US3439760A - Automatic printing price scale with photoelectric encoder including range and motion detectors - Google Patents

Description

April 22,1969

K. c. ALLEN 3,439,760

AUTOMATIC PRINTING PRICE SCALE WITH PHOTOELECTRIC ENCODER INCLUDING RANGE AND MOTION DETECTORS sheet of 7 FOURTEEN COLUMNS INVENTOR KENNETH c. ALLEN BIN'ARY DECIMAL CONVERTER COMPUTER f- ATToRNEYa April 22, 1969 K. c. ALLEN 3,439,760

' AUTOMATIC PRINTING PRICE SCALE WITH PHOTOELECTRIC v ENCODER INCLUDING RANGE AND MOTION DETECTORS ,2

Fil'ed Oct. l5. 1963 SheekI Of '7 FIG-4 INVENTOR KENNETH C. ALLEN ATTORN EYS 22, K. C. ALLEN AUTOMATIC PRINTING PRICE SCALE WITH PHOTOELECTRIC ENCODER INCLUDING RANGE ANDvMOTION DETECTORSV 1965 sheet of7 Filed Oct. L5

R. N m 9 m .A r m m. M uw W www F, a, L" 1 w. m N mw mw v W a C; y@ s@ c MM: mw Tlfllgl. Tv o om Tlf yr T Wo u, Tm vN f m .0.o w om om Q QE m N f 0.,. x j?. Y u @w w# i nu mw LTS om?. 1 m A w OW TG V .ll A v .QQ m da... 99.59F wm Apr1l22, 1969 K. c. ALLEN 3,439,760

` AUTOMATIC PRINTING PRICE SCALE WITH PHOTOELECTRIC ENCODER INCLUDING RANGE AND MOTION DETECTOR'S l5, 1963 sheet 4 of 7 Filed Oct.

0.0 w m L L A. U C ,A H T M E l N N E K v. B

K. c. ALLEN 3,439,760

RC ON DETECTORS April '22, 1969 AUTOMATIC PRINTING PRICE SCALE WITH PHOTOELECT ENCODER INCLUDING RANGE AND MOTI 1965 Sheet Filed Oct. l5,

all lil .Il

INVENTOR.

KENNETH C. ALLEN www? ATTORNEYS Aprxl 22, 1969 K. c. ALLEN 3,439,760

- AUTOMATIC PRINTING PRICE SCALE WITH PHOTOELECTRIC l ENCODER INCLUDING RANGE AND MOTION DETECTORS Flled oct. 15, 1965 sheet 6 of 7 INVENTOR.

KEN NETH C. ALLEN ATTORNEYS 8f n @i April 22, 1969 K. c. ALLEN 3,439,760

AUTOMATIC PRINTING PRICE SCALE WITH PHOTOELECTRIC ENCODER INCLUDING RANGE AND MOTION DETECTORS Filed oct. 1,5, 1965 sheet 7 of 7 F IG-IO 309 C l l 232 L .P-s P4 Q-e 13 308 F J309 w 7T /50 P v Q C bala [-263 l "f I COMPUTER START L SIGNAL *fo-1 *lsf M- -1 7 #f2 M|| 3l6 J o-:s ,n-5 A-l ;P5 N'Z i 309 347 l (3-3# N" R-e Ml- Tm .Im-5 C I FIG-11 INDEX OF ELECTRICAL SYMBOLS I A *A RELAYA NORMALLY OPEN NORMALLY CLOSED CONTACTS OF CONTACTS OF RELAYA RELAYA -\MMM -N- L|GHT RESISTOR DIODE CONDENSER DIODE Y KENNETH C. ALLEN www@ ATTORNEYS t weighing and computing scale systems.

Int. rCl. G01g 23/28 U.S. Cl. 177--3 13 Claims ABSTRACT oF THE DISCLOSURE An automatic weighing; scale comprisingphotoelectric digital encoding means and computing means with printout means for computing the price of a commodity and issuing a printed ticket. The price is based upon the weight of the commodity multiplied by the price perfunit of weight. The digital encodingfmeans comprises a chart bearing a pattern of translucent and opaque binary markings. Weight magnitudes are converted intobinary electrical signals by photoelectric transducers clustered inthe path of light projectedy throughthe chart and through .a slitted mask. vCertainof` Said same photoelectric transducers are also utilized inr an out-of-range control system and ina motion detection system,rb`oth of which systems act to prevent the automatic operation of the Overall i system unless proper conditions obtain. e

This application'is a continuation-in-part of my application Ser." No. 220,765, ltiled Aug;` 31,' 1962. l

This invention relates to weighing scales for automatic The invention more particularly relates to a weighing scale including a weight readout system whichprovides an yelectrical digitalsignal-corresponding to ,the weight of an article placed on a platten'The weight signalwhich is formed is such that it may then be combined with the cost perunit weight of the commodity weighed in a-computer to'fproduce a valu'ecf the article,*andfa ticket printed and issued'of such values.`

United States Patent 3,439,760 Patented Apr. 22, 1969 ten, and from this a binary-cyclic-biquinary code has been derived. A principal feature .of this code is that only one bit of information changes with any incremental change in Weight. The readout system ot'this invention provides an output which is compatible with the computer described and claimed in applicants United States Patent No. 3,045,229, assigned to the same assignee as this invention. i

The mechanical portion of the weighing'mechanisml of the scale may be constructed according to the projecting scale shown in the United States patent to Meeker et al., 2,723,113, assigned to the same assignee as this application, wherein a chart is supported for movement according to the weight on the platter. In this invention a lensv systemprojects an enlarged image of the encoded chart onto a mask positioned adjacent the transducers. Apparatus may also be employed to project an image of the weight in arabic numerals for viewing by the operator.

The digital readout portion of this Vinvention includes switches or relays which are operated according to the illumination of selected ones of the transducers.` The `The scale of this invention provides rapid analogue to digital conversion and is read outby the employment of photoelectric transducers which are arranged-to respond to a projected portion of a movable chart carried lby the scale'balance mechanism, The weight represented by ditferentchart positions is :encoded on the chart in a pattern of translucent and opaque binary markingswhich control the energization of the transducers."

kOne of the principall problems'in reading out any continuous encodedfindicating device concerns the necessity for making distinctions'between adjacent numbers, such as between 19.99 and 20.00. As the scale moves from the lower number to the higher number, or vice versa, four digits in the above example must change simultaneously. lf the device responding to the encodedl chart responds to less than foury ofthe digit changes in this example, an error results.

