Method and apparatus for evaluating and sorting sheets in a high speed manner

Abstract

Electronic solid state circuitry incorporating the microprocessor for automatically controlling document handling apparatus. Sheets are advanced from a stack of sheets arranged in an infeed stacker and are moved one at a time at high speed through an examining location where sensors examine the sheets to determine their condition. The microprocessor periodically initiates an adjustment in the brightness level and gain control level of the lamps and sensor elements employed in the sensor array; tracks each sheet as it moves through the document handling apparatus; and evaluates the outputs developed by the sensors to determine the fitness of each sheet. A gating roller assembly, under control of the microprocessor, is operated to divert each sheet toward one of a plurality of output stackers according to the results of the evaluation performed by the microprocessor. Timing of the electronic circuitry is controlled by timing signals derived from the document handling apparatus to synchronise the electronic circuitry with the document handling apparatus. The sheets are tested during their high speed movement for a variety of conditions which are operator selectable and which conditions are operator adjustable.

Claims

What is claimed is:

1. Microprocessor-based control means for operating document handling apparatus comprising means for moving sheets in a first direction at spaced intervals along a predetermined path;

sensing means for detecting the passage of sheets and for detecting predetermined characteristics of said sheets;

at least one of said sensing means being adapted to generate a signal upon the passage of the leading edge of each sheet at the location of said one sensing means;

means for generating timing pulses at a rate which is a function of the velocity of said sheets moving along said predetermined path;

multi-stage counter means being pulsed by said timing means;

memory storage means;

means responsive to a leading edge signal for transferring the contents of said multi-stage counter means to a predetermined location in said memory storage means;

said microprocessor-based control means further including means for periodically sampling the contents of said counter means and for determining the difference between the count stored in said predetermined location in said memory storage means and the count developed by said multi-stage counter means at the time said counter means is sampled wherein the difference in said count represents the location of said sheet along said predetermined path;

document condition examining means positioned at spaced intervals along said path;

means for temporarily storing examined conditions;

means responsive to predetermined difference values between the status count stored in said predetermined location in said memory storage means and the count sampled from said multi-stage counter means for examining the conditions in said temporary storing means.

2. The apparatus of claim 1 further comprising means adjacent to the output end of said predetermined path for diverting sheets meeting predetermined criteria toward a first outfeed stacking location and for diverting sheets failing to meet said predetermined criteria toward a second outfeed stacking location;

said microprocessor-based control means further comprising means responsive to a predetermined difference between said status count in said first predetermined location in said memory storage means and the count in said counter means for operating said diverting means in accordance with said evaluated data whereby operation of said diverting means is initiated sufficiently prior to the leading edge of the sheet in question arriving at said diverting means to be assured that said sheet is diverted to the proper outfeed location.

3. The apparatus of claim 2 wherein said microprocessor-based control means further comprises means responsive to a leading edge of a sheet passing said sensing means for shifting the status count in said first predetermined location to a second predetermined location in said memory storage means and for shifting the status count of the sheet whose leading edge is now passing said sensing means from said multistage counting means into said first predetermined location in said memory storage means;

said microprocessor-based control means including means for periodically determining the difference between the status counts stored in said first and second predetermined memory locations and the status count presently in said multi-stage counter means for determining the time at which the outputs of said sheet evaluation means should be transferred to said memory storage means.

4. A method for evaluating sheets to determine their fitness including plural sheet evaluating means arranged at predetermined spaced locations comprising the steps of;

moving sheets at spaced intervals in a first direction along a predetermined path which moves said sheets past said sheet evaluating means;

generating pulses at a rate representative of the velocity of the sheets moving in said predetermined direction;

counting said pulses;

automatically restarting said count when said count reaches a predetermined maximum value;

storing said count at a predetermined location in a memory when the leading edge of a sheet passes a predetermined location;

periodically comparing the present count being developed against the count stored in said predetermined location for determining the difference therebetween;

examining the results of one of said sheet evaluating means when said difference lies within a predetermined range;

examining the results of a second one of said sheet evaluating means when said count lies within a second predetermined range different from said first range.

5. The method of claim 4 comprising the steps of operating a diverting means when said difference lies within a third predetermined range, said range being selected to assure operation of said diverting means prior to the time that the leading edge of the sheet to be diverted reaches said diverting means to provide sufficient time to operate said diverting means.

6. The method of claim 5 further comprising the step of operating said diverting means to divert sheets toward a fit stacking location when the evaluation of the sheet indicates that it has met certain criteria and for diverting the sheet to an unfit output stacking location when the data evaluated indicates that the sheet has failed to meet the aforesaid criteria.

7. The method of claim 4 further comprising the steps of:

counting the number of fit sheets;

counting the number of unfit sheets, and displaying said counts.

8. The method of claim 4 further comprising the steps of:

counting the number of fit sheets;

counting the number of unfit sheets;

comparing the number of fit sheets against an adjustable setting; and

halting the movement of sheets when the number of fit sheets reaches said setting.

9. The method of claim 4 further comprising the steps of:

counting the number of fit sheets;

counting the number of unfit sheets;

comparing the number of unfit sheets against an adjustable setting; and

halting the movement of sheets when the number of unfit sheets reaches said setting.

10. The method of claim 4 further comprising detecting for the presence of genuine sheets;

diverting the sheet which is other than genuine to a location for receiving unfit sheets; and

halting the further feeding of sheets.

11. The method of claim 4 further comprising the steps of detecting the presence of sheets at the input end of said predetermined path to halt the apparatus in the absence of sheets at said input end.

12. A method for operating document handling, counting and evaluating apparatus comprising the steps of:

moving sheets at spaced intervals in a first direction along a predetermined path;

detecting the presence of the leading edge of the sheets as they pass a predetermined location;

generating pulses at a rate which is a function of the velocity of the sheets moving along said path;

counting said pulses;

beginning a new count each time the count reaches a predetermined maximum value;

storing, in a first predetermined memory location, the count present at the time a leading edge of a sheet passes a first predetermined location;

transferring the count stored in said first predetermined memory location to a second predetermined memory location when the leading edge of the next sheet passes said first predetermined location along said path and storing the count being developed at that time in said first predetermined memory location;

periodically determining the difference between the counts in said first and second predetermined memory locations and the aforesaid instantaneous count;

sampling the results of one of the evaluation means when said difference lies within a first predetermined range;

sampling the results of a second one of said evaluation devices when said difference lies within the second predetermined range different from said first predetermined range;

diverting said sheets to a first output stacking location when the sheet meets the evaluation criteria and diverting the sheet to a second output location when the sheet fails to meet the evaluation criteria.

Description

FIELD OF THE INVENTION

The present invention relates to document handling apparatus and more particularly to electronic circuitry for high speed automatic control of document handling apparatus to examine sheets as they move one at a time at high speed through said apparatus and to control delivery of the sheets to one of a plurality of output stackers according to the results of the evaluation and at no reduction in handling speed.

BACKGROUND OF THE INVENTION

A document handling apparatus is utilized for handling sheets such as, but not limited to, checks, paper currency, food and premium coupons and other like documents. There are a number of applications in which it is desired to be able to handle such sheets at high speed, to evaluate said sheets to ascertain whether they meet or fail to meet certain criteria and to divert the evaluated sheets to an output path associated with the results of the evaluation. For example, in the handling of paper currency, it is extremely desirable to be able to sort paper currency in accordance with certain criteria. Many banks and other like institutions utilize automated facilities sometimes referred to as 24-hour banking equipment, in which it is possible to withdraw money at any hour of the day or night simply by inserting a plastic card into an appropriate slot and manipulating certain buttons upon a control panel for the purpose of withdrawing money, such as paper currency, for example. Such automated banking equipment has been found to operate successfully only with the use of new or nearly new paper currency, since paper currency which is worn or has any tears or folded corners will not be properly fed by the automatic teller equipment and will, in fact, cause it to jam. Since new or nearly new paper currency is often difficult to obtain on a regular basis from the Federal Reserve, one of the best techniques of obtaining new or nearly new paper currency which will meet all of the criteria necessary for use in automatic banking equipment, is to examine paper currency taken in by the bank and sort out all new or nearly new paper currency for use in the automatic teller equipment. This technique is presently being done manually which constitutes an extremely tedious and time-consuming procedure.

Other operations which banks and other similar institutions are interested in performing at high speed are evaluation of paper currency for purposes of sorting unfit paper currency from fit paper currency, in order to withdraw unfit paper currency from circulation and return same to the Federal Reserve for subsequent destruction. Paper currency which, although it may not quite meet the stringent criteria which must be met for use in automated teller equipment, may nevertheless be in satisfactory condition for use by the bank or other similar institutions in normal day-to-day transactions. It thus becomes desirable to sort otherwise fit paper currency from unfit paper currency in order to provide tellers with paper currency acceptable for continuing circulation and to remove unfit paper currency from circulation and for return to the Federal Reserve. Operations of this nature are also being performed manually. It is thus extremely desirable to be able to perform such operations in an automatic, high-speed manner and to be able to evaluate sheets such as paper currency to determine whether they are too stiff or too limp; too light or too dark; ripped, torn, perforated or otherwise damaged; have torn and/or folded corners and even evaluate such sheets to determine or aid in a determination of their authenticity.

BRIEF DESCRIPTION OF THE INVENTION

The present invention is characterized by comprising document handling apparatus and method and apparatus for controlling document handling apparatus at a high speed in order to perform all of the above-mentioned examinations, to evaluate the results of the examinations, and to sort sheets in accordance with said results in a substantially uninterrupted manner, wherein the examination and control means enables such operations to be performed at no reduction in system operating speed.

The document handling apparatus of the present invention comprises an infeed stacker for receiving a large stack of sheets to be evaluated. The mechanism delivers sheets from the stack on a one-at-a-time basis and at high speed through an examination station where the sheets are examined to develop signals representative of their condition. These signals are evaluated by the electronic circuitry, which incorporates a microprocessor, to determine whether the sheets meet certain criteria which, to increase the versatility of the equipment, are operator selectable.

The microprocessor utlizes signals from the sensor array to track each sheet as it moves through the document handling apparatus in order to control gating apparatus to gate sheets to appropriate output stacker locations depending upon whether the sheets have met or failed to meet the criteria for fit documents.

The document handling apparatus is capable of having several sheets at varioius locations within the path of movement of the document handling and counting apparatus. The microprocessor is capable of tracking each such sheet, receiving data for each sheet and storing said data in addresses allocated to the associated sheet, all of said operations being performed accurately and at high operating speeds. The microprocessor evaluates the sensor signals for each sheet and controls gating means to divert sheets to the appropriate output stacker location in accordance with the results of the evaluation.

A variety of sensing means are provided to produce signals which assure that the gating means have operated properly. These signals are evaluated by the microprocessor to assure proper operation or, in the alternative, to take appropriate action to halt the document handling apparatus, or portions thereof, to prevent the equipment from being damaged in the event of any improper or erroneous operation.

The sensing apparatus utilizes solid state electronic circuits which assure the provision of highly sensitive and precision sensing signals to facilitate highly accurate evaluation of the sheets to be assured that they meet the desired operator adjustable criteria.

Novel digital type sample and hold circuitry is employed for retaining signal conditions over long time intervals and especially in the event that the document handling apparatus is halted, for example, for batching purposes.

The aforesaid type circuitry is further employed in conjunction with the sensing arrays in order to continuously monitor signal levels of the sensor devices and lamp sources used in cooperation therewith to provide constantly updated compensation for any abrupt and/or gradual changes in components of the sensor array due to aging, accumulation of dust or dirt or for any other reason.

The sensing circuitry is utilized for counting and length measuring purposes as well as the criteria mentioned hereinabove to add still further versatility to the system.

The document handling apparatus utilizes a plurality of motors whose operation is controlled by the microprocessor. Timing means associated with the output of one of said motors generates timing pulses utilized to synchronize the electronics of the system with the mechanical document handling apparatus. The signals are used in each of the detection circuits.

The microprocessor controls a lamp source which is utilized as the light source for the sensor array. A lamp regulator circuit is employed to exert control over the operating voltage level of the supply source to assure that the lamp source is operated within tight voltage tolerances to control its output brightness level to achieve a long, useful operating life.

The sensors cooperating with the light source are coupled to amplifying means having automatic gain control circuitry for automatically and constantly updating the output levels of the amplifying means to be assured that any changes in output level which may be due to accumulation of dust or dirt, component aging, or other causes, is automatically fully compensated for in order to prevent erroneous operation of the detection circuits.

