, at most, five visibly separate rows of dots. If four additional needles were added so that seven rows could be printed, the maximum symbol height would be 5.6 mm, much greater than the size chosen by the Japanese government.

Fig. 5. illustrates a geometrical analysis of a modified printing structure that makes it possible for at least seven horizontal rows to be printed in a space having a total height of 3.7 mm. A structure having N needles 41A, 41B...41(N-l) . 41N is arranged so that the needles are tilted at an angle -θ- with respect to the horizontal. Two shaded lines 42 and 43 printed by the needles 41A and 41C, respectively, are indicated. These lines are illustrated as having constant width, but they actually consist of dots, just like the lines 37 and 38 i Fig. 3. All of the needles are of equal diameter D and the width of each line is defined as D, although the line width may actually be slightly different from the needle diameter. The width of the space 44 between the closest visibly separate lines 42 and 43 is defined as w.

The width W is less than the line width D, con¬ trary to the space 39 between the lines 37 and 38 in Fig. 3, but the space 44 is still plainly visible, and the visibility of this space is a most important factor in this invention. Fig. 5 shows a vertical overlap V between two adjacent needles 41(N-1) and 41N, and the same overlap exists between each adjacent pair. The center-to-center distance between each adjacent pair of needles, as exempli- fied by the needles 41A and 41B, is X, and therefore the center-to-center distance between the end needles 41A and 41N is (N-l)X.

The vertical distance between the centers of adjacent needles is equal to the radius R of each needle minus the vertical overlap V, that is, 2R-V. Since D = 2R, the vertical distance may be written as D-V. The total vertical distance between the centers of needles 41A and 41N is (N-1)(D-V) and the total height T of the tallest symbol that can be printed is:

T = (N-1)(D-V) + 2R (1)

Equation (1) may be rewritten as either

T = ND - V(N-l) (2) or

T = N(D-V) + V (3)

The angle -θ- is

X (4)

The width W is

W = D - 2V (5)

The width W of a space 44 as a percentage of the width D of a line 42 is

W = D - 2V D ^% D ⁽¹⁰⁰⁾ (6)

Several numerical illustrations show the ad¬ vantage of offsetting adjacent printing elements both horizontally, or more generically, in the printing direc¬ tion, and vertically, or more generically, in a direction at an angle (usually perpendicular to the printing direction. For example, with fourteen needles 41A-41N having a diameter of 0.35 mm, which is a diameter often used at the present time and for some time heretofore, and spaced apart center-to-center 0.37.mm., it is possible to print seven separate horizontal rows in a space having a maximum height of 3.7 mm by offsetting the adjacent needles at an absolute angle of -6- of approximatel 45°degrees. The angle-0- is referred to as an absolute angle because the row of needles may be tilted upwardly to the right, as shown in Fig. 5, or upwardly to the left.

The vertical overlap V between any two adjacent needles, for example the needles 41A and 41B, will be approximately 0.09 mm. The space W between- two rows of dots, or horizontal lines, 42 and 43 printed by any pair o most closely adjacent needles in a first group of first needles comprising alternate needles 41A, 41C,... is approximately 0.17 mm, which is about 47% of the width of any line, for example, the line 42. The spaces, such as the space 44, between each pair of spaced horizontal rows can be filled in by printing dots with appropriate printing elements of a second group consisting of the elements 41B, 41D,..., each of which is capable of printing a row of dots centered between the two rows on each side and overlapping those rows.

If any two dots are to be printed in a vertical line in spite of the fact that the needles are offset as shown in Fig. 5, a delay time must be maintained between • the printing of such vertically aligned dots. If the vertically aligned dots are to be printed by next-adjacent needles 41A and 41B, the delay must be related to X cos -θ, i.e. to the amount of horizontal offset, or displacement, between these dots. If it is the dots printed by the farthest apart needles 41A and 41N , the delay will be related to (N-l) X cos -θ.