'This difficulty is eliminated yin this invention by the employment on the chart of a pattern formed in a cyclic binary code which represents'a range of weight. This code is read along paths or columns of binary markings by transducers after the platter has come to rest with a weight thereon falling within the weight range of the scale. The binary code ofthe chart. is preferably cyclic in form representing decimal equivalents. The code of this invention is a specialderivation of Grays reected code.

' However, the cycle variation is based on iive'rather than transducers are preferably photocells which undergo a decrease in resistance upon the incidence 0f light. The -circuitsfurther include a motion ldetector for signaling when the scale has come to rest with a load on the platter. The balance position of the chart is read and stored in the readout circuits with such speed that there is no need to lock the scale at balance. The transducers are made insensitive to any further movement of the scale following the readout of the chart'to prevent the possible input of erroneous information.

Provision is also included for preventing automatic computer operation when there is less than a predetermined minimum weight on the scale platter. Another circuit senses a balance position of the scale and 4chart either above or below a i given ran'ge of weights for the system and prevents an automatic start under such conditions. A nonrepeat circuit assures that only one weighing and readoutcycle occurs for each separate weighing operation. t Y

The readout circuits of this invention provide direct conversion from the binary-cyclic-biquinary code into the corresponding decimal equivalent representing the Weight of an article placed on the scale platter. This results in the simplification and increased reliability of the weight readout circuit. y

It is therefore a principal .object of this invention to provide a Weighing scale as outlined above incorporating apparatus for the digital readout into decimal equivalents of a weight represented by a binary cyclic code. A further important object of this invention is to provide a weighing scale as outlined above having a cyclic digital system with direct decimal readout employment in a value computer.

Another object of this invention is the provision of a weighing scale as outlined above characterized by a minimum of moving mechanical components and the elimination of the motors, scanners,'mechanical readout switches, the platter switch, and the like. y

A still further object ofthis invention is the provision of an electrical digital readout weighing scale characterized by high speed. A i

A still further object of this invention is the provision of an automatic weighing scale as outlined above which reads out weight information ,sufficiently fast so that there is no need for any separate provision for holding or locking the scale to prevent error due to motion during readout.

' Another object of this invention is the provision of a load sensing 'circuit-A operating by means of a signal `from atente@r the binarycode Aonda chartto sense when more than a` minimum weight is on the scale platter and prevent automatic operation withless than such minimum weight.

A further object of lthis invention is the provision of an error interlocking operating through the code on the chart to prevent automatic operation when the scale is at balance below z'ero or above a predetermined weight. Another object of this invention is the provision of an automatic weighing scale utilizing a binary-cyclic-biquinary encoded` counting or weight indicating and translating system.

o Further objects and advantages of the invention will be apparent from the following description, the accompanying drawings and the appended claims.

In the d'rawingsf t `f FIG. 1 isa perspective view of ya computing Vscale systern which mayhave a scale made according to this invention; "`FIG. 2 is a diagrammatic view of a portion of the weighing scale of the invention;

FIG.`3 is an elevational view of and mask;`

,e FIG.I4,is a fragmentary view of a portion of the binary code of `this invention;

FIGS. 5 and 6 together form the photocell and relay circuits for each ofthe rows or columns of binary markings onfthe chart; i FIG. 7 is a circuit diagram showing the ladder arrangement of the relay contacts for determining the fevenness" or oddness of a translated true number and for operating a` complements indicating relay accordingly; l

FIGS. 8 and 9 are wiring diagrams of the output contacts of the read out relays and complements indicating relays providing direct decimal indications of weight;

FIG. 10 is a wiring diagramV of some of the control circuits of this invention; and t e FIG. 1l illustrates the electrical symbols employed in FIGS. 7-10.

t Referring to the figures of the drawings which illustrate a preferred embodiment of the invention, an automatic computing scale system is shown in FIG. l as including a weighing scale having a scale platform 21. The *scale 20 is shown as including a window 22 for reading the weight, and the usual tare adjustment 24.

The computer is shown as mounted adjacent the scale 20 .and receives the weight information from the scale for combining with a price per unit of weight to compute the value of a weighed article. The computer is preferably constructed according to the teachings of the United States patent to Allen, No. 3,045,229 of July 17, 1962, which will be referred to herein asthe Allen computer patent. A series of price input knobs 26, 27 and 28 are mounted on a printer 30 and provide the means by which the unit price may be set into the computer 25. The printer 30 may be constructed according tothe teachings of the United States patents to Allen, No. 2,948,465, and t0 Allen et al., No. 2,948,466, and operates to print and issue a label showing the unit price, weight and computed value.

The mechanical weighing mechanism ofthe scale 20 may be constructed according to the teachings of United States patent to Meeker et al., 2,723,113, assigned to the same assignee as this application. Referring to the diagram of FIG. 2, it will be seen that the scale 20 includes the usual lever connected for movement with the platter 21. An optical chart 36 is supported for unidirectional movement in a vertical plane by the lever 35 according tothe balance position of the scale platform 21. `Weight information, i.e., arange of weights, is encoded onthe chart 36 into` closely spaced rows 37 of binary marks shown with exaggerated spacing in FIG. 2. and comprising patterns of alternating opaque and transparent areas. As described hereinafter, there are fourteen rows or columns 37 employed in the illustrated embodiment of the the photocell block 4 invention, and the chartv36 forms a part of an optical projection system shown as including a tixediprojection lamp 40 and a lens 41 to concentrate thek light of the t lamp filament on the chart 36. A projection lens 42 projects an enlarged image 37 of a small vertical extent of the chart rows 37 onto the surface of a photocell mask 45, and in each balance position of the scale, the transparentand opaque portions of the columns 37 form a unique pattern indicative of the corresponding particular weight.

The invention includes photoelectric means for reading out the portion of a code on the chart 36 corresponding to the weight on the platter, consisting of a plurality of photocells Sti, one for each of the rows 37. Each photocell 50 is positioned immediately behind a slit 55 in the mask 4S. However, for the purpose of illustration, the photocells 50 are shown in FIG. 2 as being spaced away from the mask. The photocells 50 undergo a decrease in resistance with light falling on the photocell window, to operate as current gates or valves. A particularly useful photocell for this purpose is designated type CL I604, manufactured by Clairex Corporation, 19 W. Twenty-Sixth St'., New York 1G, NY.