Limpness detection is performed by a limpness detector assembly which converts physical displacement of a movable detector member within the limpness detector assembly into an electrical signal. The signal is compared against adjustable reference levels for detecting the presence of sheets which may be either too limp or too stiff. A comparator output signal is presented to the microprocessor to indicate the condition of the bill, i.e., either too limp, too stiff or neither of the above.

The sensing signals, after undergoing the aforementioned automatic gain control, are simultaneously applied to the hole detection circuitry, folded corner detection circuitry and average density and length detection circuitry.

The hole detection circuitry, which is operator adjustable, is designed to indicate the presence of holes and/or tears within the sheet and further to indicate the portion or portions of the sheet in which the hole or tear is present. The sensor array utilizes a plurality of sensors arranged in side-by-side fashion, each designed to sense an associated "strip" of the sheet. The absence of light from all of said sensors is interpreted as the presence of a document, which information is utilized for document counting purposes. The presence of a hole or tear by any one or less than all of the sensors causes the detecting sensor or sensors to abruptly generate a pulse of large magnitude. The entire waveform undergoes slew rate limiting and an offset adjustment. The resultant signal is compared against the original unaltered signal, whereupon any portion of the unaltered signal exceeding the altered signal, which functions as a dynamic threshold level, causes a pulse or pulses to be generated at such time, which pulse is interpreted as the presence of a hole or tear. The duration of the pulse represents the size of the hole or tear measured in the direction of movement of the sheet and is utilized to control the accumulation of pulses by a digital sample and hold circuit whose output is compared against an adjustable threshold in order to detect the presence of a hole or tear greater than a predetermined size, the adjustable threshold enabling the size of the hole or tear to be ignored being operator selectable. The results of the evaluation for each sensor is stored within a bistable flip-flop for subsequent examination by the microprocessor.

The folded corner detection circuitry derives signals from the sensors in the sensor array which scan the sheets. The signals developed by the sensors are examined to determine the delayed occurrence of the corners of the leading edge of a sheet relative to the central portion of the leading edge of the sheet and the early occurrence of a corner of the trailing edge of the sheet relative to the central portion of the trailing edge of the sheet to detect the presence of a missing leading and/or trailing edge corner due to the fact that the corner is either torn or folded. These signals are stored for use by the microprocessor for subsequent evaluation.

A combined average density and length detection circuit utilizes the document detected signal developed by the hole detection circuit to initate the accumulation of timing pulses in a digital sample and hold circuit. The occurrence of the trailing edge of a sheet terminates the count, which is present in analog form to comparator means which provide signals representing either the presence or absence of a sheet which is too short or too long.

An average density signal is developed by accumulating timing pulses when the sheet is either too light or too dark, which thresholds are selected by operator adjustable controls. The count is compared against a density reference level to develop a signal when the document is darker (or lighter) than a machine adjustable threshold. Density readings are taken during each half of the sheet which serves to increase the sensitivity of the detection circuitry and which further serves as a means for aiding in the detection of the possible feeding of overlapping sheets. These signals are temporarily stored in bistable circuits pending their examination by the microprocessor.

Paper currency is examined for genuineness and the results of these tests are also made available to the microprocessor, which abruptly halts the apparatus so that the suspect sheet is the last sheet to be delivered to the unfit output location when the apparatus is turned off.

The microprocessor exerts control over all of the electronic circuits, energizing the lamp source employed as part of the sensor array assembly upon the occurrence of initial set-up conditions; initiates automatic gain control adjustments for the sensor adjustable amplifiers only upon the occurrence of intervals during which sheets are absent; collects the signals representative of the conditions observed by the sensor array for further processing; and controls the various motors and brake means based upon the observed conditions.

The timing signals for the electronic circuitry are derived from the document handling apparatus and applied to a timing counter which is repetitively stepped to a full count, automatically reset and subsequently stepped to a full count so long as the document handling apparatus is in operation.

As soon as a sheet passes a predetermined point within the document handling apparatus, this condition is detected by sensor means causing the microprocessor to store the count, hereinafter referred to as a status count, in the aforesaid timing counter at that instant, which status count is stored in a memory location assigned to that sheet, said count being unique to the last mentioned sheet. A second artificially generated offset count is simultaneously stored in a second memory location associated with the last mentioned sheet. The offset count is periodically updated by comparing the status count against the count in the counter which is continuously incremented by the timing pulses. Each updated offset count represents the advancement of the sheet to a particular location in the document handling apparatus. The microprocessor periodically examines the offset count and executes a sub-routine comprised of a plurality of steps to be performed at the time that the sheet reaches the locations in the document handling apparatus associated with the present offset count. As the next sheet comes "on line", the status and off-set counts previously stored in the memory locations assigned to the first sheet to come "on line" are transferred to a second pair of memory locations utilized to store the status and offset counts representative of a sheet which is moved a predetermined distance downstream from the "on line" location. The status and offset count for the sheet just coming "on line" are then stored in the first-mentioned pair of memory locations. This operation is repeated for several sheets wherein the document handling apparatus is capable of keeping track of as few as one and up to five sheets each moving at spaced intervals through the document handling apparatus between the infeed hopper and the outfeed stackers.

When certain counts are developed within the offset counters of each of the sheets in process within the document handling apparatus, said counts trigger the microprocessor to sample certain of the conditions being observed. The states of the signals are examined by the microprocessor which controls the gating roller to divert sheets toward the appropriate output stackers in accordance with the observed conditions. In one preferred embodiment, output stackers for fit and unfit documents are provided and sheets are selectively diverted thereto in accordance with the observed conditions. Detector means are provided along each of the alternate output paths and their conditions are sampled and observed by the microprocessor to be assured that sheets have, in fact, been diverted to the proper output stacker. In the absence of the condition which is anticipated to be present based upon the control signal applied to the gating roller, the microprocessor halts all but the stacker motor to prevent the document handling apparatus from being damaged.

The microprocessor also interfaces with visual display means and a control panel for exerting control over the adjustable thresholds of the detection circuits and the display means in accordance with operator selections undertaken through manipulation of the panel controls.

The detection circuits are adapted to retain any count developed therein during the examination of a sheet or sheets in the event that the document handling apparatus is temporarily halted, for example, during a batching operation. When the machine is restarted, the counts pick up precisely where they left off, assuring that the observed conditions are accurate and in accordance with the sensitivity adjustments selected by the operator.

The microprocessor also cooperates with sensor means to prevent the lamp from being illuminated when no sheets are present in the infeed stacker and also to turn off the lamp and the motors of the document handling apparatus when all the sheets in the infeed hopper have been processed through the document handling apparatus.

In the event that the operator controls are manipulated to perform a counterfeit detection operation, the microprocessor, in the presence of a signal representing a "suspect" document, causes the "suspect" sheet to be the last sheet to be transferred to the "unfit" output stacker, whereupon the document handling apparatus is abruptly halted. A display indication alerts the operator to the "suspect" condition enabling the suspect document to be removed for subsequent observation.

The system may be employed for document counting or may be employed for document counting and sorting whereupon one, more than one, or all of the aforementioned conditions may be sensed depending upon the desires of the operator.

OBJECTS OF THE INVENTION AND BRIEF DESCRIPTION OF THE FIGURES

It is, therefore, one object of the present invention to provide a novel, high-speed document handling apparatus and cooperating electronic solid state control means therefor to provide for high speed handling, examination and sorting of sheets in accordance with one or more operator selectable criteria.

Another object of the present invention is to provide novel detection circuits for use in conjunction with document handling apparatus for detecting certain conditions of the examined sheet and for comparing said conditions against adjustably selectable thresholds to ascertain the relative fitness or unfitness of the sheet in accordance with preselected criteria.

Still another object of the present invention is to provide novel, solid state electronic detection circuits for use with document handling and examining apparatus and incorporating novel, adjustable amplifier means provided to compensate for changes in signal levels of the sensors due to the accumulation of dust or dirt, aging of circuit components and the like, said adjustment being made during operation but at a time when gaps between sheets are passing the sensors.

Still another object of the present invention is to provide novel, electronic solid state detection circuits responsive to signals of a sensor array to determine the fitness of a sheet with respect to certain preselected criteria, and which incorporate digital sample and hold circuits capable of indefinitely storing the count of an accumulated condition during an interruption in the operation of the document handling apparatus, for example, due to batching.

Still another object of the present invention is to provide a novel document handling apparatus incorporating solid state electronic control means utilizing a microprocessor for monitoring and controlling all of the operations of the document handling apparatus, evaluating the signals developed by the detection circuits and diverting examined sheets to an appropriate one of plural outfeed stackers in accordance with the examined conditions.

Still another object of the present invention is to provide document handling apparatus incorporating a microprocessor and related electronic circuits wherein sheets passing through the document handling apparatus are automatically tracked by the microprocessor to assure performance of all operations of the document handling apparatus in accordance with the location and condition of each of the examined sheets.

The above as well as other objects of the present invention will become apparent when reading the accompanying description and drawings in which:

FIG. 1 shows a simplified plan view of a document handling apparatus designed in accordance with the principles of the present invention.

FIG. 1a shows a simplified plan view of the cooperating light source and sensor array of FIG. 1.

FIG. 2 shows a block diagram of the electronic control means for controlling the document handling apparatus of FIG. 1.

FIGS. 3a and 3b show a detailed block diagram of the central processor unit employed in the control system of FIG. 2 and embodying a microprocessor.

FIG. 4 is a schematic diagram showing the lamp control circuit for regulating the operating voltage provided to the lamp which is employed as part of the sensor circuitry.

FIG. 5 is a schematic diagram showing the automatic gain control circuit of FIG. 2 in greater detail.

FIG. 6 is a schematic diagram showing the limpness detection circuit of FIG. 2 in greater detail.

FIGS. 6a and 6b are diagrams showing an alternative arrangement for detecting for limpness.

FIGS. 7a and 7b together comprise a schematic diagram showing the hole detection circuit of FIG. 2 in greater detail.

FIGS. 7c through 7g show waveforms useful in describing the operation of the hole detection circuit of FIG. 7.

FIG. 8 is a schematic diagram showing the folded and/or missing corner detection circuit of FIG. 2 in greater detail.

FIG. 9 is a schematic diagram showing the average density detection and length measuring circuit of FIG. 2 in greater detail.

FIG. 10 shows a block diagram of the display circuit of FIG. 2 in greater detail.

FIG. 11 is a plan view showing the keyboard and control board of FIG. 2 in greater detail.

FIG. 12 shows a family of waveforms useful in describing the operation of the microprocessor of the present invention.

FIGS. 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28 and 29 are flow diagrams useful in explaining the operations performed by the microprocessor-based control of FIGS. 3a, 3b.

DETAILED DESCRIPTION OF THE BEST MODE OF CARRYING OUT THE INVENTION

FIG. 1 shows document handling, examining and counting apparatus 10 embodying the principles of the present invention. A detailed description of the apparatus of FIG. 1, as well as the power train and other associated apparatus is set forth in detail in copending application Ser. No. 188,906 filed Sept. 19, 1980 and assigned to the assignee of the present application. For purposes of understanding the control system of the present invention, it is sufficient to understand the operation of the apparatus of FIG. 1 which will be described hereinbelow. A more detailed understanding of the power train and the manner in which it operates the document handling examining and counting apparatus 10 can best be derived from the aforementioned copending application Ser. No. 188,906 whose teachings are incorporated herein by reference thereto.

The sheet feeding mechanism 10 is comprised of a plate 18 which is mounted to a supporting surface 12 supporting mechanism 10 by suitable mounting means such as rods 20a through 20e which are secured to the face of plate 12 and extend outwardly therefrom and in turn have plate 18 secured thereto by suitable fastening members. Plate 18 cooperates with supporting surface 12 to serve as cover means for the mechanisms arranged therebetween. Plate 18 and spacers and supports 20a through 20e serve as the means for positioning and supporting an elongated plate 22 which functions as both an infeed hopper and a guideplate for guiding sheets through the apparatus. The upper end 22a of plate 22 supports a substantially C-shaped channel 24 shown in FIG. 1a of application Ser. No. 188,906, whose base portion 24a rests upon the upper surface of plate 22 and whose upright arms (only arm 24 being shown in FIG. 1) extend upwardly therefrom, to serve as a means for receiving a stack S of sheets to be processed by apparatus 10, said stack S being supported between the aforesaid upright arms and upon surface 24a.