The concept of offsetting the needles even makes it possible to provide nine separate horizontal rows in the vertical height of 3..7 mm with needles having a diameter of about 0.35 mm and spaced 0.37 mm apart, center-to-center. The angle -θ- must be decreased to approximately 32° and the overlap V will then be about

0.16 mm, which is about 45% of the width of a line. The minimum clear space between two visibly separate lines will be almost 11% of the line width D .

The diameter of the needles 41A-41N need not be

0.35 mm. However, if there are to be nine separate horizontal lines, and eighteen needles, the diameter of each needle must be less than about^" 0.39 mm, for at the diameter, V = — so that W = 0, which means that there 2 is no space between the lines 42 and 43 that should be visibly separate. The corresponding, limiting minimum value of fr is a little less than 29°.

The reason for selecting even numbers of needles 41A-41N, e.g. fourteen and eighteen, to produce seven or nine visibly separate lines of dots like the lines 42 and 43 is that many existing matrix printers use printing heads with seven or nine needles, and it is desirable to be able to utilize the mechanism of such existing machines with a minimum of revision. By simply tilting the existing seven needle or nine needle printing heads, or even just the jewel bearing, it is possible to use such heads to print approximately the upper or lower half of a symbol at a time. Alternatively, many symbols could be printed using just the lower set of dots. Instead of printing half of just one symbol, it may be, and usually is, desirable to print the upper or lower half of all of the symbols on a given line in one scanning sweep of the printing head across the entire line and then to feed the paper the required distance to make it possible to print the other half in a second sweep, preferable in the reverse direction. Alternatively, the head could be shifted instead of the paper. It is well- known practice to print alternate lines of full alpha- numeric symbols in opposite directions, and the memory and control circuits capable of carrying out such bidirec¬ tional printing can be applied to printing half-height symbols.

Using the same needles twice in such bidirec¬ tional printing is equivalent to doubling the number of needles, and doubling any number means that the effective number will be even, not odd. The paper may be fed just enough to cause the uppermost needle printing the top row of the bottom half of the symbols to fall exactly of the positions of the needle printing the bottom row of the top half of the symbols. The advantage of feeding the paper just that much less than half the maximum symbol height is that it causes the space W between the closest visibly separate lines of dots to be greater, e.g. W is approx¬ imately 0.06 mm for seventeen needles of 0.35 mm. di¬ ameter, each, but only about 0.04 for eighteen needles of the same size. Thus, the space 44 for a seventeen needle arrangement would be about 17% of the line width instead of about 11% for an eighteen-needle arrangement. The angle θ- would be about 34°

Although a printing arrangement in which the actual or effective number of needles is even is capable of printing only the same number of visibly separate lines as a printing arrangement having one less needle, the extra needle allows one pair of the lines to be spaced farther apart than the other pairs. Any pair can be selected for such additional spacing, which may be of some benefit in adding to the legibility of the symbol.

For a printing arrangement with a smaller number of needles, e.g. thirteen, the value of W for needles having diameters of 0.35 mm is approximately equal to 0.20 mm., which is about 56% of the width of a line. The angle -θ would be about 49°

If all of the needles are to be used, or at least available, for both halves of a symbol, the relative vertical shift between the paper and the printing head should be ^{NX Sln θ}. _{If t}h_e bottom row of the upper half 2 is to overlap, precisely, the top row of the lower half, the vertical shift shoul_d be ⁽N-l⁾ X sⁱn θ_^

2

Typically, although not absolutely necessarily needles such as the needle 19 in Fig. 2, are round cylin¬ ders. The perpendicular distance through the axis of a needle from one surface to the opposite surface is re- ferred to as the diameter, even if the cylinder does not have a circular cross section. If the diameter is made much less than about 0.35 mm, the needle tends to be too flexible so that it may bend in response to the impact of the armature 26. Another advantage of offsetting the needles in accordance with this invention is that it becomes reasonable to reduce the diameter of the forward part of the needle adjacent the end 23 between the bearing 21 and the ribbon 24. Fig. 6 shows a needle 46 so modified. Its main portion 47 has a uniform body diameter of about 0.35 mm, although the body diameter may be larger than that or even somewhat smaller. The forward cylindrical portion 48 has a lesser diameter, which may be 0.2 mm, or even smaller. This forward portion should be long enough to prevent the shoulder 49 from pressing the ribbon 24 (Fig. 2) against the paper 12, for example about 0.2 mm long. The diameter of the end 51 should not be so small that it will pierce the ribbon, and the portion 48 should be entirely forward of the bearing so that the shoulder 49 will never enter the bearing where it might rub on the bearing surface and cause excessive wear of that surface.