The photocells 50 are retained in a cell block 52, as shown in FIG. 3. There are provided as many of the cells 50 as there are rows of binary information on the chart 36, and in this embodiment, fourteen photocells are employed. In order to conserve space within the optical system, and to position the cells 50 as closely to the center of the optical axis as practicable, the photocells are arranged in clustered relation as shown, with five in each of two rows and four in the third row, with each of the cells 50 being spaced laterally from the adjacent cells and from the cells in the adjacent rows.

The mask 45 is secured to the front face of the cell block 52 and includes image-defining openings or slits 55, one for each of the photocells 50. The slits 55 may be approximately ten thousandths of `an inch wide and are accurately positioned in relation to the rows 37 of the projected image of the chart 36, as indicated by the representation of the projected image 37' of the columns 37 in FIG. 2 and as also shown in FIG. 3. Accordingly, the relative positionsof each of the rows on the chart 36 are staggered vertically in order to conform to the position of the slits 5S on the mask 45 Preferably, to provide tolerance in the physical alignment in positioning of the photocells, the effective width of the slits is smaller than the smallest division of the projected image of the code formed on the chart 36 and accurately aligned with respect to the projected image. The slits 55 obviate the necessity for accurate positioning of the photocells with respect to the projected pattern, and since each slit is smallerin effective area than the smallest division of the projected pattern, it restricts the chart image which is projected onto `its associated photocell to a small and clearly defined portion of the projected image.

The cell block 52 may be provided with heater means for maintaining a constant temperature of the photocells 50 in order to eliminate the effects of drift due to variations` in temperature. For this purpose, a low wattage heater 56 may be attached to one surface of the cell block 52 for the purpose of maintaining the photocells 50 at such a constant temperature.

The output of the photocells 50 is applied to circuits for converting the binary coded information into its decimal equivalents, as represented by the block 58 in FIG. 2, and as shown in FIGS. 5-10. The binary equivalent of the Weight to the closest one hundredth 0f a pound is then applied to the computer 25.

The cod e wherein one can count to infinity by changing only one bit of information for each numerical change in thegcommon decimal system. An example of this system is shown in Table I, wherein there is acolumn of decimal numbers on the left anda column of cyclic decimal numbers on the right.- It willk benoted from an examination of Table I that as one moves from,9 to 10 in the decimal system, the cyclic decimal changesl from 9to` 19 and countsl downwardly tothe cyclic decimal 10, which, in turncorrespo'nds to 19 in thedecimal system. In changingv from 19 to 20 in the decimal system, both digits must be switched. But, inthej corresponding change in the cyclicf'system, the digits change from 10 to 20 and only one digitis switched, that ofy a l to `a 2 in the second column. Obviously, such a, counting system can be extendedindefinitely. y v

.Therulefor translating cyclicdecimals into natural decimalsucan be stated as follows: Examining the cyclic f decimal columns, the kfarthest left number is always a correct decimal number. If this number is even, the succeeding number to the right is also a correct decimal number.I However, if the farthest left number is.'` odd, thenthe 9s complement of the succeeding number must be used. The correct meaning ofthe 'third `digit from the left'is determinedin the same manner, depending upon whether or not vthe translated true decimal number of the vprevious digit is odd or even. Thus, the` use of the natural number or its 9s, complement is dependent upon the oddness or` the evenness ofthe translated true numberthat appears to its left.

The cyclic biquinary numbering system shown in Table II, hasbeen derived from the above system. Here, a pair ofv digitsv are employed to represent a single decimal numberwherein the digits to 4 are yused inthe lower order,\of the pair and a 0 o r a 1 inthe higher `order of the pair. The occurrence of a 0 or 1 in the higher order of the pair directs whether the lower order represents a true numberv or whether it represents the 9s complement.

.The significance of the O :or the 1 inthe Ahigher order of the pair of numbers is dependentupon whether the translated true number of the next higher order pair of digits is odd or even. If the next higher decimal number is even, ythe 0 .directstheuse of the significant number and the 1 directs the useof 'its 9s complement-The meaning of() or 1 isreversed if the higher order translated true number is odd. For example, in Table II, the cyclic biquinary number 13 dictates that the true decimal number is6, the 9s ,complement of 3. VIn this example, the higher order decimal is an even `(lso thaty the 1 inthe second columndirects the use of the 9s complement of 3'. By"'contrast, lthe kcyclic biquinary number 01 13 calls first for a true decimal 1. in the higher order. This being odd, the significance of the 1 in the lower order pair is reversed and the 'true decimal number is therefore 13. Table II shows `additional examples of this system of counting.

Table III represents the cyclic biquinary code of Table II in binary form. In Table III, four columns or rows of binary indications are employed to indicate a single decimal column. The first three rows of each grouping of four rows represent whole decimal numbers of 0 through 4, and the fourth' Vrow indicates whether or not the 9s complement of the number represented by the first three rows is intended.

The arrangement of binary indications representing 0 through 4 may take kany one of several forms, but once this is determined, it is maintained uniformly throughout the counting system. In Table III, the numeral 1 -is employed to indicate the occurrence or presence of a` binary indication or digit, and the 0 'is used to indicate the abvsence ofabinary indication. Preferably, some form of indication is employed to represent a true zero rather than the lackof indication as is commonly employed in binary systems. Therefore, 0001 represents the whole decimal number of 0in each of the groups of rows or columns.

A further examinationy of Table III shows that the reflective repetition of the `binary indications in the first three rows occurs in live digit increments. In other words, one counts from O to 4 in the first three columns, and then a 1 indication is added to the fourth or complements indicating column while the binary lindications of 5 through 9 representa reflection or reversal ofthe indications of 0 through 4.

Y The preferred embodiment of this invention lconsists of an automatic weighing scale which is'partcularly adapted for use as the automatic weighing scale'in a computing scale system such as described and claimed in the above Allen computer patent. In this patent there is described a. scale system having a maximum weighing capacity of 24.99 pounds. Thus, the scale described herein uses three groups of four columns of binary markings which represent .respectively the hundredths of pounds, tenths of pounds, and pounds of Weight on the scale platter. Only two binary rows are required to represent the tens of pounds since this will either be 0, l or 2 for a scale of the given capacity of 25.00 pounds. In Table III, fourteen columns of binary indications are employed to count to 25.00 (which are, in fact, sufficient to count to 29.99)

by changing only one binary indication for each successive change in digital value of the weight.