The sheets are supported by portion 22a and the next adjacent portion 22b of plate 22 and occupy the region generally as represented by the trapezoidal shaped dotted region S. A hole is provided in the base portion 24a of channel 24. A light source 25a and sensor 25b, also shown in FIG. 2, are positioned adjacent to said hole. When sheets are stacked in the infeed hopper, light is reflected from the bottom sheet in the stack S towards sensor 25a. In the absence of a stack, no reflected light reaches sensor 25a which develops a signal compared against a predetermined threshold which is adjusted to provide a sheet present signal which is higher than the signal due to ambient light. Alternatively, the signal from source 25a may be a predetermined frequency and a filter 25c passes light of only the aforesaid frequency to prevent ambient light from being interpreted as the presence of a document.

A shaft 26 supports an eccentric picker roller assembly 28 mounted to rotate upon shaft 27 and having a central eccentric portion 28a and opposing concentric outwardly extending ends. Only end section 28b and central section 28a are shown in FIG. 1 for purposes of simplicity. The outer ends each have annular grooves about their periphery for supporting and receiving a resilient O-ring belt. As shown in FIG. 1, O-ring belt 30 is entrained about the annular groove provided therefor in end section 28b of picker roller 28. The ends of roller 28 are concentric about shaft 27, while center portion is eccentric relative to shaft 27 as was mentioned hereinabove.

Resilient O-ring 30 is also entrained about a roller 32 having an annular shaped groove surrounding said roller and adapted to position and seat said O-ring. Although not shown for purposes of simplicity, a second annular groove is provided on the opposite end of roller 32 and has extrained therearound a second O-ring, similar to O-ring 30 and which is entrained about the opposite end projection of eccentric roller 28, which opposite projection has not been shown for purposes of simplicity.

A roller 34 is positioned downstream from roller 32 and is mounted to rotate about shaft 34a. Roller 32 is mounted to rotate about shaft 32a. Additional annular grooves, which are axially spaced from the previously mentioned grooves, are provided about the periphery of roller 32 to position and seat O-ring 36 and an additional O-ring (not shown). Cooperating grooves are provided at like locations about the periphery of roller 34 to seat the last-mentioned O-rings, only O-ring 36 being shown for purposes of simplicity. A roller 38 is positioned downstream from roller 34 and is mounted to rotate about shaft 38a. An O-ring 40 and a second O-ring (not shown) similar thereto are entrained about rollers 34 and 38 which rollers are both provided with annular grooves for seating and positioning a pair of such O-rings, only one O-ring, namely O-ring 40, being shown in FIG. 1 for purposes of simplicity.

A roller 42 mounted to rotate about shaft 42a is positioned just above the surface of roller 34.

Plate 18 is provided with an opening for receiving shaft 44. An elongated arm 46 is secured to shaft 44 and has its right-hand edge resting against the left-hand surface of member 48 which is secured to the left-hand end of threaded member 50. Member 50 threadedly engages a tapped opening 52a in a mounting block 52, secured to plate 18. Threaded member 50 is provided with a slotted end 50a for receiving the head of a screwdriver to facilitate its adjustment. A spring means 54 extends between a pin 55a provided on mounting block 52 and a pin 55b provided near the upper end of arm 46 for normally urging arm 46 clockwise about shaft 44. By adjusting threaded fastening member 50, the angular orientation of arm 46 about its axis of rotation, i.e. the center of shaft 44, may be simply and readily adjusted.

Also pivotally mounted upon shaft 44 is an elongated stripper assembly supporting arm 56 which is locked to swing with shaft 44. The forward free end of mounting arm 56 is provided with a pin 58 for supporting swingable stripper support 60. A solid stripper member 62 is secured to the underside of swingable support member 60, typically by suitable fastening means (not shown). A torsion spring 64 has its opposing ends respectively secured to arm 66 and a swingable member 60, urging member 60 counterclockwise about the axis of pivot pin 58 relative to arm 56. Thus springs 54 and 64 tend to resiliently urge stripper member 62 into engagement with the adjacent portion of roller 32, while at the same time being yieldable to relieve a possible jam condition, i.e. to relieve the sudden build-up of a curled document or two or more overlapping documents which move between stripper member 62 and roller 32.

The confronting surfaces of members 62 and 32 have differing coefficients of friction whereby, when a single document passes therebetween, the surface of roller 32 exerts the prevailing influence upon a single document, enabling the document to pass in the forward feed direction, as shown by arrow 68. In the event that two documents are simultaneously fed between members 62 and 32, the coefficient of friction between the two documents is substantially less than the coefficient of friction between the lower document and the surface of roller 32, allowing the lower document to move in the forward feed direction 68. The coefficient of friction between member 62 and the upper document is also greater than the coefficient of friction between the two documents causing the upper document to be impeded from moving in the forward feed direction, thereby stripping the overlapping sheets fed therebetween to substantially assure that the sheets will be fed in a single file past the position of the nip formed between members 62 and 32.

The support members 20f, 20g and 20h which substantially perform the same functions as support members 20a through 20e, in addition to supporting plate 18, support an upper plate 70 having a plurality of bends therein which define flat portions between said bends, said flat portions being designated 70a through 70e. Portions 70a, 70b and 70c cooperate with portions 22a and 22b of guideplate 22 to define a stacker region for supporting a stack S of sheets and further, to define a tapering entrance throat portion between plate portions 70b-70c and 22b.

Sheets in stack S which rise above portion 70b have their leading edges resting against plate portion 70a which serves to relieve the portion of the stack therebeneath from a part of the weight exerted on the stack S by sheets arranged above the corner between portions 70a and 70b.

The central portion 28a of eccentric roller 28 is preferably fitted with a pair of O-rings (not shown) to provide good frictional engagement between the O-rings and the bottommost sheet in the stack S of sheets. The eccentric portion 28a of roller 28 together with the last-mentioned O-rings, serve to "jog" the stack upwardly and to exert a frictional force on the bottommost sheet, to drive the bottommost sheet in feed direction 68 to cause the sheet to be moved between members 62 and 32 for the feeding and stripping operations, as was described hereinabove.

Sheets moving past members 62 and 32 pass between plate portions 22c and 70d and are guided by the upper runs of O-rings 36 and 40 and the surface of roller 42, causing the sheets, being fed in single file, to undergo a change in direction, initially being fed generally diagonally downward as shown by arrow 68 to being fed generally diagonally upward as shown by arrow 68a. Roller 42, freewheeling mounted on shaft 42a, is arranged to smoothly guide sheets as they make the transition from being moved off of the upper run of O-ring 36 and on to the upper run of O-ring 40.

As sheets move along the upper run of O-ring 40 and pass over roller 38, the sheets are guided between the surface of roller 38 and guideplate portion 70e where they are caused to enter into the nip between roller 74, mounted to rotate upon shaft 74a, and idler rollers 76.

A pair of idler rollers are resiliently positioned above roller 74 and are resiliently mounted by suitable leaf spring means. As shown for example in FIG. 1, one such idler roller 76 is mounted to rotate about shaft 76a which is supported by the free end 78a of leaf spring 78 whose opposite end is secured to swingable plate 99 by fastening means 80, swingable plate 99 forming part of a swingably mounted unit 101, to be more fully described.

The rpm (revolutions per minute) of roller 74 exceeds the rpm of roller 38, so that, as documents enter into the nip between rollers 76 and 74, they are abruptly accelerated to move at a higher linear velocity, causing the trailing edge of the document fed through the nip formed by rollers 74 and 76 to move a predetermined spaced distance from the leading edge of the next document to be fed to said nip, providing a gap between said trailing and leading edges sufficient to perform counting and sensing operations on said sheets.

The roller 74 preferably has a surface with a high coefficient of friction. The rollers 76 are provided with grooves for receiving and supporting an O-ring, such as O-ring 84 to be assured that the accelerating force is imparted to sheets with a minimum of slippage.

Positioned immediately downstream of the acceleration roller 74 and idler roller 76 is a light source assembly 84 and a light sensor array 86. Light source 84 is comprised of a housing containing a lamp, preferably a halogen lamp (not shown). The cover plate 88 over the end of housing 84 adjacent to the feed path 68a is transparent. An opaque mask is provided upon the cover plate to enable only an elongated slit of light to be passed upwardly through transparent plate 88 toward the light sensor array 86. Array 86 is comprised of a plurality of sensors, such as for example the sensor 86a. The remaining sensors 86b-86d are arranged in an end to end fashion so as to be substantially aligned with the elongated slit provided in transparent cover plate 88. A similar transparent cover plate 86e is provided across the bottom surface of array housing 86.

As shown best in FIG. 1, the array assembly 86 is comprised of a housing aligned with a slit 92 in swingable plate 90 which slit 90 is divided into four compartments, each of which receives and supports the sensing surface 86a through 86d of an associated sensor element 86. As can be noted, each sensor surface has a rectangular shape. Elongated narrow dotted rectangle 88a represents the slit provided in the mask formed over the upper end of the light source housing 84 to define the region over which light is emitted from the light source assembly 84 and toward the light sensor array 86.

A preview sensor 94 is positioned above an opening in plate portion 70e and cooperates with a light source, preferably an LED 96, to function as a preview sensor for a purpose to be more fully described. Note also FIG. 2.

The swingable plate portion 99 upon which the idler rollers, such as idler roller 76 and the sensor array 86 is mounted, forms part of swingably mounted unit 101 having a plate 102 with a mounting portion 102a provided with an opening 102b for cooperating with the opening 18a in plate 18 for swingably mounting assembly 101. Assembly 101 has a cover lid portion 103 mounted upon a pair of spaced parallel side plates 102 and a plate (not shown) similar thereto, which lid rotatably mounts a fastening member 104 in a freewheelingly fashion. The lower end 104a of freewheeling mounted fastening member 104 is adapted to threadedly engage a tapped aperture 106a in block 106 which is secured between plate 18 and mounting plate 12. Thus, the swingably mounted assembly 101 serves to facilitate examination of the sensor array assembly as well as other internal mechanisms and/or components contained therein.

Lid 103 supports a group of spacer rods 108 which are secured at their upper ends to lid 103 and which position and support a printed circuit board 110 at their lower ends, said printed circuit board 110 supporting electronic components which cooperate with sensors 86a through 86d of the sensor array 86 for providing signals utilized for sheet examination and evaluation purposes, as will be more fully described.

The rollers 28, 32, 34, 38 and 74 are all driven by the feed motor M.sub.f (note also FIG. 2), the driving coupling as was described hereinabove, being obtained through the power train described in detail in copending application Ser. No. 188,906. Motor Mf as shown in FIG. 2, has an output shaft 112. A gear 114 is mounted upon shaft 112. Gear 114 is provided with a plurality of gear teeth 114a about its periphery and is secured to the feed motor output shaft 112 to rotate in unison with shaft 112. A light source 118 and a light sensor element 120 are positioned on opposite sides of gear 114 adjacent to the periphery thereof whereby teeth 114a pass between members 118 and 120 to cause light from source 118 reaching sensor element 120 to be modulated in a pulse-like fashion for generating system timing pulses to be employed in a manner to be more fully described.

The aforementioned power train is designed in one exemplary embodiment to cause the picker roll 28 to rotate at a speed which imparts movement to the document so as to be capable of achieving a velocity of 113 ips (inches per second). The feed roller 32 is rotated at a speed capable of moving documents along feed path 68 at a linear velocity of 106 ips. The acceleration roller 74 rotates at a speed sufficient to accelerate sheets so that they reach a velocity of 176 ips.

A limpness detector assembly 142 is located downstream from the light source and sensor array 84, 86, and is comprised of a pair of elongated generally cylindrical-shaped members 144 and 146, each mounted to rotate about shafts 144a and 146a and each having a gear-like periphery 144b and 146b respectively. Shaft 146a is mounted upon a a swingable arm (not shown) which is resiliently biased to normally urge gear-like roller 146 toward gear-like roller 144. As sheets pass therebetween, a counterforce is exerted upon gear-like rollers 144, 146, the magnitude of the counterforce being a function of the relative stiffness or relative limpness of sheets passing therebetween, thereby limiting the movement of gear-like member 146 toward gear-like member 144. Members 146 and 144 are mechanically coupled and driven so that the teeth of one of said gear-like rollers at least partially enters into the grooves arranged between the teeth of the other of said gear-like rollers and vice versa, in order to impart an undulating configuration to the sheet passing therebetween. The degree of said undulations is a function of the interaction between the force exerted upon the sheet by gear-like rollers 144 and 146 and the counterforce exerted by the sheet passing therebetween upon gear-like rollers 144 and 146. For example, very stiff sheets do not experience any bending, while extremely limp sheets such as thin onion-skin sheets, undergo a maximum amount of bending. A detailed description of the limpness detector is set forth in copending application Ser. No. 188,906 filed Sept. 19, 1980 and assigned to the assignee of the present invention.

