Automatic Duplex Control System for a Reproduction Machine

This invention relates to electrostatographic xerographic type reproduction machines, and more particularly, to an improved control system for such machines.

The advent of higher speed and more complex copiers and reproduction machines has brought with it a corresponding increase in the complexity in the machine control wiring and logic. While this complexity manifests itself in many ways, perhaps the most onerous involves the inflexibility of the typical control logic/wiring systems. For as can be appreciated, simple unsophisticated machines with relatively simple control logic and wiring can be altered and modified easily to incorporate changes, retrofits, and the like. Servicing and repair of the control logic is also fairly simple. On the other hand, some modern high speed machines, which often include sorters, a document handler, choice of copy size, multiple paper trays, jam protection and the like have extremely complex logic systems making even the most minor changes and improvements in the control logic difficult, expensive and time consuming. And servicing or repairing the machine control logic may similarly entail substantial difficulty, time and expense.

To mitigate problems of the type alluded to, a programmable controller may be used, enabling changes and improvements in the machine operation to be made through the expediency of reprogramming the controller. However, the control data which operates the machine and which is stored in the controller memory pending use, must be transferred to the various machine components at the proper time and in the correct sequence without unduly interfering with or intruding unnecessarily upon the other essential functions and operations of the controller.

The present day reproduction machine may include a variety of features such as the ability to make two-sided or duplex copies, as well as sorting individual sets of copies for easy retrieval by the user. The present invention is directed to a control system for automatically producing duplex copies and for collating sets of copies in proper numerical order. It was discovered that special consideration must be given to the control of the machine parameters when the number of original documents to be copied is odd. Otherwise unnecessary duplication of machine cycles may be required thereby resulting in the degradation of the speed of the machine.

Therefore, it is the primary object of this invention to provide a control system for a reproduction machine to permit duplex copying automatically, while at the same time optimizing the throughput capability of the machine.

This and other objects of the invention are accomplished by the interactive control of various machine components to provide automatic duplex copying capabilities, preferably under the command of a digital computer. The machine includes an automatic document handler (ADH) for locating originals to be copied on an exposure platen whereby images are subsequently formed on a photoreceptor. The images are transferred to copy sheets and fused to form finished copies on at least one side. If duplex copies are desired, a deflector mechanism prevents the finished copies from proceeding to an output receptacle such as an output tray or sorter, but directs them back to a container from which the copy sheets are subsequently fed to transfer another image on the back or second side of the sheets. If the last original document to be copied is odd in number the deflector is inhibited by the control system so that the finished copies proceed directly to the output receptacle without being fed back to unnecessarily complete the entire duplex copying feed back cycle.

Other objects and advantages will be apparent from the ensuing description and drawings in which:

FIG. 1 is a schematic representation of an exemplary reproduction apparatus incorporating the control system of the present invention;

FIG. 2 is a vertical sectional view of the apparatus shown in FIG. 1 along the image plane;

FIG. 3 is a top plane view of the apparatus shown in FIG. 1;

FIG. 4 is an isometric view showing the drive train for the apparatus shown in FIG. 1;

FIG. 5 is an enlarged view showing details of the photoreceptor edge fade-out mechanism for the apparatus shown in FIG. 1;

FIG. 6 is an enlarged view showing details of the developing mechanism for the apparatus shown in FIG. 1;

FIG. 7 is an enlarged view showing details of the developing mechanism drive;

FIG. 8 is an enlarged view showing details of the developability control for the apparatus shown in FIG. 1;

FIG. 9 is an enlarged view showing details of the transfer roll support mechanism for the apparatus shown in FIG. 1;

FIG. 10 is an enlarged view showing details of the photoreceptor cleaning mechanism for the apparatus shown in FIG. 1;

FIG. 11 is an enlarged view showing details of the fuser for the apparatus shown in FIG. 1;

FIG. 12 is a schematic view showing the paper path and sensors of the apparatus shown in FIG. 1;

FIG. 13 is an enlarged view showing details of the copy sorter for the apparatus shown in FIG. 1;

FIG. 14 is a schematic view showing details of the document handler for the apparatus shown in FIG. 1;