The A through N indicated in Table II above the several columns designate these columns and also designate relays A through N which are individually operated in accordance with the' binary indication'at any particular balance position within the columns. An examination of Table III indicates that particular groups of the relays A through N will be operated for each decimal position. The tabulation shown in Table IV illustrates the operation of the relays A, B, C and the complements indicator D in the column through the digits0-9. 1

TABLET` Decimal: Cyclic decimal 0 0 1 1 2 2 3 3 4 4 5 5 `6 6 7 7 8 8 9 9 10 19 11 18 l2' 17 13 16 14 15 15 14 16 13 17 l2 18 l1 19 10 20 20 TABLE II Decmal: Cyclic biquinary 0 OO 00 l 00 0l 2 00 02 3 00 03 4 00 04 5 00 14 6 00 13 7 00 12 8 00 1l 9 00 lO 10 0l l0 1l O1 1l l2 01.12 13 01 13 14 01 14 TABLE IIContinued Cyclic biquinary TABLE III.-BINARY-CYCLlC-BIQINARY "NM LKJI HGFE DCBA 01 0001 0001 0001 01 0001 0001 0011 01 000]. 0001 0010 01 0001. 0001 01 1 0 01 0001 0001 0100 01 0001 0001 1100 01 0001 0001 1110 The above-described cyclic biquinary code is applied in binary form to the chart 36 in fourteen vertically arranged columns A through N. A fragment of columns A throughV D on chart 36 is illustrated in FIG. 4, beginning at zero pounds and extending through twenty-eight hundreths of a pound (0.2.8 pound in one hundredths of a pound increments. The dark binary indications of the code as shown in FIG. 4 are actually formed on the chart as clear or transparent areas wherein light is per- Vmitted to fall on the photocells thus effectingV a decrease vin the resistance, signalling the coincidencev of the associated slit 5S with the projected image of the chart at one of the 1" binary indications. The 0 binary bits comprise the opaque areas on the chart 36.

The physical arrangement of the rows A through N of binary indications may be varied on the chart 36Iwithin wide limits, as long as the associated photocell Si) and slit 55 reading the particular column are correspondingly located to receive the projected image of the particular column which it is to scan. Thus, the columns A-N lmay be formed concentrical-ly on a disk, or wrapped about a drum. Preferably, they are arranged linearly side-by-side in a plane on a` photographic pla-te.,The binary indications of the columns are stagged vertically one in relation to another to correspond to the physical placement or grouping of the Ycorresponding cells 50 withinv the block 52, but forconvenienceof understanding the principlesof the invention, they are represented in FIG, 4 asif the fourteen photocells were all arranged in alphabetical order and in -a single horizontal line. y y p Referring ,to FIG. 4, they A column comprises dark areas each having a vertical extent of four `units and spaced six units apart, and since as already noted, these dark areas represent transparent portions of the chart, it follows that light will be transmitted to the A photocell only when the digit in the hundredths of a pound column is 8, 9, 0 or l.

The C column is identical with the A column except that it is displaced with respect thereto so that light falls on the C photocell only when the digit in the hundredths co1- umn'is 3, 4, 5 or 6. The 'B column contains sections of dark areas each three units in vertical extent and spaced two units apart, while the D column comprises alternating light and dark units each of a vertical extent often units. p The fourcolumns E, F, G and H are similartorthe columns A, B, C and D respectively except that thelight and dark areas therein are each ten times the vertical extent ot the corresponding column for the hundredths of a pound digits because they yrepresent the next higher decimal order. It will also be noted that the E and G columns include dark areas extending below zero and above the indicatedvcapacity of 25.00 pounds, so that the E and 'G photocells both receive light whenever the balance position of the scale is below zero or above 25 pounds, for a reason pointed out hereinafter. It will be apparent th-aththe I, J, K and L columns are also similar to the columns A, B, C and D, respectively, but have the light and dark area therein each one hundred times the vertical extent of the corresponding A, B, C or D column. The M and N columns correspond in relative arrangement to the A and B columns, but each area therein is one thousand times the vertical extent of the corresponding area in the A or B column, with the exception that the dark area of the M column is removed below .1l pound -for a reason pointed out hereinafter.

Cha-rt reading and converting circuits In. the following description, reference may be had first to FIGS. 5 and 6 of the electrical wiring diagram wherein respective photocells 50'associat`ed with the code columns A through N are shown in conjunction with their associated relays A through N. In the following description, the photocells 50 are designated individually as Fa, Fb, Fc, et cetera, according to the binary column A, B' or C with which they are associated. Similarly, the relays controlled by the photocells are correspondingly designated as relays A, B and C, et cetera. a

Each of the photocells 50 is connected to a common power lead which has applied thereto, through either normally open contacts P-1 or normally closed contacts 1 of control relays P or Q,'a` negative full wave recticd and unfiltered DC voltage which may be inthe order of 300 volts. The photocells are each similarly connected to a common return lead 81 through load resistorsZ. Accordingly, 4when one of the 'photocells' Fa-Fn, at the balance position of the weighing mechanism, is coincident with a projected binary marking on the chart 36, the irnpedence of the photocell may drop in the order of 5:1, such as from .5 megohm to .1 megohm, thereby developing a lmore negative voltage at the junction 84 between the photocell 50 and the load resistor 82.. Relay operating means for the relays A-N mayinclude thyratron tubes which are grid controlled by the photocells, but preferably include four-layer diodes 85 which are identified by the identifications Va-Vn, in conformance with the above-established notations. The four-layer diodes may be of the Shockley type 4E-200 which have the characteristic of thyratrons in that a negative pulse at the base of the diode will set the diode into a state of conduction, and it will continue to conduct until the voltage isA removed from the anode. u i It will be seen from an examination of FIGS. 5 and 6 that the four-layer diodes 85- are connected in series with the associated relays A-N between a lead 86 andthe common lead.81. A full wave unfiltered DC voltage for operating the relays A-N is applied to the lead 86, andmay be in the order of 24 volts. It willalso be noted that a conventional di'ode 88 extends between the, junction ofthe cathode of the four-layer diodes 85 and the load resistor 82 to ground, for each of the circuits associated with the photocells F, through Fn. The-diodes 88V provide a low im- 4pedence conducting path for the four-layer diodes 85 while blocking the dissipation ofthe load signal effected 9. by the photocells 50. Additionally, each of the relays A throughN is provided with a resistor S9 connected effectively in parallel with the relay coil kbetween the line 86 carrying the power for the relays, and the junction of each relay and its diode 85. The purpose of the resistors 89 is to provide an instantaneous current for the four-layer diodes Va n in order to hold the diodes conducting while the current is building in the associated relay coils.