The aforementioned driving coupling assures substantially synchronised rotation of gear-like rollers 144 and 146 in order to assure the proper entry of the teeth of gear-like roller 144 into the grooves arranged between the teeth of the other gear-like roller 146, and vice versa.

A pair of elongated O-rings, only O-ring 152 being shown in FIG. 1, are entrained about pulleys 154, 156, 158, 160, 162 and 164. Another pair of O-rings, only one such O-ring 166 being shown in FIG. 1, are entrained about pulleys 160, 168, 170 and 172. Pulleys 154, 156, 158, 162, 164, 170 and 172 are all freewheelingly mounted so as to be driven by O-rings 152 and/or 166. Pulleys 154, 156, 158, 160, 162 and 172 are all mounted to rotate about shafts 154a, 156a, 158a, 122b, 162a, 172a and 168a, all of which are mounted in a stationary fashion so that they are capable only of rotating about their central axes.

Pinch rollers 164 and 170 are rotatably mounted upon shafts 165, 171 provided at the free ends of a pair of swingable levers 176 and 178 in a freewheeling manner, each being pivotally mounted to support surface 12 by pivot pins 176a and 178a respectively. Centrally located pulleys 164a and 170a, shown in dotted fashion, are also freewheelingly mounted upon shafts 165, 171 and rotate independently of pinch rollers 164, 170. Pulleys 164a, 170a have recesses for receiving and seating O-rings 152, 166. The diameter of pinch rollers 164a, 170a, is greater than the diameter of the pulleys 164a, 170a to prevent O-rings 152, 166 from engaging pinch rollers 190, 192 and O-rings 194, 196.

The linear portion 152a, curved portion 152b and linear portion 152c of the path defined by O-ring 152, cooperates with the linear portion 166a, curved portion 166b and linear portion 166c of the path defined by O-ring 166 to cooperatively define a conveying path between which sheets exiting from the limpness detector assembly 142 are caused to be fed in a generally diagonally upward direction along path portions 152a, 166a and thereafter experiencing movement along a curved path portion 152b-166b, whereupon the documents are then moved in a generally downward vertical direction, as sheets move between path portions 152c-166c. Based upon the exemplary values set forth hereinabove, the sheets are moving at the same linear velocity through the path defined by O-rings 152 and 166 as they move through the limpness detector assembly 142 and the acceleration roller and cooperating idlers, namely, 176 inches per second.

Still considering FIG. 1, a pair of pinch rollers 190 and 192 are mounted to rotate about shafts 190a and 192a each having entrained thereabout an O-ring 194, 196. O-rings 194 and 196 are seated in grooves provided at the central portion of the pinch rollers 190, 192 and are further entrained about a directly driven large diameter pulley 198 and 200 respectively, each rotating about a shaft 198a, 200a respectively. Each of the pulleys 198, 200 has integrally joined thereto and extending from both sides thereof a pair of smaller diameter pulley portions so that the pulley 198 is arranged between the aforesaid smaller diameter pulley portions. Only one such small diameter pulley portion, namely portions 198c and 200c, is shown in FIG. 1 for purposes of simplicity, it being understood that each of these pulley portions receive and support an O-ring 202 and 204 which is further entrained about a cooperating pulley 206 and 208 respectively, each rotating about shafts 206a and 208a respectively.

The pulleys 198 and 200 further support rotary fan-like stacker wheels 205, 207 comprised of a plurality of curved resilient blades 210, 212, arranged at spaced intervals in the manner shown so as to form pockets 211, 213, between adjacent pairs of resilient blades 210, 212. Each sheet is adapted to be driven into one of said pockets in a manner to be more fully described. The sheets are subsequently stripped from their pockets by O-rings 202 and 204 and thereafter deposited upon an associated stacking plate 216, 218, each having upright sides 216a, 216b and 218a, 218b, for supporting and gathering sheets thereon. Upright walls 216b and 218b are provided with clearance slots to permit unimpeded movement of the run 202a, 204a, of O-rings 202 and 204 respectively.

One exemplary embodiment is designed so that the runs 194a and 196a of O-rings 194 and 196 are driven at a velocity such that sheets passing through the nip between pinch rollers 164-190 and 170-192 are accelerated to achieve a linear speed of the order of at least 178 ips in order to quickly "grab" the leading edge of the sheet after it has been deflected by the gating roller 250. The stacker wheels 205, 207, are mounted upon the shafts 198 and 200 which also rotatably support the pulleys 254 and 252. The stacker wheels 205, 207 are rotated so that the tip speed at the free ends of curved flexible fingers 210 and 212 is of the order of 28 ips. The much higher linear speed of the sheets assures insertion of each of the sheets deeply into a pocket 213. The curvature of the pocket 213 serves to decelerate each sheet as it enters the pocket 213.

A gating roller 250, mounted for rotation upon the gating roller motor shaft 252, is adapted to rotate in either a clockwise or counterclockwise direction, dependent upon the polarity of the driving signal applied to the gating motor Mg by the microprocessor.

The gating motor Mg is preferably a d.c. motor capable of rotating at a speed in the range of 2,000 to 8,000 rpm and preferably of the order of at least 3,600 rpm, and which is capable of rapidly reversing direction and reaching its maximum rpm in the reverse direction within an extremely short time interval.

In order to be assured that sheets are deflected in the proper direction by gating roller 250, as will be more fully described hereinbelow, a pair of sensor elements 260 and 262 are arranged just downstream of pinch rollers 190 and 192. A pair of light sources 264 and 266, which may for example be light emitting diodes (LEDs) are arranged adjacent to the sensors 260, 262 respectively so that, as sheets pass therebetween as represented by arrows 270 and 272, the light rays from each source are attenuated by the presence of the sheet causing the reduced brightness condition detected by sensors 260 and 262 to be interpreted as the passage of a sheet, which information is utilized by computer control means 280 (see FIG. 2), to control the operation of the apparatus.

The operation of the document handling examining and stacking system is as follows:

A stack of sheets which may, for example, be paper currency, are placed in the infeed hopper. See stack S of FIG. 1. When the document handling apparatus 10 is turned on, the rotation of the eccentric picker roller 28 jogs the stack S upwardly and its O-rings (see O-ring 72) frictionally engage the bottommost sheet, accelerating the bottommost sheet in the forward feed direction 68 whereby the bottommost sheet is advanced through the tapering throat portion to move into the nip formed between stationary stripper member 62 and feed roll 32. Members 62 and 32 cooperate in the manner described hereinabove to assure that sheets are fed in a single file as they pass through the aforesaid nip between members 62 and 32 and are advanced in the feed direction shown by arrow 68. The sheets undergo a turn at rollers 42 and 34 and thereafter move in an upward diagonal direction shown by arrow 68a. The sheets are abruptly accelerated by acceleration roller 74 and cooperating idler 76 in order to form a gap between the trailing edge of the document accelerated by accelerator roll 74 and the leading edge of the next document to be fed therethrough, said gap being of a length sufficient to prevent overlapping between documents and thereby facilitating counting of documents as well as providing an interval between sheets sufficient to enable the control circuitry to perform certain functions such as the gain control adjustment of the sensors, as will be more fully described hereinbelow.

As the sheets, which in the example given hereinabove, are moving at a linear speed of the order of 176 ips, pass between lamp source 96 and sensor 94 and subsequently between sensor array 86 and light source 84, the signal conditions from the sensors of array 86 and from sensor 97 are fed to computer control 280. The type of tests performed on the sheets, which may for example be paper currency, are: density of the sheets, i.e. are they "clean" or "dirty"; do the sheets have tears, cuts, slits or perforations; are there folded or torn corners; and are the sheets of the proper length, i.e. are they too long or too short.

The limpness detection assembly 142 is designed to detect the relative limpness or stiffness of the sheets and is further designed to indicate the presence of foreign material or members affixed to the sheets, for example, masking tape or transparent tape, staples and the like, which materials are often used to repair a torn bill. The limpness detector assembly 142 also serves as a means for indicating the presence of folded corners, as well as erroneous double feeding of documents by providing a "too stiff" signal in the event of passage of two documents in an overlapping fashion. Signals from the limpness detector apparatus 142, as will be described in detail hereinbelow are also provided to computer control circuit 280 in order to control the apparatus in accordance with the test or tests being performed.

In addition to the above, it is also possible to provide a counterfeit detection apparatus 284 which is positioned between the limpness detector assembly 142 and the sensor array 86, in order to detect the presence of suspect (i.e. possible counterfeit) bills. Counterfeit detection apparatus of this type is described in U.S. Pat. No. 4,114,804 issued Sept. 19, 1978 and assigned to the assignee of the present application. The counterfeit detection apparatus represented by black box 284 also provides its signals to one input of the computer control 280 which further receives signals from the post gate detectors 260 and 262 and which further provides control signals to the feed, stacker and gate motors Mf, Ms and Mg respectively.

The computer control 280 is provided with selection means for example, adapted to select those tests which are to be performed, it being understood that none, all or less than all of the tests can be performed simultaneously depending upon the setting of the selection members to be more fully described in connection with FIG. 11.

The first and second output stacking platforms 216 and 218 may arbitrarily be assigned to respectively stack fit and unfit documents, fit documents being described as those which meet the desired criteria based upon the tests being performed, and unfit documents being those which fail to meet the desired criteria. For example, documents which are too limp and/or too stiff may be collected upon stacker plate 218 while documents that meet the desired criteria, i.e., fall between the criteria of being too stiff and too limp, are stacked upon stacker plate 216.

Based upon receipt of the appropriate information, computer control 280 is designed to apply a signal of the appropriate polarity to gating motor Mg in order to rotate the gating roller 250 in the proper direction. Let it be assumed that the last document tested is now entering into the sheet conveying path formed by O-rings 152 and 166, and that this sheet, in accordance with the signals applied to computer control 280, has been classified as unfit. Computer control 280 will therefore apply a signal to gating motor Mg at a time sufficiently in advance of the sheet entering into the conveying path formed by O-rings 152 and 166, to be assured that gating roller 250 reaches its desired operating speed before the leading edge of the document to be appropriately diverted reaches gating roller 250.

The path along which the next sheet to be diverted to the appropriate output stacker is moved, advances the sheet along a path 290 which lies substantially along an imaginary diameter 250b of gating roller 250 so that the leading edge of the sheet will strike the surface of gating roller 250 at an angle which is substantially perpendicular to an imaginary line which is tangent to the surface of gating roller 250 and which intersects path 290 at point 292. By rotating gating roller 250 at a sufficiently high speed (i.e. rpm), proper deflection of the sheet is made possible. For example if the rotating speed is too low, since the sheet is moving at a very high rate of speed, in the example given 176 ips, the surface of roller 250 acts as a stationary wall and the sheet will simply bounce off of the surface of the gating roller and will not be properly deflected. However, when the tangential speed of the gating roller 250 is sufficiently high and is much greater than the linear velocity of the sheet, it is thus possible to deflect the sheets in a rapid and effective manner. In the example given, the gating roller 250 is caused to rotate clockwise, deflecting the leading edge of the sheet toward the right and causing the sheet to move into the nip formed between pinch rollers 170 and 192. The pinch rollers 170 and 192 "grab" the sheet and cause it to be accelerated as it is moved through the nip formed by pinch rollers 170, 192 and downwardly along the right-hand run 196a of O-ring 196 which serves as a means for moving sheets therealong as well as guiding said sheets toward and into the pockets 211b formed by adjacent pairs of fingers 211a. The O-ring 196 which may also be a flat belt, if desired, is formed of a resilient material having a relatively high coefficient of sliding friction which engages the sheet and serves to urge the leading edge of the sheet deeply into one of pockets 211b in stacker wheel 211. The curvature of each pocket 211b, defined by the curved fingers 211a serves to hold the sheet as the stacker wheel 211 rotates clockwise. The leading edge of the sheet in each pocket 211b bears against the right-hand run 204a of O-ring 204 which serves to strip the sheet from each pocket 213 as the inner ends of the fingers 211a begin to move past O-ring 204. Preferably a pair of O-rings are placed on opposite sides of each of the stacker wheels 209, 211. The stripped sheets are then caused to move downwardly where they are collected upon stacker plate 218. The leading edges of the sheets engage the right-hand run 204a of O-ring 204 which serves to drive the leading edges of the sheets downwardly to form a neat stack whereby the O-ring 204 serves the dual functions of stripping documents from the stacker wheel assembly 211 and serves to urge the leading edges of the documents downwardly towards the stacker plate 218.

Stacker wheel assembly 209, O-rings 194 and 202, and stacker plate 216, function in a manner identical to the corresponding elements 211, 196, 204 and 218 described hereinabove.