if needles having effective printing dot di¬ ameters smaller than their body diameters, as is true of the needle 46 in Fig. 6, are used for the needles 41A-41 in Fig. 5, the legibility of the printed symbols may be further enhanced. For example, fourteen such needles producing dot diameters of about 0.26 mm will have no overlap V, even though their body diameters are 0.35 mm and their axes are spaced 0.37 mm apart. For such an arrangment, the space W will be approximately equal to the dot diameter and the angle 0 will be about 45°.

If there are eighteen needles that have ef¬ fective printing diameters to produce dot diameters of abo 0.20 mm, there will also be no overlap V, and the angle -& will be about 36°.

The angle ϋ has been calculated as if the printing apparatus included a delay line or other delay means capable of being set to any desired delay time. If the apparatus has no delay means as such, the printing of a vertical line of dots may be obtained by selecting the timing of the pulses that actuate the needles so that the pulses actuating any two needles to produce dots in a vertical alignment occur in the proper time relationship to do so. This time relationship must take into account the scanning speed of the printing head and the vertical distance between the two dots. If there is a fixed timing system for actuating the needles, it may be nec¬ essary to tilt the row of needles farther, which is the same as reducing the absolute value of the angle 0. Then, the needles must be spaced farther apart so that hori- zontal rows of dots printed by adjacent pairs of needles will not overlap excessively.

Fig. 7A shows printing head 52 with two jewel bearings 53 and 54 mounted rigidly in a frame 56. The bearings may be divided into two parts similar to the parts 21a and 21b in Fig. 4, and they may be held in place by being clamped or otherwise fixedly held in the frame 56. The bearing 53 serves as a guide for nine needles 57a-57i, and the bearing 54 serves as a guide for an additional nine needles 58a-58i. The needles 57a-57i are actuated by solenoids 59a-59i in a support 61 rigidly attached to the frame 56. Similarly the needles 58a-58i are actuated by solenoids 62a-62i held in a support 63 that is also rigidly attached to the frame 56. The frame 56 is supported on rods such as the rod 16 in Fig. 1 to slide back and forth in the directions indicated by a double ended arrow 64.

The bearings 53 are placed in the frame so that rows of dots printed by successive actuations of the needles 57a and 58i will either overlap each other exactly or will overlap only to the same extent as the overlap, if any, between each other adjacent pair of needles. The bearings 53 and 54 are also spaced apart a precise hori¬ zontal distance to permit a single symbol to be printed with all eighteen of the needles 57a-57i and 58a-58i, just as if all of the neeedles were in a single bearing. This makes it unnecessary to print the upper half (approxi¬ mately) of each symbol on the first scan of the printing head and the lower half (approximately) of each character during a second sweep. All printing of each symbol is completed during a single sweep. Furthermore, the paper, similar to the paper 12 in Fig. 1, need not be fed less than a full line increment after each sweep. It need not be fed a small increment after alternate sweeps, and a slightly larger increment after the other alternate sweeps.

Fig. 7B shows another embodiment similar to tha in Fig. 7A. In Fig. 7B, only a front bearing support 66 similar to the frame 56 and two jewel bearings 67 and 68 similar to the bearings 53 and 54 are shown. The electro magnetic actuating means are not shown, but can be identi¬ cal to the means 59a-59i and 62a-62i in Fig. 7A. The bearings 67 and 68 have needles 69a-69i and 71a-71i, respectively, supported in parallel vertical rows in them. Unlike the needles 57a-57i in Fig. 7a, which are spaced as closely as possible, the needles 69a-69i in the bearing 67 in Fig. 7B and the needles 71a-71i in the bearing 68 are spaced apart by a sufficient distance to permit any two next-adjacent needles, for example, the needles 69a and