The circuit components described above apply to each of the circuits for reading out thecolumns A-N. In addition, the circuits associated with the photocells Fa, ,Fc and Fm, Fn are provided with additional components since these photocells are also used as motion detectors and weight presence indicators, and their functions will be described in greater `detail below.

The contacts of the relays A-N are combined rwith those ofthree complements indicating relays X, Y, and Z as shown in FIGS. 7 8 and 9 to provide direct binary to decimal readout cir-cuit means for the binary code ou the chart 36. Referring to FIG. 8, it will be seen that the contacts of relays A, B and Care arranged' in a matrix to apply a ground signal selectively to the digits -.4 in the one-hundredths of a pound leads indicated generally at 100. The 9s complementsare applied through normally open contacts of the complement relay X.

Referring to FIG. Sand comparing the circuit with the relay operating sequence shown in Table` IV, if a zero in the one-hundredths of a pound column is indicated, then only relay A is operated bykphotocell Fa, thus closing a line through a pair of normally closed contacts of the relay B to the zero lead. If thernumeral "1 is indicated in this column, theny both relays A and B are operated according to the code, resulting in the application of ground to the number 1 lead, through front contacts of relays A and B. The circuit may easily be traced for the numerals 2, 3 and 4 according to the schedule of operation for the relays as shown in Table IV.

For the numerals 5, 6, 7, 8 and 9, the complements relay X is operated by the circuit shown in FIG. 7. Thus, in order to indicate a 5 in the one-hundredths of a pound column (C-I-D), a ground is applied to the number 5 terminal through aclosed contact of the complements indicator relay X and a closed contact of relay C. The circuits may easily be traced for the numerals 6, 7, 8 and 9 as outlined in Table IV'.

The complements indicator relays X, Y and Z, respectively, representing the complements of the hundedths of a pound, tenths of pound and pounds, are operated by a signal applied through the ladder circuit of FIG. 7. This circuit electrically determines the evenness or oddness of the preceding translated binary number and determines whether the significant number or the 9s complement of the next lower order number is tobe used.

The ladder circuit includes yboth normally open and normally closed contacts of the relays D through N. The arrangement of contacts is not unlike that of an electric circuit where a light is controlled by a plurality of switches. The full complement of M and N relay contacts is not needed since, in this example, the maximum tens of pounds column digit is limited at two. Only a pair of contacts of the D complement indicator relay is needed, since it will control only the X complement relay for the hundredths of pound column. Also, no contacts of the A, Bor C relays arev needed, since these represent the lowest order and cannot have any effect upon any further complements. Otherwise, for each of the relays E through L, two normally open relay contacts are cross connected by two normally closedcontacts of the same relay so that a ground signal will be applied on one side of the ladder or carried across to the opposite side depending upon Whether any particular relay D through N is operated or 1a'The complements relay Z for the pounds column is connected with the ladder for operation just 4below the indicator relay `L for the pounds column. The complements relay Y for the tenths of pounds column is connected just below the H indicator relay in the tenths of pound column. rThe complements relay X for the hundredths of pounds column is connected at the bottom of the ladder just below the indicator relay D.

An examination of the circuit of FIG. 7 will reveal that the complements indicating relay Z for the pounds column will be operated or not operated depending upon whether relays for the tens of pounds columns N and M are operating, and whether the indicator relay L for the pounds column is operated. The connection of these relay contactsis such that any one of them can direct whether the relay Z is operated or not. In further explanation, if the highest order number lOs of pounds) is an even number, then it is either a zero or a 2, the output or the ground is applied by the ladder at position in FIG. 7, through the normally closed contact N. Now, if relay L is not energized, the next lower order number is a true one and the complements indicator relay Z would not be energized.

If on the other hand, the highest order number is an odd number, then a ground is applied at position 106 on the ladder. The higher order number being an odd number indicates that the 9s complement of the lower order number must be used, `and the complements relay Z is operated. lf relay L is operated and the highest order num- Iber is an even number, relay Z is operated. But, if relay L is operated and the highest order number is an odd number then, according to the rule for translating the cyclic code, the two odds make an even and the complements relay ZZ will not be operated since the number indicated by the relay I, l and K would then .be a true number.

The same description applies to the operation of the complements indicating relays Y and X. In this manner, the evenness or oddness of the preceding higher order number is determined, and it directs whether the 9s complement below is used depending upon whether the indicator relay (D, H, or L, depending on the column) for the particular column is operated or not operated.

FIG. 9 shows the additional contact connections for indicating the tenths of a pound, pounds, and tens of pounds of weight and for application of a weighting scale computer. The circuits of FIGS. 8 and 9 are particularly adapted for use with the computer of the Allen computer patent. In this patent, the computer includes a programmer for selecting, in sequence, the hundredths of a pound, tenths of a pound, pounds and tens of pounds stored weight information. The outputs 100 in FIG. 8 correspond to, and be electrically connected to, the contacts 185 in FIG. 16 of the Allen computer patent. The wiper 190 of that patent would not be used, but the ground common lead 101 of the hundredths of a pound contacts corresponds to that wiper. The combined output leads of FIG. 9, which are sequentially applied to the tenths of pounds, pounds and tens of pounds circuits, correspond to, and may be electrically connected to, the fixed contacts 265 in FIG. 17 of the Allen computer patent.

A relay 316 in the Allen computer patent is operated to apply the information stored in the matrix corresponding to the closed contacts of the relays A, B, C or X to the Allen computer patents weight storage` relay 3110-315 for hundredths of a pound. When one of these relays is operated, indicating that this weight information has been received, the computer relay 316 is released and, in doing so, applies power to the relays S20-325 of the Allen computer patent. lUp to this time during the computing cycle, relay 326 has been operated, and thus a ground is supplied through contact 326A in lead 121 of the tenths of a pound matrix (FIG. 9). The weight information thus stored in the matrix comprising the closed contacts of relays E, F, G and Y is then applied to the computer and stored by the tenth of a pound storage relays S20-325 in the Allen computer patent.

Similarly, when one of these tenth of a pound storage relays has operated, signaling that the `weight information has 4been received, the holding circuit to the relay 3'26`is broken, and the ground is applied through Contact 326B and closed contact 336A to the lead 125 of the pounds storage matrix. The release of relay 326 also applies power to computer relays 331)*335. The input of weight information represented by the matrix of the relay contacts I, I, K and Z is then stored by computer relays S30-335 following which the holding circuit for the relay 336 is broken. This applies the ground through contacts 336B to the tens of pounds matrix. In this manner, the stored weight information of FIGS. 8 and 9 is applied sequentially to the computer for utilization.