As was mentioned hereinabove, pinch rollers 164 and 170 are swingably mounted upon arms 176 and 178. The rollers 164 and 170 are designed to rotate clockwise and counterclockwise respectively, as shown by arrows 294 and 296. In order to permit the clearing of jam conditions occurring in the nips formed between rollers 164-190 and 170-192 respectively, arms 176, 178 are also free to swing in an over-center fashion in the event of a jam to provide an automatic arrangement for clearing a jam. Sensors in the form of microswitches 297 and 298 may be provided to indicate the release of swingable arms 176 and 178 from their operative position in order to provide indications to the computer control 280 to enable the computer control to take appropriate action.

The post gate sensors 260 and 262 function in a somewhat similar manner to provide signals to the computer control 280 in order to be assured that documents have been deflected in the proper direction by gating roller 250. Thus, for the example given hereinabove, assuming gating roller 250 to be rotating clockwise, computer control 280 will examine the signal derived from sensor 260 to be assured that a document has passed between sensor 260 and LED 264 at the proper time. In the event that this signal is not derived and/or an erroneous signal is derived from sensor 262, even though gating roller 250 is rotating clockwise, computer control 280 will interpret this data as an error condition and take appropriate action which preferably takes the form of deenergization of the feed motor Mf and the gating motor Mg, preferably allowing the stacker motor Ms to continue rotation to clear any documents from the region of gating roller 250 and collect said documents at stacker plates 216, 218.

The computer control 280 receives a signal from sensor 120 shown in FIG. 2 in order to provide proper timing for the apparatus. For example, assuming an ideal condition in which a local supply source provides an operating voltage of a precise voltage and frequency, all pulleys, belts and the like will be likewise rotating and moving at an ideal speed. However, in the event that there are any sudden surges and/or gradual changes in the operating voltage and/or frequency of the local source, and/or in the event that the motor undergoes an abrupt or gradual change in its operating characteristics, this will directly affect the operating speed of the aforesaid feed, stacker and gating motors Mf, Ms and Mg. However, by deriving timing pulses directly from one of said motors, namely the feed motor Mf, any changes, whether gradual or sudden, in the local supply source, are immediately reflected in the timing pulses developed off of timing gear 118 to assure proper operation of the apparatus due to the synchronous operation of the mechanical system and the system electronics.

The gating roller 250 is preferably a low mass member to facilitate its rapid acceleration and deceleration. To accomplish this, the gating roller 250 may assume a variety of configurations having low mass. One gating device suitable for this purpose is the cylindrical roller 250 shown in FIG. 1 which is comprised of material of low mass, such as cork. Other suitable gating rollers are described in detail in copending application Serial No.

FIG. 2 shows the solid state electronic control circuitry 280, comprised of a central processing unit 302 incorporating a microprocessor as will be more fully described in connection with FIGS. 3a and 3b. Output control line 304 is coupled to driver circuitry 306 whose outputs in turn are connected to brake means 308 and the gate, stacker and feed motors Mg, Ms and Mf respectively, to provide the signals for stopping and starting the motors and for energizing and deenergizing the brake means 308. Brake means 308 applies a braking force, represented by dotted line 308a, to the output shaft 112 of feed motor Mf.

As was mentioned hereinabove, the teeth 114a of timing gear 114 modulate light from light source 118 which may, for example, be an LED, which light is directed to sensor element 114 which may, for example, be a phototransistor. Appropriate power is applied to the light source and sensor elements 118 and 120 by coupling circuit 310 which receives the power source at terminal 310a and which provides the signal developed by light sensing element 120 at its output 310b. Output 310b is simultaneously coupled to associated inputs of the hole detection circuit 312, folded corner detection circuit 314, average density and length measuring circuit 316 and an 8-bit timing counter 320 provided as part of the central processing unit 302. The manner in which the timing pulses are utilized will be described in greater detail hereinbelow.

The aforementioned light emitting element 25a (FIG. 1) and cooperating light sensing element 25b shown also in FIG. 1, are coupled through line 322 to stack sensor input line 302a of microprocessor 302 to indicate to the microprocessor whether or not sheets are provided in the infeed stacker.

The aforementioned preview sensor means shown in FIG. 1 and comprised of a light source 96 and light sensing element 97, is located upstream relative to the sensor array 84 and is utilized to initiate a gain adjustment operation in a manner to be more fully described. The outputs of detector 97 are coupled through preamplifier circuit 324 and its output line 324a to one input of the automatic gain control circuit 326.

The limpness sensor is comprised of a permanent magnet 328 mounted in a stationary fashion and cooperating with a swingably mounted, hall-effect device 330 which undergoes displacement relative to magnet 328 as a function of the displacement between movably mounted gear-like roller 146 and stationary mounted gear-like roller 144. The hall-effect sensor 330 develops a signal applied to a preamplification circuit 332 whose output signal 332a is coupled to inut 334a of limpness detection circuit 334. The limpness detection circuit compares the limpness sensing signal against predetermined threshold levels as will be more fully described hereinbelow, to develop a pair of output signals each capable of assuming either one of two binary states, said signals being applied through a pair of lines, represented in FIG. 2 by line group 334b, which signals are applied to an associated pair of inputs identified as 302b in FIG. 2, said inputs being described in greater detail hereinbelow.

The post gate sensors 260, 262 and their associated LED's 264, 266 as were described in connection with FIG. 1, are also shown in FIG. 2. The outputs 260a and 262a of sensors 260 and 262 are coupled to input ports 302c, 302d respectively, of the central processing unit 302.

The lamp assembly 84 referred to in FIG. 1 utilizes a halogen lamp 85 whose voltage is regulated by lamp regulator circuit 346 to provide a supply voltage whose levels are accurately controlled within tight tolerances in order to prevent the halogen lamp from developing temperature levels which are either too high or too low, either of which conditions tend to significantly shorten the life of the lamp. In addition, the lamp regulating circuit also assures that the lamp operates at the optimum brightness level.

As was described hereinabove, the output of the halogen lamp is passed through the transparent slit 88a provided in the mask position over the lamp housing (see FIG. 1a). The sensors 86a through 86d have their outputs respectively coupled to inputs of associated preamplifiers within preamplifier circuitry 324. The amplified output signals appearing in the output line group 324c are applied, together with the preview sensor signal in line 324d to associated inputs 326a and 326b respectively, of the automatic gain control circuit 326.

The outputs of sensors 86a through 86d after undergoing amplification and gain control adjustment, are applied through line group 326c simultaneously to the inputs 312a, 314a and 316a of hole detection circuitry 312, folded corner detection circuit 314 and average density and length measurement circuit 316, respectively.

The amplified output of the preview sensor 97 appears at output 326d of automatic gain control circuit 326 and is applied to input 302e of central processing unit 302 for initiating a gain control adjustment operation, as will be more fully described. A gain reference voltage level is applied to input 326d of circuit 326 from voltage reference level circuit 336 which provides all of the required reference levels necessary for use as supply voltages, as well as reference level voltages utilized by the various detection circuits.

The hole detection circuit 312 utilizes the system timing pulses derived from the feed motor Mf at input 312c, the signals from the sensors of sensor array 86 at input 312a and various d.c. reference levels for detecting the presence of holes, through the use of a digital sample and hold circuit and cooperating comparator circuits, as will be more fully described hereinbelow.

The folded corner detection circuit 314 utilizes the system timing pulse signals at input 314b and the signals of sensors 86a through 86d of sensor array 86, as well as reference level signals derived from circuit 336 to detect the presence of folded or missing corners at both the leading and trailing edges of a sheet.

Average density detection and length measuring circuit 316 utilizes the signals from all of the sensors in array 86 at input 316a and the timing pulses at input 316b, as well as a document detected signal, derived from output 312b of hole detection circuit 312, as well as reference level signals derived from circuit 336, to determine the average density over first and second halves of each sheet and to determine whether the sheet being handled is either too long, too short, or neither of the above.

Selection of the operations to be performed and adjustment of the threshold levels in accordance with the particular needs of the operator are accomplished through the keyboard and control board 340 which provides control signals to a group of inputs represented as 302f of central processing unit 302, which signals are derived from the group of lines represented as output 340a. Control signals derived from ports 302g of the central processing unit 302 are coupled through control line group 302h to inputs 340b of the keyboard and control means 340.

A group of output lines 340c are coupled from the operator adjustable controls to be more fully described to reference level circuitry 336, whereupon the sensitivity controls provided as part of the control panel array are utilized to adjust the various reference levels to suit the needs of the particular application.

The central processing unit 302 provides observable information to the operator by means of output lines represented by line group 302m, for application to the inputs 342a of display circuit 342 which, as will be more fully described hereinbelow, is capable of displaying information such as, but not limited to, the number of fit sheets which have been processed; the number of unfit sheets which have been processed; the total number of suspect documents which have been processed; the number of fit and unfit sheets within a batch, when the batching mode is selected; to name just a few of the possible displays. Outputs 302k-1 to 302k-4 reset the evaluating circuits 334, 314, 312 and 316 after sampling the results of their evaluations.

FIG. 4 shows a detailed block diagram of the lamp regulating circuitry 346 for regulating the voltage levels of the voltage applied to halogen lamp 85. In order to maximize the useful operating life of halogen lamp 85, the lamp is provided with a d.c. supply voltage which is alternated in a regular, periodic fashion. This is accomplished by the use of an "H-type" circuit 350 comprised of transistors Q4, Q5, Q6 and Q7, only two of said transistors, i.e. either Q6 and Q7 or Q4 and Q5, being conductive at any given time. The switching of transistors Q4 through Q7 is controlled by a pair of switching transistors Q3 and Q8, transistor Q3 having its collector and emitter electrodes respectively coupled to the base electrodes of transistors Q4 and Q5, transistor Q8 having its collector and emitter electrodes respectively coupled to the base electrodes of transistors of Q6 and Q7. The switching circuitry, as will be more fully described, functions so that while terminal 85a of lamp 85 is coupled to ground through transistor Q7, terminal 85b is coupled to a positive d.c. level through line 354 and transistor Q6 to terminal 356a of voltage regulator 356. Alternatively, when terminal 85b of lamp 85 is coupled through line 354, and transistor Q5 to ground, terminal 85a is coupled through line 352 and transistor Q4 to line 356a of voltage regulator 356.

The voltage regulator 356, which may for example be a type u A78H12SC is provided with a voltage at terminal 356b which is of the order of 18 volts. Voltage regulator 356 functions to maintain precisely a 12 volt difference between the voltage level at its control input 356c and its output 356a. For example, if the voltage level at input 356c is at +5 volts, the output level at output terminal 356a will be +17 volts; if the voltage level at 356c is at +4 volts, the voltage level at output 356a will be +16 volts, if the voltage level at input 356c is at -1 volt, the voltage at output 356a is +11 volts, and so forth.

The switching of the "H-type" circuit 350 is performed at a rate to reverse the polarity of the d.c. signal applied across lamp 85 at a frequency in the range from 100 Hz to 1 kHz cycles. The signals are derived from switching circuit 358 as will be more fully described hereinbelow. Due to the fact that the transistors Q4 through Q7 have storage delays, i.e. due to the fact that the transistors Q4 through Q7 are capable of turning on more rapidly than they are capable of turning off, the switching control signals are derived in such a manner as to assure that the transistor of "H-type" circuit to be switched off is switched off early and the transistor which is to be switched on is switched on at a predetermined time interval sufficient to allow the transistor which has just received a switching signal to be fully switched off before the transistor to be turned on is in fact turned on.

The switching control circuit is comprised of a clock source 360 for applying clock pulses to input 362a of an Octal Johnson type counter 362. Output 362b of counter 362 applies output pulses at the same clock rate as clock 360, to the clock input 364a of a second multi-stage binary counter 364. Clock pulses from source 360 are also applied to one input of AND gates 370 and 372.

Output 362b of counter 362 is coupled in common to one input of AND gates 366 and 368. The output 362c of counter 362, which is coupled to the last stage of counter 362, which is preferably an 8-stage counter, is coupled in common to one input of each of the AND gates 370 and 372 and applies pulses to these gates at phase delay relative to output 362b.

Output 364b of counter 364, which is coupled to the first stage of the counter, applies signals to one input of AND gate 366 and the clock input 374a of a D-type bistable flip-flop 374. Output 364c, which is coupled to the fourth stage of counter 364 applies signals to one input of AND gate 370. Output 364d, which is coupled to the fifth stage of counter 364 applies its signal level simultaneously to one input of gate 368 and to the clock input 376a of bistable flip-flop 376. Output 364e of counter 364, which is coupled to the last stage of counter 364, applies its signal level to one input of AND gate 372.