69b, to print visibly separate rows of dots. The needles 71a-71i are spaced the same distance apart as the needles 69a-69i but are offset vertically so that the level of the center of a row of dots printed by successive actuation of, for example, the needles 71a is midway between the levels of the centers of dots printed by the needles 69a and 69b. The dots printed by the other needles are spaced correspondingly. Of course, the horizontal spacing between the needles 69a-69i and 71a-71i requires delay means to effect printing of dots in vertical alignment from the two sets of needles. The bearings 67 and 68 may be spaced to accommodate any time delay, but preferably, the bearings 67 and 68, like the bearings 53 and 54 in Fig. 7A, are located as close together as possible, taking the rela¬ tively large cross-sectional areas of the actuating means 59a-59i and 62a-62i illustrated in Fig. 7A into account.

The specific numbers of printing elements and the dimensions set forth above are not to be considered as limitations on the scope of the invention. All are to be considered as merely illustrative of the invention.

BUR, ^;

Claims

Claims:

1. Printing apparatus comprising-: supporting means; a first group of first printing elements sup- ported by said means and each having a certain body diameter and a front end comprising a printing area of a certain size to print a dot having a predetermined dot diameter each time one of said printing elements is ac¬ tuated, said elements being arranged in a row, the center-to- center distance between adjacent ones of said elements being greater than twice said body diameter, said elements being oriented so that the centers of two rows of said dots printed by two adjacent ones of said printing ele¬ ments will be spaced apart a distance greater than said dot diameter and not substantially greater than twice said dot diameter; and a second group of second printing elements sup¬ ported by said means, each of said second elements being substantially identical with each of said first elements and oriented so that the centers of rows of dots printed by said second printing elements are substantially midway between rows of dots printed by adjacent ones of said first printing elements.

2. The printing apparatus of claim 1 in which all of said first elements are uniformly spaced apart.

3. The printing apparatus of claim 1 in which all of said second elements are uniformly spaced apart.

4. The printing apparatus of claim 1 in which said second printing elements are in the same row with said first printing elements.

5. The printing apparatus of claim 4 in which the center- to-center spacing between each of said second elements and the nearest element of said first elements is substan¬ tially uniform and is less than 10% greater than said body diameter.

6. The printing apparatus of claim 1 in which the centers of said rows of dots printed by each next-adjacent pair of said first elements are spaced apart by a distance at least 5% greater than said dot radius.

7. The printing apparatus of claim 1 in which the orienta¬ tion of each of said first elements is such that a row of dots printed thereby overlaps a row of dots printed by the next adjacent one of said second elements by a distance V, which is less than half the diameter of dots printed by the respective ones of said first and second elements.

8. The printing apparatus of claim 1 in which said row of first elements is at an angle -θ of between.32° and 49° with respect to said printing direction.

9. The printing apparatus of claim 4 in which the center of each of said second elements is offset with respect to the center of the next adjacent one of said first elements by an absolute angle 0 of between 29° and 49° measured from said printing direction.

10. The printing apparatus of claim 1 in which the diameter of the front end of each of said printing elements is smaller than said body diameter measured midway between said front end and the opposite end.

11. The printing apparatus of claim 1 in which each of said printing elements comprises a ballistically actuated needle having a uniform body diameter along most of its length and a substantially cylindrical portion of lesser diamter adjacent the front end thereof.

12. The printing apparatus of claim 11 in which the length of said substantially cylindrical portion of lesse diameter is less than four times as great as said body diameter.

13. The printing apparatus of claim 11 in which said bod diameter is approximately 0.35 mm and said diameter of said front end is approximately 0.20 mm.

14. Printing apparatus comprising: a printing head; a plurality of printing elements supported in said head, each of said elements having an end with a maximum effective printing cross-sectional diameter that is small in comparison with the length of the respective element; means for supporting a record medium in front o said head; means for actuating each of said elements individually to effect printing of an elemental area dot on said record medium; means for providing relative lateral movement i a straight line between said head and said record medium, whereby combined repetitive actuation of any one of said elements and said relative lateral movement causes a substantially straight row of said dots to be printed on said record medium, said end of each of said printing elements being located relative to said end of the next adjacent one of said elements such that the maximum effective printing cross-sectional diameters of adjacent ones of said elements are greater than the distance between the centers of rows of dots printed by said adjacent elements.