In this manner, a computer may receive the directly converted numerical outputs of this invention and use them to combine with unit price for computing the value of the article which has been weighed. It is obvious that the invention is not limited to sequential read out, since all of the closed ones of the contact groups of FIGS. 8 and 9 may be sensed simultaneously and used directly, or the information stored.

Control circuits and operation In the automatic weighing system ofthe Allen computer patent, there are provided controls and interlocks which initiate or permit certain automatic operations and prevents or block others under various circumstances. For

example, that system included control circuits for indicating that a weight had been placed on the platter, and that the scale `had come to balance, and the system of this invention also includes weight detecting means which must be operated, or be Ibypassed manually in orderv to initiate the automatic scale and computer operation. The weight detecting circuit in the Allen computer patent included a platter switch, namely the switch 55, but in the apparatus of this invention, a signal is taken or generated from the chart itself to indicate that more than a small weight is on the platter, such as one tenth of a pound or more. Provision is also included for manually overriding the weight detecting means so that weighing, computing and printing operation may be completed when the weight is less than the small predetermined weight, such as less than one tenth of a pound, and the circuits for accomplishing this result include the relays in the M and N columns (tens of pounds), as now explained.

The binary code from zero to two in the tens of pounds col-umn is as follows: O=M, 1=MN and 2=N. However, zero is also equal to no M, and a small portion rof the M column on chart 36 adjacent zero is accordingly removed for a distance corresponding to approximatelythe iirst one tenth of a pound, so that zero to .1()` pound in the M and N column gives no output or signal to the photocells. This causes M to come in only above a tenth of a pound, and thus a signal in either the M column or the N column will indicate that a weight is on the scale in excess of .l pound. The control circuit for the readout includes the start relay P, and as shown in FIG. l()l this circuit includes a pair of normally open contacts M-l and N1 of the N and M relays connected in parallel so that either one of these contacts can operate relay P through a normally closed contact `Q-3 of the power control relay Q and a closed contact O-1 of the nonrepeat relay O. It neither of relays M or N is operated, this indicates that there is less than a minimum weight on the scale, and relay P does not operate.

The balance or scale motion detecting circuit may be as described and claimed in the copending application of Allen, Ser. No. 220,765, tiled Aug. 3l, 1962, and assigned to the same assignee as this invention. In that application, there is described a portion of the circuit including the relays A and C employed as motion sensors or detectors. Only so much of the motion sensing control circuits is described herein as necessary to an understanding of the operation of this portion of the control circuit.

Referring to FIG. 4, it will be seen that two of the columns of binary marking, A and C, have markings which do not mutually coincide. The motion detection circuit uses a portion o'f the photocell readout circuit of relays A and C during scale movement, and includes a circuit for sensing the dropout ofeither relay A or C for signaling that the platter has come to rest. Since the binary markings of columns A and C are mutually exclusive in the sense that relays A or C willnever be operated simultaneously at any balance position of the scale, then it follows that when either of relays A or C is released following a delayed dropout, the scale is at balance, regardless of the state of the other relay. A feature of this arrangement, which is described and claimed in the application Ser. No. 220,765, is that the photocell motion detecting circuit, using two photocells, is not adversely aifected by small vibrations, since the arrangements of the photocells and the binary markings are such that at any balance position, at least one of the photocells A or C will be at a steady state condition.

In the following description of an operating cycle, `it is assumed that the scale and readout system are connected as noted above with the computing and printing circuits of the Allen patent. `In addition to the relay contacts already listed which are identified by the samereference characters as in that patent, it should be noted that certain o'f the components shown in FIGS. 9 and 10 similarly correspond to components in the same patent. Thus relay O, corresponds to relay NR in the Allen computer patent, and any part bearing a reference character above 200 correspond-s to the similarly numbered part in the same patent.

Referring to FIGS. 5, 6 and 10, when the scale is in balance with no load on the platter, relays 30S, O, P. Q and R and all of relays A to N will be released, as well as the relays having the contacts 309, 316, 326, 336, 346, 347 and 613, and the signal light 232 will be lighted. When a load is applied to the platter, the chart will begin to move and cause the photocell F, to operate its associated four-layer diode Va, causing relay A to close `and to be locked in this position for the reason that the inductive currents in photocell la are not bypassed in any manner, and these currents hold the four-layer diode operated across the null point of the full-wave unfiltered power supply. Immediately thereafter, relay C will operate and similarly be held by the memory function of its four-layer diode Vc. Y

The simultaneous operation of relays A and C will now operate relay R. If the weight exceeds .l0 lb., then simultaneously with the operation of relays A and C, either or both of relays M and N will operate, and this will cause the operation of relay P and the extinguishing of the ready light 232. The closing of contacts P-Z now establishes a shunt path around relays A and C, and since relay R is already closed, these two relays are shunted by condensers and 131 and their associated current limiting resistors 132 and 133.

As the scale proceeds towards its balance point, photocells Fa and Fc will be alternately operated by the coded chart 36 at slower and slower speeds, and when the balance point is nearly reached, one or both of photocells Fa and Fc will remain unlit. Since contacts P-2 and contacts R-l and R-2 are now bypassingthe inductive currents of relays A and C, these relays are completely under the -control of their associated photocells, because the four-layer diodes Va and Vc will no longer be receiving current during the `null period of the DC power supply. Therefore, one of `these four-layer diodes will stop conducting, and the associated relay A or C will remain operated only sufficiently long to discharge its associated condenser 130 or 131.

When one of relays A or C does release,'this causes the release of relay R,` after a `very short time delay,l

duirng which the inductive current of relay R recirculates through diode 135. The resulting opening of the contacts R-1 and R-Z removes the time-delay condenser from the other of relays A and C, thus permitting it to drop out immediately if its associated photocell is not calling for it to he operated. With the release of relay Rv andthe A13 closing of contacts R-3 and R-f4, relays A and C are now bypassed b` y diodes 136 and 137 so that the operation of relays A and C is still under the full control of the associated photocells, due to the continued bypassing of the inductive currents ofthe relay coils.