The outputs of gates 370, 372 are coupled to the reset inputs 374b and 376b of bistable flip-flops 374, 376. The outputs of gates 366 and 368 are coupled through inverters 378 and 380 to two inputs of AND gate 382. The remaining input of AND gate 382 is coupled to the collector of transistor Q1 through inverter 384 whose output is also coupled to the D-inputs 374c and 376c of bistable flip-flops 374, 376. Transistor Q1 has its collector and emitter electrodes coupled between positive 12 volts through resistor R5 and ground potential respectively. A lamp enable signal is applied to line 383 when it is desired to illuminate the lamp, causing transistor Q1 to conduct. The level at the collector of Q1 goes low, which condition is inverted at 384 to apply a high level at the D-inputs 374c, 376c of flip-flops 374, 376 and to apply a high level to AND gate 382. In the absence of a lamp enable signal, the collector of Q1 is high, which state is inverted by inverter 384 causing low levels to be applied to AND gate 382 and the D-inputs 374c and 376c, preventing these circuits from operating.

Assuming that a lamp enable signal is present, transistor Q1, through inverter 384 applies a high level to inputs 374c and 376c. Output 364b of counter 364 applies a pulse to the clock pulse input 374a of bistable flip-flop 374, causing its Q output 374d to follow the level applied at input 374c. The leading edge of the pulse 390 is thus initiated at time t.sub.o. The pulse is terminated by means of a reset input signal applied to reset input 374b by the output of gate 370 when the output 364c of counter 364 changes, developing the trailing edge at time t.sub.1.

The Q output 376d of bistable flip-flop 376 goes high upon receiving a clock pulse from output 364d of counter 364 which occurs at a predetermined time t.sub.2 after the trailing edge of pulse 390 goes low. Bistable flip-flop 376 is reset by gate 372 from output 364e of counter 364, the trailing pulse occurring at time t.sub.3. The output 374d of bistable flip-flop 374 goes high again at a time t.sub.4 after the trailing edge of square pulse 392 occurring at time t.sub.3. Time t.sub.4 actually coincides with time t.sub.o due to the reinitiated count of counter 364. Output 374d goes low again at time t.sub.5 by the reset signal from gate 370 while output 376d of bistable flip-flop 376 goes high at time t.sub.6 which is a predetermined time after time t.sub.5. These waveforms are applied to the base electrodes of transistors Q3 and Q8 to control switching of the H-type circuit 350. For example, when a high level is applied to the base of Q3, the output at the collector of Q3 drops while the level of the common point 394 between resistors R23 and R24 goes high to turn on transistor Q5. The level at the collector of Q3 turns off transistor Q4. When a low level is applied to the base of Q3, Q3 is turned off causing the level at common point 394 to go to ground and turning transistor Q5 off. The collector of Q3 goes high causing transistor Q4 to be turned on. Switching transistor Q8 controls the operation of transistors Q6 and Q7 in a similar fashion. However, the delay between leading and trailing edges of the pulses as described hereinabove provide sufficient time for transistors being turned off to experience the aforementioned storage delay and be fully turned off before the transistor in series with the lamp is turned on. Thus for example, if transistor Q4 is on and transistor Q7 is off, and the states of these transistors are to be reversed, the application of a turn-on level signal to the base of Q7 is delayed relative to the turn-off level applied to Q4 for a period sufficient to be assured that Q4 is completely turned off.

The outputs of bistable flip-flops 374 and 376, which develop the aforementioned square pulse waveforms, which waveforms have a duty cycle of less than 50% and preferably of the order of 48% (to compensate for the storage delays of transistors Q4-Q7), are also connected to respective inputs of gates 398 and 400, whose remaining inputs are coupled in common to the output of gate 382 which applies an enable signal to gates 398 and 400 during the time intervals that the associated transistors are turned on. The collector of Q1 is coupled through line 404 to the base of Q2 for the purpose of activating regulating circuit 356. The outputs of gates 400 and 398 are respectively coupled to the control inputs 408a-414a of solid state switches 408-414, and control inputs 410a-412a of solid state switches 410-412. Each of these switches respectively couple associated terminals 85a and 85b of lamp 85 to a pair of sample and hold circuits 416 and 418. For example, when terminal 58a is at a +d.c. voltage and terminal 85b is at ground, these voltage levels are applied through line 352 and resistor R19 simultaneously to one input of solid-state switches 408 and 412 and through lead 354 and resistor R20 simultaneously to one input of each of the solid-state switches 410 and 414. Terminal 85a, being at a high level d.c., gate 400 is enabled to close solid state switches 408 and 414, while gate 398 is disabled to open solid state switches 410 and 412. The sample and hold circuits 416 and 418 sample the positive d.c. voltage level and temporarily store this level. The outputs of sample and hold circuits 416 and 418 are coupled to the inverted and non-inverted inputs respectively of an amplifier 420 to develop a difference signal which is applied to the inverting input of comparator 422, where it is compared with a reference voltage level applied to the non-inverting input of comparator 422 by the adjustable arm R11a of potentiometer R11. The difference voltage is applied to control input 356c of voltage regulator 356 only during the presence of a lamp enable signal which is applied to the base of transistor Q2 through line 404 and resistor R18 to turn Q2 off and apply a regulating level to the voltage regulator input 356c. As was described hereinabove, output terminal 356a of series regulator 356 maintains a predetermined precise voltage difference between a level at output 356a and a level at regulator input 356c, in the example being given, 12 volts.

When the polarities of the voltage applied to lamp 85 are reversed, terminal 85b going positive and 85a going to ground, gate 398 closes switches 410, 412 while gate 400 opens switches 408 and 414 to apply the positive voltage level from terminal 85b through solid state switches 410 and 412 to the sample and hold circuits 416 and 418 respectively. The difference between the voltage level and the reference level, measured by comparator 422, is developed at the output of error amplifier 422 in order to appropriately adjust the output level at 356a of series regulator 356. Thus, the polarity of the signal voltage applied to halogen lamp 85 is enabled to be switched at high speed and is further enabled to be constantly regulated so as to control the voltage level within very tight tolerances.

In spite of the fact that the regulation circuit of FIG. 4 provides excellent regulation of halogen lamp 85, other changes may occur in the sensing array due to the aging of components, accumulation of dirt and dust in the region of the optical path between light source 85 and sensor array 86, and the like.

In order to automatically compensate for such changes, be they gradual or abrupt, the sensors in sensor array 86 undergo initial amplification by amplifier circuit 324 and thereafter are coupled to an automatic gain control circuit 326 for automatically and precisely regulating the output level of the sensor signals by comparison with a gain reference voltage applied at input 326e. The automatic gain control circuit 326 is shown in greater detail in FIG. 5 and is comprised of a gate 434 which is enabled when signals from preview sensor 97 and a document detected signal is developed at output 312b of hole detection circuit 312 to enable gate 436 which passes clock pulses to the clock pulse input of a bistable flip-flop 438. The interval during which an automatic gain control adjustment is performed is the time interval measured between the passing of the trailing edge of a document beyond sensor array 86 and the movement of the leading edge of the next document over preview sensor 97. The Q output of bistable flip-flop 438 goes high causing gate 440 to pass one clock pulse from oscillator 435 through inverter 422 to gate 440, resetting the bistable flip-flop. The Q output of bistable flip-flop 438 is also coupled to the reset input 462b of a multi-stage counter 462 which is an integral part of the automatic gain control circuit 446 for sensor 86a. Automatic gain control circuits 448, 450 and 452 have been shown by black boxes simplifying FIG. 5, it being understood that each of the automatic gain control circuits 448 through 452 are identical to automatic gain control circuit 446.

Automatic gain control circuit (hereinafter AGC circuit) 446 also includes gate 454 which is enabled, when the Q output of bistable flip-flop 444 is high, to pass one pulse from clock pulse source 435 through gate 445. The other input of gate 454 is high, so long as the output of comparator 460 has yet to indicate that the gain adjustment compares with the desired reference level. The output of bistable flip-flop 444 goes high upon the switching of bistable flipflop 438 as was described hereinabove. As a result, clock pulses are passed from clock pulse course 435 through gate 445 and 454 to the clock pulse input 462a of multi-stage counter 426. The output of each stage of counter 462 is coupled to the control terminal 464a-470a of an associated solid state switch 464 through 470. Each switch is coupled in a branch circuit with a resistance element R.sub.A through R.sub.D. A final resistance element R.sub.E forms a complete branch circuit. The sensor signal of sensor 86a is applied to the inverting input of operational amplifier 474. The output of operational amplifier 474 is fed back to the input through one or more of the branch circuits containing resistors R.sub.A through R.sub.E, resistor R.sub.E always being in the feedback circuit.

The clock pulses are applied to clock input 462a and when a reset reference level is removed from reset input 462b, counter 462 starts to count from a zero count toward a maximum count wherein switches 464 through 470 are closed in a predetermined sequence, switch 464 closing, thereafter switch 466 closing as 464 opens, switches 464 and 466 both closing, and so forth. The resistance values of resistors in the branch circuits are weighted in that resistor R.sub.D is double the value of resistor R.sub.E, resistor R.sub.C is double the resistance value of resistor R.sub.D, and so forth. The output of operational amplifier 474 is a function of the feedback impedance coupled between its output 474b and its input 474a divided by the input resistor R.sub.F, whereupon the voltage at the output is regulated by selective coupling of the resistors in the aforementioned branch circuits. The counter 462 is operated in such a manner as to begin at a zero count and count-up in order to decrease the gain from a maximum value at the output of operational amplifier 474, as the count in counter 462 is advanced from a zero count toward the maximum count.

Comparator 460 compares the signal level at the output of operational amplifier 474 with the reference level signal developed across the voltage divider circuit comprised of resistors R.sub.G, R.sub.H and R.sub.J and is coupled to comparator 460 through adjustable arm 476. As soon as the voltage level at the output of operational amplifier 474 equals the reference level, the output of comparator 460 goes low to block gate 454 from passing clock pulses from clock pulse source 435 through gates 445 and 454.

This gain control level is maintained until the next time an adjustment is performed, at which time the operation described hereinabove is repeated. Gain control adjustments are made between each interval during the passage of the trailing edge of a document and the passage of the leading edge of the next succeeding document in the region of sensor array 86 and preview sensor 97.

As was mentioned hereinabove, automatic gain control circuits 448 through 452 operate in substantially the identical manner. The output of comparator 460 is also coupled to one input of gate 480 which resets bistable flip-flop 444 to block gate 445 from supplying pulses to each of the counters forming part of their associated automatic gain control circuits. It should be understood that gate 480 resets bistable flip-flop 444 only after all of the reference levels have been adjusted.

Although the automatic gain control circuit 446 shows a 4-stage counter for selectively closing solid state switches 464-470 provided in four associated branch circuits, it should be understood that the total number of branch circuits may be increased or decreased and accordingly that the number of stages of counter 462 may be increased or decreased depending upon the particular accuracy level desired.

Once an adjustment is made, the count in each counter 462 is maintained until at least the next adjustment is made. The outputs of each of the operational amplifiers 460 provided in each automatic gain control circuit are simultaneously applied to associated inputs of the hole detection circuit 312 (see FIGS. 2 & 7), folded corner detection circuit 314 (FIGS. 2 and 8) and average density and length measuring circuit 316 (FIGS. 2 & 9).

A detailed diagram of the hole detection circuit 312 is shown in FIGS. 7a and 7b.

A separate hole detection circuit 490 through 496 is provided for each of the four sensors in sensor array 86. Since the circuits are substantially identical in design and function, only one of the circuits will be described in detail, for purposes of simplicity. To further simplify FIGS. 7a and 7b, hole detection circuits 492 through 496 have been shown in amplified black box form.

The signal of sensor 86a is simultaneously applied to one input of hole comparator 514 and to one input of an amplitude limiter circuit comprised of operational amplifier 498, the remaining input being coupled to the output 536a of an operational amplifier 536 forming part of adjustable amplitude limited adjustment circuit 504. The amplitude limited signal output of amplifier 498 is coupled to terminal 506b of the diode bridge 506 by operational amplifier 500 which is operating as a voltage follower comparator. The signal level applied to terminal 506b is limited in accordance with the adjustment of the amplitude limit adjustment circuit. Comparator 502 operates as an inverter and compares the output of comparator 514 with the output 536a of operational amplifier 536. The increase in light intensity due to the presence of a hole causes comparator 502 to develop an output which, when applied to input 506a, causes the slew rate of the slew rate limiter 506 to be further reduced. Terminal 506c of diode bridge 506 is coupled to one input of operational amplifier 508 which, together with capacitor C1 and bridge circuit 506, forms a slew rate limitor circuit which limits the rate at which the output of operational amplifier 508 can follow a positive going signal, having a very sharp leading edge.