"B^U

O

15. Printing apparatus according to claim 14 in which a straight line between corresponding points of dots printed simultaneously by said next adjacent elements makes an angle -β having an absolute value of between about 29° and 49.

16. The printing apparatus of claim 15 in which the angle -0- is equal to sin^" D-V , where D is the diameter

X of each of said dots, V is the overlap of two dots printed in a line perpendicular to said straight line by adjacent ones of said printing elements, and X is the distance between the centers of said adjacent printing elements.

17. Printing apparatus according to claim 14 in which the distance between centers of two rows printed by adjacent ones of said printing elements are spaced apart by a distance D-V, where D is the diameter of each of said dots and V is the overlap of said two rows.

18. The method of printing a symbol that consists of a plurality of dots at predetermined points in a matrix pattern of N such points in each row in a first direction and a plurality of said rows in a second direction, said dots being printed by means of printing elements arranged in diagonal row that makes an angle -Θ- with said second direction, each of said dots having a diamter D that is greater than the spacing between each adjacent pair of said points in one row in said first direction but less than approximately twice such spacing, said method com¬ prising printing said dots in said one row at different times determined by the cosine of 0 times X,the center-to- center spacing between the printing elements printing said dots.

^"BU EΛ

OMPI

19. The method of claim 18 comprising printing any dots required for said symbol in approximately half of such points in the first direction on a recording medium by means of N/2 of said printing elements as said N/2 printing elements scan approximately half the area of sai symbol; shifting said N/2 printing elements in said one direction by a predetermined distance relative to said recording medium; and scanning the remainder of said area by said N/2 printing elements and simultaneously printing any of said dots required for said symbol in said remainder of said area.

20. The method of claim 19 in which said shifting is substantially equal to — N X sin θ.

2

21. The method of claim 19 in which said shifting is substantially equal to (N-1 ) X sin -θ.

2

22. Printing apparatus comprising: supporting means; a first group of first uniformly spaced printing elements supported by said means and each having a certain body diameter and a front end comprising a printing area of a certain size to print a dot having a predetermined dot diameter each time one of said printing elements is actuated, said elements being arranged in a first row, the center-to-center distance between adjacent ones of said elements being greater than said body diameter, said elements being oriented so that the centers of two rows of said dots printed by two adjacent ones of said printing elements will be spaced apart a distance greater than half said dot diameter and not substantially greater than twice said dot diameter;

O ^* a first group of actuating means equal in number to said printing elements but much largervin cross-sectional area than said body diameter, said actuating means being rigidly affixed to said supporting means; a second group of second printing elements supported by said means, each of said second elements being substantially identical with each of said first elements and arranged in a second row offset from and parallel to said first row of said first elements and uniformly spaced apart by the same distance as said first elements and oriented so that rows of dots printed by said second printing elements are parallel to rows of dots printed by said first printing elements; and a second group of actuating means equal in number to said printing elements but much larger in cross-sectional area than said body diameter, said second group of actuating means being rigidly affixed to said supporting means.

23. The printing apparatus of claim 22 in which said first row of first printing elements and said second row of second printing elements are perpendicular to the rows of dots printed thereby.

24. The printing apparatus of claim 23 in which said first printing elements are spaced apart by a distance sufficient to produce two visibly separate dots when any two next-adjacent ones of said first printing elements are actuated, each of said second printing elements being located to print dots centered at a level midway between the centers of levels of dots printed by one pair of said next-adjacent first printing elements.

OMPI

25. The printing apparatus of claim 22 in which said first row of first printing elements and said second row of second printing elements are angularly disposed at an angle of between about 32" and 49° with respect to the rows of dots printed thereby.

26. The printing apparatus of claim 25 in which said second group of printing elements is offset horizontally and vertically with respect to said first group of print- ing elements.