The release of relayl R closes lcontacts R-S, and since contacts P-4are already closed, relay Q will operate and set up its own holding circuit through contacts Q-6. Closing of relay Q will al-sofclose its contacts Q-S as one of the conditions for starting the computer and will open contacts Q-4 and thereby remove the bypass through diodes 140 and 141 from relays lM and N, and this will establish the holding conditions for the four-layer diodes Vm land Vn. At the same time, closing of contacts Q-2 will provide power to the relays B and D through line S6, and the openingof contacts Q-1 will eliminate one of the sources of power for line 80.

The opening of contacts Q-3 will remove power from relay P, and after a very short vperiod of time, during which the inductive current of relay P recirculates through diode 150, relay Pr will release and thereby cut offpower through line 80 to the photocells. The release of relay 'P opens contacts P-2, thus removing the bypass circuits'frorn relays A and C and thereby establishing the memory functions of their associatedfour-layer diodes Va and Vc; More specifically, this action rende'rsthe systeni insensitive to any 'further scale motion, since all of the four-layer diodes were ii'red or not according to the position of the chart 36, and those which were fired remain in that condition after power was removed from the photocells. The firing of the four-layer diodes during the short period of concurrent operation of relays P and Q isk so fast that there is no need for mechanically locking or holding the weighing mechanism in balance, but the rcmoval of power to the photocells as described, incombination with the self-locking' action of the red fourlayerV diodes, provides in eect an electronic lock preventing the readout system from responding to further scale motion until the cycle is completed as now described;

Thel closing of contacts P-5 applies a gnound on the computer startline 151 to lclose relay 303 and thereby 'to start the readout cycle of the computer, providedA the other conditions Iwith respect to relays 309 and 615` of the Allen cotmputer patent are satisfied as described therein. At the start ofthe computing cycle, the relays having the contacts 309 and 316 4in FIG. l0 will close as described in the Allen computer patent. This will operate relay O, which will lock itself in through its contacts O-3, anld light 232 will be operated through contacts 0 2 to indicate that a new load may now be placed fon the scale. Closing of relay 309l will also release relay 308, and this will in turn close the relays having the contacts 326,336, and 346 as described in the Allen computer p-atent. At the conclusion of the read out cycle as described in the Allen computer patent, a relay 347. in the computer operates, thus opening the holding circuit to relay Q as shown in FIG. l0. Relay Q releases, and conditions are now identical with those which existed at the start of this description with the exception that relay O has been operated, and this prevents any further computer operation through the opening of contacts 0 1 which prevent any operation of relay P.

As soon as the l-o-ad is changed after operation of light 232, relays A and C will close inthe manner previously described and cause the operation of relay R, which in turn will cause the release 'of relay O. When the scale arrives at a new balance point in the 'iweighing range above .l0 pound, a new cycle of readout will take place, initiate-d as already described by'closing of lrelays R and P, provided 'the cycle for 'the previous load has been completed and the label has been removed as indicated rby closure of the relay contacts 309 and 613 respectively. If, on the other hand, the scale has come to balance at` the zero point, a second operation would not take place because with neither of relays M nor N operated, relay P could 14 not be operated. Under these conditions, signal light 232 willbe lighted through contacts Q-3, N-Z and M-2.

Provision is made for causing the computer to function in the low weight range of .l0 pound or less, or when relay O has not been operated by movement of the scale, as at zero or to print duplicate labels Without removing the load from the scale. In such cases, neither of relays M and N is operated, or relay O is not released, and there is therefore no energizing circuit available for relay P. A manual switching 262 is accordingly connected in the circuit as shown in FIG. 10 to -bypass the M, N and O contacts and thereby vto provide for manual closing of relay P. This will initiate operation of the complete cycle as already described following the closing of relay P.

vThe scale further includes means for automatically sensing if the scale mechanism is in a position less than zero, or if the mechanism is in a position which exceeds the weighing capacity of the scale. Referring to Table 1V, it will be seen that relays E and G are never operated simultaneously in accordance with the code. Therefore, the chart is formed arbitrarily below zero, and above a maximum weight (such as above 25 pounds) in such a manner that E and G are caused to operate together, as already noted in connection with FIG. 4. Thus, a weight of 00.01 provides a signal of A, Q Q, I. A weight of 25.01 pound arbitrarily vprovides a signal of N, L, K, H, Q, E, A. It will be seen that in each case G and E are operated together, and this maybe used to indicate error and prevent'automatic operation. A circuit for utilizing this condition is shown in FIG. 9 as including a contact G-1 connected to receive a signal through a Contact E-l on the E relay. This signal, provided both the G and E relays are energized, is applied to shunt down the error detecting relay 348 of the Allen computer patent when relay 326 is closed as the result of the reopening of relay 308 as noted above and described in the Allen computer patent, thereby preventing completion of the computing cycle.

It will therefore be seen that this invention provides an automatic scale system in which a chart and associated reading transducers are used to determine when a weight has been placed on the platform, to sense motion, and to sense conditions of over and below capacity' situations. Theinventionfurther provides direct readout and cyclic binary to decimal conversion of the binary chart.Y A unique cyclic biquinary code is employed which reduces the number of binary columns or paths which arev required and therefore reduces the number of readout cornponents required. Also, a unique converting circuit is employed for translating the cyclic biquinary into true decimal numbers.

What is claimed is:

1. A computing scale system comprising, a weighing mechanism, a chart operatively connected with said mechanism having formed thereon a binary pattern representing a range of decimal weights, a mask having means forming a plurality of openings therein, a projecting system including a light source arranged to project an image of said chart onto said mask, each of said openings being smaller in effective area than the smallest division of said projected pattern, a separate photocell positioned adjacent each of said openings to receive light therethrough, and converter circuit means connected to said photocells and operative to provide an output of the decimal value of the pattern intercepted at said openings.

2. A computing scale system as dened in claim 1 wherein said pattern comprises a plurality of rows, there being one of said photocells` for each of said rows, and comprising means maintaining said photocells in clustered relation about the optical axis of said projecting system, said rows being staggered in the direction of movement of said chart to conform with said clustered placement of the corresponding said photocells.

3. An automatic weighing and price computing system having maximum and minimum weight limits, cornprising a scale mechanism, a chart operatively connected with said mechanism having a binary code formed thereon representing a range of weights which extend beyond said limits, photoelectric transducers positioned to respond to a portion of the code on said chart representing Vweight on the scale, electric circuit means connected to said transducers for translating the observed portion of said code into output signals representing a digital value of said weight, and means responsive to the occurrence of a characteristic combination of transducer outputs signaling the movement of said scale mechanism to a balance position outside of said system limits.