The output of the slew rate limitor circuit 506 is coupled to one input of operational amplifier 510 and to output line 512 for a purpose to be more fully described. Automatic threshold circuit 510 raises the level of the slew rate limiter circuit by a predetermined value, provided by operational amplifier 536 and applies this signal to one input of comparator 514. The other input of comparator 514 is directly coupled to receive the sensor level signal through line 516. The output of comparator 514 is coupled to one input of gate 518 whose remaining inputs are coupled to the output of gate 520 and hole detection inhibit line 521 to permit gate 518 to pass the signal representing a detection of a hole only when the hole detection circuit is not inhibited and when less than all of the hole detection circuits indicate the presence of a hole. When all of the hole detection circuits 490 through 496 detect maximum brightness, this is interpreted as the passage of the trailing edge of a sheet beyond the sensor array 86. Under this condition, gate 520 inhibits all of the gates 518 and 522 through 526. When less than all of the hole detection circuits indicate the presence of maximum brightness, this condition is interpreted as a hole or tear in the sheet enabling all of the gates 518 and 522 through 526 to pass a hole detection signal if one is present.

The outputs of gates 518 through 526 are coupled through inverters 527-533 to the reset inputs 538a of digital sample and hold circuits 528 through 534 respectively. Since all of the digital sample and hold circuits are substantially identical in design and function to one another, digital sample and hold circuits 530 through 534 have been shown in black box form and only a description of digital sample and hold circuit 528 will be given herein for purposes of simplicity.

The signal developed by gate 520 is utilized in detection circuit 316 as will be more fully described and is also utilized by a central processing unit 302 for developing a count of the documents being handled.

The operation of the hole detector circuit may be better understood from consideration of the waveform diagrams of FIGS. 7c through 7d. FIG. 7c is a waveform diagram showing the signal level developed by a piece of paper currency. At time t.sub.o, the output of the sensor is high indicating the absence of a document. At time t.sub.1, the leading edge of the document passes over sensor array 86 causing the output level of the associated sensor to drop considerably. The output level remains relatively low until, at time t.sub.2, when the trailing of the document passes array 86, the output of the sensor goes to maximum.

Assuming that a fairly large hole is present in the paper document, the sensor level output abruptly goes high at time t.sub.1 +A, stays high for a period of time and then drops at time t.sub.1 +B back to a lower output level. A smaller hole is represented by the pulse occurring during time t.sub.1 +C and t.sub.1 +D, the holes H1 and H2 being shown immediately beneath these pulses. An extremely small hole H3 is represented by the small pulse occurring between time t.sub.1 +E and time t.sub.1 +F.

The signal applied to circuit 490 by sensor 86a is thus represented by the waveform of FIG. 7c.

The amplitude limiter circuit limits the output of the sensor for example, to the threshold level Th1 shown in FIG. 7c, limiting any signal to a maximum of the threshold level. The resulting waveform at the output of the amplitude limiter circuit is thus shown in FIG. 7d.

The slew rate limiter limits the rate at which the signal appearing at the output of operational amplifier 508 can follow the actual signal being developed by sensor 86a. Thus the slew rate limiter cannot follow the rapid change in the leading edge of pulse P1 and builds up at a slower rate to a signal level represented by the sloping portion 530' of altered pulse P1'. The slew rate limiter is, however, able to follow the rapid negative change in the signal, as shown by waveform 7d. The slew rate limiter yields similar results for the pulses P2 and P3, as shown by the sloping portions 532' and 534' of pulses P2' and P3' respectively, shown in FIG. 7d.

Considering the waveform A of FIG. 7e, the waveform of 7c, after undergoing slew rate limiting, has its d.c. level adjusted upwardly and thereafter applied to one input of comparator 514. The other input of comparator 514 receives the original sensor signal applied at input 490a and represented by waveform B shown in FIG. 7e. As can clearly be seen, waveform A is at a higher level than waveform B over most of the period during which a document is passing over the sensor array. However, waveform B can be seen to exceed waveform A in instantaneous signal level at time t.sub.x, which results from the inability of the output of the slew rate limiter circuit to follow abrupt positive going changes. When waveform B exceeds waveform A, comparator 514 develops a hole detected signal which is passed by gate 518. This signal remains high, as represented by the pulse of FIG. 7f, until the instantaneous level of waveform B drops below the instantaneous level of waveform A, the duration of the pulse shown by the waveform of FIG. 7e representing the length of the hole or opening in the document measured in the feed direction.

The pulse passed by gate 518 is applied to the reset input 538a of multistage counter 538 which forms part of the digital sample and hold circuit 528. In the absence of a reset input signal, the timing pulses, developed by sensor 120 (see FIG. 2) which are coupled to clock input 538b of counter 538, are passed to allow counter 538 to count up from a zero count toward a maximum count. Each output stage of counter 538 has its output line 538c through 538g coupled in a branch circuit having a branch circuit resistor R1 through R5 respectively. All of the resistors R1 through R5 are coupled to a common terminal 542, which is coupled to a +12 volt source through resistor R6 and to one input of an operational amplifier 544. Each of the branch circuits contain resistors R1 through R5 which have weighted resistance values, wherein the value of resistance R2 is one half that of R1, the value of resistance R3 is one half that of R2, and so forth. The amount of current each of these resistances delivers to one input of comparator 544 continually increases depending upon the number of resistances in the circuit. As the count in counter of 538 is incremented, the voltage level of output terminal of operational amplifier 544 drops from a maximum value is a staircase fashion, as represented by the descending staircase waveform shown in FIG. 7b and in FIG. 7g. The level at the output of operational amplifier 544 follows the level at input 542. The staircase waveform is applied to one input of comparator 546 whose remaining input is coupled to the voltage divider circuit comprised of potentiometer 547 having adjustable switch arm 547a which is manipulated by means of a control knob provided on the control panel of FIG. 11, as will be more fully described hereinbelow.

As soon as the level of the staircase signal drops below the threshold level applied to the noninverting input of comparator 546, which threshold level Th2 is shown in FIG. 7g, comparator 546 generates a pulse as shown by waveform E of FIG. 7g. This pulse is applied to the clock input 548a of D-type bistable flip-flop 548 whose D input 548b is maintained at a high level. The Q output 548c follows the level at the D input 548b upon receipt of the aforesaid clock pulse as shown by waveform E of FIG. 7g. The bistable flip-flop 548 temporarily stores the hole condition detected for purposes of "remembering" the hole condition until such time that the central processing unit 302 is free to examine the state of the bistable flip-flop 548, which occurs at a particular time during a sample routine performed by the central processing unit 302. Similar bistable flip-flops 548'-548'" are provided for each of the other sensor units. Adjustment of the threshold level applied to comparator 546 allows extremely small holes to be ignored, if desired. Alternatively, the adjustment may be made quite sensitive to detect even small holes. The sensitivity of the system is designed to detect holes, tears, perforations and the like having a dimension of the order of 0.04 inches, or less, measured in the sheet feed direction.

Gate 552 shown in FIG. 7a derives signals from all of the hole detection circuits 490 through 496 to provide a signal indicating the presence of a hole by at least one of the sensors, regardless of the size of the document hole, which information is made available to the density circuit 316 (FIG. 9) for processing in a manner to be more fully described.

The average density detection and length measuring circuit 316 is shown in FIG. 9. The signals from all of the sensors 86a through 86d are summed by operational amplifier 570. The output of operational amplifier 570 is simultaneously applied to the noninverting input of comparator 572 and the inverting input of comparator 574. The signal level is compared against the "light" reference threshold represented by an internally adjustable potentiometer 576 having adjustable arm 576a. The signal level is also compared against a "dark" reference threshold represented by internally adjustable potentiometer 578 having adjustable arm 578a. The output of comparator 572 is applied to one input of AND gate 580. The remaining input of AND gate 580 receives the document hole signal at the output of gate 552, shown in FIG. 7a, in order to prevent the presence of a document hole from being interpreted as a document having density lighter than the light density threshold. The output of comparator 574 is applied to one input of gate 582, the remaining input being coupled to the output of gate 580. The output of gate 582 is applied to one input of gate 584, whose remaining input receives double frequency timing pulses from the timing source, said timing source being coupled to the input of frequency doubler 586 whose output is coupled to the remaining input of gate 584. Thus, when the density of the document is either lighter than the light threshold or darker than the dark threshold, double frequency timing pulses are applied through gate 584 to gate 588, whose remaining input is coupled to the output 590e of bistable flip-flop 590, to be more fully described, for the purpose of inhibiting the application of timing pulses to the counter 594 when the marginal portions of the leading and trailing edges of the sheet being examined are passing the sensor array 86.

Comparators 650 and 664 provide density detection inhibit signals to inhibit density signals when the marginal portions of the leading and trailing edges of a sheet are passing the sensors. The descending staircase voltage signal developed at the output 660a of operational amplifier 660 due to the count being developed by counter 618, is applied to the inverting input of comparator 650 for comparison against a threshold level applied to the non-inverting input of comparator 650 by resistor R37. As the leading edge of a sheet moves over sensor array 86, maximum staircase voltage is applied to comparator 650. The staircase voltage drops as the leading edge of the sheet moves past the sensor array and a comparison occurs when the marginal portion of the leading edge has passed the sensor array 86. The change in output level of comparator 650 is coupled through gate 668 to the reset input 590a of bistable flip-flop 590. The output level of gate 668 changes when the output level of comparator 650 changes relative to the level of the document detected signal applied to the remaining input of gate 668, causing output 590e of flip-flop to change, thereby removing a level which blocked gate 588 from passing double frequency timing pulses from frequency doubler 586, enabling counter 594 to accumulate timing pulses whenever a "too light" or "too dark" condition is present for an area of the sheet beyond the leading marginal portion.

Counter 594 is reset at the leading and trailing edge of each sheet and at the mid-length point of each sheet by comparator 598, a timing circuit 671 comprised of resistor R.sub.T1 and capacitor C.sub.T1 and Exclusive-OR gate 669. When a leading edge of a sheet passes sensor array 86, counter 618 is reset, causing the output level of comparator 598 to change. This level is applied to timing circuit 671 and one input of gate 669. Timing circuit 671 delays the signal level transition applied to the remaining input of gate 669, which develops a pulse whose pulse width is determined by the delay of timing circuit 671. Counter 594 is thus reset in readiness for receipt of timing pulses when gate 588 is enabled.

When the staircase signal reduces to a level equal to the reference level applied to comparator 598 by resistor R38, indicating the mid-length point of a sheet passing over sensor array 86, the output level of comparator 598 shifts, causing timing circuit 671 and gate 669 to apply a pulse to reset input 594b, resetting the counter 594 in readiness for a density measurement for the last half of the sheet.

Presuming that the aforesaid marginal portion has now passed through the sensor array 86, the reset level is removed and double frequency clock pulses are applied to the clock input 594a whenever a too light or too dark condition is detected, causing counter 594 to step from a zero count towards a maximum count. The outputs of counter 594 are each coupled into a branch circuit having a resistor R10 through R15 wherein the resistors are weighted so that the resistance value of R11 is one half the resistance value of R10, R12 is one half the value of R11, and so forth. All of the resistors have their opposite ends coupled in common to a 12-volt source through a resistor R16. The values of R10-R15 are chosen so that the parallel combination of resistors R10-R15 is substantially equal to R16. The common terminal between resistors R10 through R15 and R16 is coupled to the inverting input of operational amplifier 596 whose output is coupled to the inverting input of comparator 600 whose noninverting input is coupled to the density sensitivity potentiometer 602. Adjustable arm 602a is coupled to the non-inverting input of comparator 600 and is adjustable at the front control panel, to be more fully described in connection with FIG. 11. When the staircase signal developed by counter 594 and resistors R10 through R16 reaches a predetermined threshold, the output of comparator 600 applies a signal simultaneously to one input of gates 604 and 606. Gate 604 is enabled by the Q output of D-type flip-flop 610 which is clocked by comparator 598.

The output of comparator 598 is applied to the D input 610a of flip-flop 610, directly to one input of Exclusive-OR gate 667 and to the other input of gate 667 through the delay circuit 673 comprised of resistor R.sub.T2 and capacitor C.sub.T2. Delay circuit 673 and gate 667 function in a manner similar to gate 669 and delay circuit 671 and cooperate with inverter 675, causing Q output 610c to enable gate 604 during the time that the first half of a sheet is being examined and thereafter causing Q output 610d to enable gate 606 during the time that the last half of a sheet is being examined. If the count accumulated by counter 594 is sufficient to cause the staircase voltage level to drop below the threshold established by potentiometer 602, the dense condition is clocked into the appropriate flip-flop 612 or 682.