4. An automatic system as defined in claim 3 comprising means responsive to the occurrence of a diferent characteristic output of said transducers signaling motion of said scale and blocking operation of said electric circuit means.

5. In an automatic weighing and computing scale system including a platter on a balance mechanism, a projection system including a chart operatively connected with said mechanism and a plurality of photocells positioned to |be responsive to the balance position of said chart, and a computer, the improvement comprising a plurality of indicia on said chart representing a range of weights arranged in a binary code, said indicia and the range represented thereby extending to a maximum weight above the range of said balance mechanism, read out means connected to receive the output from said photocells and operable to translate said output into electrical signals for employment in the computer representing the weight onisaid platter, and circuit means responsive to a characteristic output of at least two ofV said photocells when said balance mechanism is in a stationary position above the range thereof and connected to prevent automatic operation of said system. Y

6.` In an automatic weighing and computing scale system including a platter on a balance mechanism, a projection system including a chart operatively connected with said mechanism and a plurality of photocells positioned to be responsive-to the balance position of said chart, and a computer the improvement comprising a plurality of inldicia on said chart representing a range of weights arranged in a binary code, said indicia and the range represented thereby extending from the minimum weight below zero to a maximum weight above the range of said-balance mechanism, read out means connected to receive` the output from said photocells and operable to translate said output into electrical signals for employment in the computer representing the weight on said platter, first circuit means responsive to a characteristic output of at least one of said photocells when said mechanism is in a stationary condition at a position with less than zero weight for preventingV automatic operation of said system, and further circuit means responsive to a further characteristic output of at least two of said photocells when said balance mechanism is in a stationary position above the range thereof and connected to prevent automatic operation of said system.

7. A weighing scale comprising a balance mechanism, a projection system including a chart operatively connected with said mechanism and having lformed thereon a binary code representing weight, photocells positioned to intercept the projected image of a portion of said chart, a converter `circuit connected to receive the output of said photocells, and circuit blocking means including one of said 'photocells responsive to a balance position of said chart with less than a minimum weight on said scale for preventing the operation of said converter circuit.

8. An automatic computing scale system including a weighing scale and a computer, comprising a projection system in said scale including a movable chart having formedthereon a binary code etectively arranged in rows and representing a range of weight, a plurality of photocells positioned to respond to a'portion of the projected image of said chart, there being one of said photocells for each of said rows, converter circuit responsive to whether each said photocell is illuminated or not by said projection system, and means including a pair of said photocells energized simultaneously by movement of said scale to a balance position outside of said weight range and connected to block automatic operation of said computer when said scale is outside of said range of weight.

9. An automatic system as detined in claim 8 wherein said computer includes motion detector means sensing a change in the output of at least one of said photocells and providing a signal blocking operation of said computer.

10. In an automatic weighing scale including a balance mechanism, the improvement comprising a projection system, a chart in said system movable in proportion to the weight of an article on said scale and having formed thereon a binary code arranged effectively into rows representing a range of weight, a plurality of photocells positioned to respond to a portion of the projected image of said chart corresponding to -the balance position of said mechanism representing the weight of an article on said scale, there being one of said photocells for each of said rows, a converter circuit including a circuit responsive to the operation of each said photocell, circuit means including a pair of said photocells operable upon motion of said weighing mechanism and connected to block the operation of said converterv circuit during the time when said scale mechanism is in motion and thereafter to energize said converter circuit, and means including another said photocell repsonsive to the movement of said scale a predetermined small amount above zero indicating the presence of a weight on said scale and connected to prevent the operation of said converterV circuit when said scale is in a range below said small amount.

11. In an automatic weighing and computing scale including a balance mechanism and a projection system including a relatively movable chart and a plurality of photocells positioned to be responsive to the projected indicia on` said chart, the improvement comprising indicia on said chart arranged in a binary code, controllable readout means for said photocells, and circuit means including at least one of said photocellsrand responsive to a characteristic output of said one photocell when there is less than a small minimum weight on said scale for blockin-gthe operation of said readout means.

12. The weighing and computing scale of claim 11 further including manually operable means for overriding said blocking means to provide for operation of said readout means with said small weight on said scale.

13. in an automatic weighing and computing scale system including a platter on a balance mechanism, and a projection system including a chart operatively connected with said mechanism and a plurality of photocells positioned to respond to an image of said chart, the improvement comprising a plurality of indicia on said chart arranged in a binary code, readout means connected to receive the output from said photocells and operable to translate said output into an electrical digital signal representing the weight on said platter at balance, and circuit means including at least one of said photocells and responsive to a characteristic output thereof at balance with less than a predetermined minimum weight on said platter for preventing the automatic operation of said readout means.

References Cited UNITED STATES j PATENTS (Other references on following page) 17 S 2,809,785 10/1957 Balde 23S-611115 2,963,222 12/1960 Auen ,'235-151 2,931,639 4/1960 Lauleret a1. 177-3 3,044,691 7/1962 Auen 23S-5s 2,948,464 8/1960 Auen. 3,289,777 12/1966 Wiuyard 177-178 2,605,965 8/1952 Shepard 23,5-61 2,855,539. 10/1958 Hoover 340-347 5 FOREIGN PATENTS 2,852,764 9/1958 Frothingham 340-347 355,963y 7/1961 France. 2,975,409 3/1961 P6111 'k 340-347 3,042,128 7/1962 Bell eertlcaL l77 210 ROBERT S. WARD, IR., PinmyExammel'. 3,061,026 7/1962 Hecox eral 177-13 3,120,287 2/1964 Allen 177-5 USC1R 2,577,815 12/1951` Saunderson et al. 250-209 10 177-25; 23S- 58; 250-208, 219

P0" O5" UNITI-LD STATES PATE NT OFFICE 559 CERTIFICATE 0F CORRECTIGN Patent No. 3,439, '760 Dated April 2.2, 19'69 Inventor-(s) Kenneth C. Allen It is certified that error appears in the above-identified patent and that said Letters Patent arehereby corrected as shown below:

Column 6, line Z4 "Table II" should read Table III. Column lO, line 40 "of a weighting scale" should read --to a weighing scale; lines 46 and 47, "and be" should read and may be Column 18, line 5 "355, 963 France" should read 355, 963 Swiss line 6 insert 1,231,229 France SIGNED ANU SEALED OCT 2 l 1959 (Smm Attest:

Edward M. Fletcher, 11% WILLIAM E. 'SGHUYLER, JR-

Attesting Officer commissioner `of Patents

Images