The outputs of gates 604 and 606 are respectively coupled to the clock inputs 612a, 682a of D-type bistable flip-flops 612, 682 for storing a density condition until such time that the central processor 302 interrogates the average density detection circuitry.

The document detected signal developed by the hole detection circuitry 312 of FIGS. 7a and 7b is utilized to measure document length and appears at input 614. The document detected signal is simultaneously applied to the reset inputs 616b and 618b of digital counters 616 and 616. The timing pulses are divided by bistable flip-flop 620 and applied to clock input 616a of digital counter 616, and are applied directly without division to the clock input 618a of digital counter 618. Counters 616 and 618 and their associated resistor circuits operate in much the same way as counter 594 and its associated resistor circuit.

The outputs of each binary stage of digital counter 616 are coupled to branch circuits each containing a resistor R17 through R23 whose opposite ends are connected in common to the inverting input of operational amplifier 624 and to the +12-volt source through resistor R24. As the counter 616 counts up from a zero count, a negative going staircase signal is developed by operational amplifier 624. This level is compared against a reference level derived from reference level circuit 630, comprised of a series circuit including fixed resistors R25 and R26 and internally adjustable potentiometers 632 and 634 having adjustable arms 632a and 634a which are preferably factory adjusted to set the shortest length of an acceptable document and the longest length of an acceptable document. The inputs of operational amplifiers 638 and 640 are coupled to adjustable arms 623a and 634a and their outputs coupled across end terminals of a potentiometer 642 whose adjustable arm 642a is coupled to the noninverting input of comparator 644 through fixed resistor R27a. Arm 642a is also connected to the noninverting input of comparator 648 and across the series connected resistors R37, R38 and R39. The common terminal between resistors R37 and R38 is coupled to the noninverting input of comparator 650, while the common terminal between resistors R38 and R39 is coupled to the inverting input of comparator 598.

The output of operational amplifier 644 is coupled through line 654 to the clock input of a bistable flip-flop 656.

The counter 618 has one output of each of its counter stages coupled to resistors R28 through R34 which are connected in common to the inverting input of operational amplifier 660, whose output is coupled to the inverting inputs of comparators 648, 650 and 664 and to the noninverting input of comparator 598. The output of comparator 648 is coupled to the clock input of bistable flip-flop 666. The output of comparator 598 is coupled to: one input of logical gates 667 and 669; delay circuits 671 and 673; and the D input 610a of flip-flop 610. The output of gate 668 is coupled to the clock input 610b of bistable flip-flop 610. The output of comparator 664 is further coupled to the clock input 590c of bistable flip-flop 590.

Bistable flip-flop 612 has its D input 612b coupled to the Q output 610c of bistable flip-flop 610 whose D input 610a is coupled to the output of operational amplifier 598 as was described above. The clock input 612a of bistable flip-flop 612 is coupled to the output of gate 604 whose inputs are respectively coupled to the Q output 610c of bistable flip-flop 610, and to the output of comparator 600. The reset input 612c of bistable flip-flop 612 as well the reset inputs of bistable flip-flops 682, 684, 656 and 666 are coupled to line 686 which receives an average data clear signal from the central processing unit 302 shown in FIG. 3.

The D input of bistable flip-flop 682 is coupled to the Q output of flip-flop 610. The clock input of flip-flop 682 is coupled to the output of gate 606, whose inputs are coupled to the Q output 610d of flip-flop 610 and the output of comparator 600. Delay circuits 671 and 673 are designed to cause counter 594 to be reset before bistable 610 is reset.

The operation of the average density and length measurement circuitry is as follows:

With respect to the density measuring circuit, the outputs of sensors 86a through 86d are summed, and applied to the noninverting input of operational amplifier 572, whose output is coupled to the inverting input of comparator 572 and the inverting input of comparator 574. Light and dark reference levels are applied to the inverting input of comparator 572 and the noninverting input of the comparator 574, respectively. In the event that the dark reference threshold is achieved, comparator 574 alters its output level, enabling gate 584 to pass timing pulses on line 697 to apply clocking pulses to counter 594 when gate 588 is enabled by flip-flop 598 when the marginal portion of the leading edge of a sheet has passed the sensor array 86. A reset level signal is applied to counter 594 when the leading edge of a sheet has passed sensor array 86 to reset counter 594 in readiness to accumulate clock pulses. As each stage of the counter 594 is enabled, a descending staircase voltage is developed at the output of operational amplifier 596. This negative going staircase voltage appearing at the output of operational amplifier 596 is applied to the inverting input of comparator 600. As soon as the signal level at the inverting input of comparator 600 drops below the reference level applied to the noninverting input of comparator 600, gates 604 and 606 are enabled. Gate 606 is the only gate able to pass a level to bistable flip-flop 682 during the last half of the examination of a sheet. The reversal of state of the bistable flip-flop 610, whose Q output 610c is coupled to gate 604 and whose Q output 610d is coupled to gate 606 alternately enables only one of these two gates. When a signal indicating that the sheet being examined is too dense (or too light) during the first half of a sheet, this condition is stored in bistable flip-flop 612 preparatory to subsequent sampling by the central processing unit 302. Bistable flip-flop 682 stores the same condition when present during the second half of examination of a sheet.

The document length measuring circuitry is initiated upon the occurrence of a document detected signal from gate 520 of FIG. 7a, at which time a reset signal is applied to reset input 618b of counter 618 and to reset input 616b of counter 616, resetting these counters to zero. Clock pulses from the divide-by-two clock pulse source and appearing on line 710, are applied to clock input 618a of counter 618. Bistable flip-flop 620 divides the already divided by two clock pulses on line 710 before applying pulses to the clock pulse input 616a of counter 616.

Clock pulses are accumulated by counter 618, whose output stages are selectively connected by resistors R28 through R34 to the inverting input of operational amplifier 660 which applies a descending staircase signal to the inverting inputs of comparators 648, 650 and 664 and to the noninverting input of comparator 598. The counter 616 cooperates with resistors R17 through R23 and resistor R24, to develop a descending staircase signal which appears at the output of operational amplifier 624. The Q output of bistable 666 is normally high. If the sheet is the proper length, the descending staircase signal of operational amplifier 660 drops below the threshold level applied to comparator 648 by resistor R41, and bistable flip-flop 666 is triggered by a clock pulse to remove the length short condition. When the descending staircase developed by operational amplifier 660 drops below another threshold level coupled thereto by resistor R38, comparator 598 retriggers reset input 594b of counter 594 and causes reversal of the Q and Q outputs of bistable flip-flop 610, indicating that one half of a sheet has been examined and examination of the second half of the sheet should now be initiated.

When the descending staircase at the output of comparator 660 falls below a third threshold level provided by operational amplifier 667, and resistors R37-R40 to comparator 664, comparator 664 clocks bistable flip-flop 590 to provide a density detection inhibit signal at output 590e to inhibit the accumulation of timing pulses in counter 594 during the time that the marginal portion of the trailing edge of a sheet is passing sensor array 86. Density measurement is inhibited at this time due to the high transmissivity of the marginal portions of the sheet.

Counter 616 is utilized to develop a signal indicative of the fact that the sheets being examined are too long. The descending staircase signal is developed by operational amplifier 624 and applied to the inverting input of comparator 644 where it is compared against the reference level applied to the noninverting input. Comparator 644 applies a trigger signal to the clock pulse input of bistable flip-flop 656 through line 654 when a document of too great a length is measured, this condition being temporarily stored in bistable flip-flop 656 for subsequent sampling by the central processing unit 302.

Bistable flip-flop 684 develops a data valid condition and its Q output when a document detected signal is terminated. The microprocessor 800 examines the state of bistable 684 and is assured that all conditions in bistables 612, 682, 656 and 666 are valid since the sheet has passed the sensor array 86, i.e., has left the system optics 86.

The central processing unit 302 samples the results of the document length and average density measurement circuitry and upon transfer of said information to the central processing unit 302, applies an average data clear signal to the reset inputs of bistable flip-flops 612, 682, 684, 656 and 666 in readiness for subsequent operation.

The limpness detector circuitry 332 shown as a black box in FIG. 2 is shown in detailed block diagram form in FIG. 6 and is comprised of automatic zero interval timer 902 which derives timing pulses from source 118 (FIG. 2) at input line 902a and the aforementioned document detected signal at input line 902b. The automatic zero interval timer 902 provides an interval during which an offset adjustment can be made.

When the trailing edge of a sheet passes the sensing means 86 (FIG. 2), the signal generated causes timer 902 to begin accumulating pulses. When the timer reaches a first count, which represents the quotient of the length of the path between sensors 86 and limpness detector 142 and the speed of movement of the sheets, a gap interval, during which no sheets are moving between the rollers of the limpness detector 142, is initiated. When the interval timer 902 reaches a second count, representing the end of a gap, the zero interval is terminated. This interval is utilized to make an offset adjustment only when no sheets are present in the limpness detector 142, i.e. during the time interval when the trailing edges of the last sheet has passed through the limpness detector apparatus 142 and before the leading edge of the next sheet enters into the limpness detector. The pulses representing the beginning and the end of the interval are supplied to automatic zero correction circuit 904. The hall-effect sensor 330, whose output is a function of the displacement between the cooperating gear-like rollers of limpness detector 142, applies its output to the input of preamplifier 908, whose output is simultaneously coupled to the inverting input of difference amplifier 910 and to input 904b of automatic zero correction circuit 904. The sensor 330 may be subject to some drift, for example, due to aging.

Automatic zero correction circuit 904 determines the offset adjustment to compensate for the drift of sensor 330 by averaging the output of sensor 330 and storing this value to provide a d.c. offset voltage for the output of amplifier 908 and its output 904c, which is applied to the noninverting input of difference amplifier 910.

The output of difference amplifier 910, which reduces the output of sensor 330 by the d.c. offset voltage present at the output 904c is coupled through amplifier 912 simultaneously to one input of comparator 914, the input 916a of synchronous average processor 916 and leading edge overshoot compensation circuit 918.

The analog signal appearing at the output of amplifier 912 is compared against the voltage reference 915 by comparator 914 to generate a document detected signal when the analog signal is greater than the voltage reference to indicate that a sheet is passing through the limpness detector apparatus, developing the square pulse signal 920 which persists during the time that the sheet passes through the limpness detector apparatus 142.

Synchronous average processor 916 is enabled, upon the presence of a document detected signal at its input 916c and measures the average value of the limpness waveform, which may vary over the length of the sheet, said average value being totally independent of feed speed. When the value of the limpness analog signal appearing at input 916a is above a predetermined threshold, the processor 916 is enabled to accumulate each timing pulse applied at input 916b. The processor 916 may comprise a counter and an analog to digital converter which circuits may be of the type shown as the counter 616 and cooperating staircase generator comprised of resistors R17-R24 and operational amplifier 624 shown in FIG. 9. The average value count is converted into analog form at output 916d which is simultaneously applied to respective inputs of comparators 922 and 924 which compare the limpness value against voltage reference levels applied to the remaining inputs of comparators 922 and 924. In the event that the limpness value is less than the lower reference level, comparator 922 applies a pulse to the clock input 928a of D-type flip-flop 928 which stores an indication that the sheet presently being examined is too limp.

Comparator 924 compares the average of the signal developed at the output 916d against an upper threshold level. When the average value of the limpness signal exceeds the upper threshold level, comparator 924 applies a pulse to the clock input 930a of D-type flip-flop 930, storing an indication that the sheet just examined is too stiff.

As was mentioned hereinabove, the limpness analog signal appearing at the output of amplifier 912 is applied to leading edge overshoot compensation circuit 918 which attenuates the limpness analog signal by a predetermined amount to compensate for the abrupt displacement experienced by the hall-effect sensor 330 relative to the permanent magnet member 328 (see FIG. 2), during the time that the leading edge of a sheet enters the limpness detector. The compensation circuit 918 comprises an R-C circuit which is coupled between amplifier 912 and detectors 936 and 938 when the leading edge of a sheet enters the limpness detector and from 5 to 15 milliseconds thereafter, to attenuate the signal developed by sensor 330, which experiences an abrupt change as the sheet first enters the limpness detector.

The compensated limpness analog signal is simultaneously applied to positive peak detector circuit 936, negative peak detector circuit 938 and one input of each of the comparators 940 and 942.

The output of positive peak detector 936 is compared against the limpness analog signal by differentia