FIG. 15 is a view showing details of the drive mechanism for the document handler shown in FIG. 14;

FIG. 16 is a block diagram of the controller for the apparatus shown in FIG. 1;

FIG. 17 is a block diagram of the controller CPU;

FIG. 18a is a block diagram showing the CPU microprocessor input/output connections;

FIG. 18b is a timing chart of Direct Memory access (DMA) Read and Write cycles;

FIG. 19a is a logic schematic of the CPU clock;

FIG. 19b is a chart illustrating the output wave form of the clock shown in FIG. 19a;

FIG. 20 is a logic schematic of the CPU memory;

FIG. 21 is a logic schematic of the CPU memory ready;

FIGS. 22a, 22b, 22c are logic schematics of the CPU power supply stages;

FIGS. 23a and 23b comprise a block diagram of the controller I/O module;

FIG. 24 is a logic schematic of the nonvolatile memory power supply;

FIG. 25 is a block diagram of the apparatus interface and remote output connections;

FIG. 26 is a block diagram of the CPU interface module;

FIG. 27 is a block diagram of the apparatus special circuits module;

FIG. 28 is a block diagram of the main panel interface module;

FIG. 29 is a block diagram of the input matrix module;

FIG. 30 is a block diagram of a typical remote;

FIG. 31 is a block diagram of the sorter remote;

FIG. 32 is a view of the control console for inputting copy run instructions to the apparatus shown in FIG. 1;

FIG. 33 is a flow chart illustrating a typical machine state;

FIG. 34 is a flow chart of the machine state routine;

FIG. 35 is a view showing the event table layout;

FIG. 36 is a chart illustrating the relative timing sequences of the clock interrupt pulses;

FIG. 37 is a flow charge of the pitch interrupt routine;

FIG. 38 is a flow chart of the machine clock interrupt routine;

FIGS. 39a and 39b comprise a flow chart of the real time interrupt routines;

FIG. 40a, 40b, 40c are a timing chart of the principal operating components of the host machine in an exemplary copy run; and

FIG. 41 is a flow chart of the routines for providing automatic control of various machine components during duplex copying.

Referring particularly to FIGS. 1-3 of the drawings, there is shown, in schematic outline, an electrostatic reproduction system or host machine, identified by numeral 10, incorporating the control arrangement of the present invention. To facilitate description, the reproduction system 10 is divided into a main electrostatic xerographic processor 12, sorter 14, document handler 16, and controller 18. Other processor, sorter and/or document handler types and constructions, and different combinations thereof may instead by envisioned.

Processor 12 utilizes a photoreceptor in the form of an endless photoconductive belt 20 supported in generally triangular configuration by rolls 21, 22, 23. Belt supporting rolls 21, 22, 23 are in turn rotatably journaled on subframe 24.

In the exemplary processor illustrated, belt 20 comprises a photoconductive layer of selenium, which is the light receiving surface and imaging medium, on a conductive substrate. Other photoreceptor types and forms, such as comprising organic materials or of multi-layers or a drum may instead be envisioned. Still other forms may comprise scroll type arrangements wherein webs of photoconductive material may be played in and out of the interior of supporting cylinders.

Suitable biasing means (not shown) are provided on subframe 24 to tension the photoreceptor belt 20 and insure movement of belt 20 along a prescribed operating path. Belt tracking switch 25 (shown in FIG. 2) monitors movement of belt 20 from side to side. Belt 20 is supported so as to provide a trio of substantially flat belt runs opposite exposure, developing, and cleaning stations 27, 28, 29 respectfully. To enhance belt flatness at these stations, vacuum platens 30 are provided under belt 20 at each belt run. Conduits 31 communicate vacuum platens 30 with a vacuum pump 32. Photoconductive belt 20 moves in the direction indicated by the solid line arrow, drive thereto being effected through roll 21, which in turn is driven by main drive motor 34, as seen in FIG. 4.

Processor 12 includes a generally rectangular, horizontal transparent platen 35 on which each original 2 to be copied is disposed. A two or four sided illumination assembly, consisting of internal reflectors 36 and flash lamps 37 (shown in FIG. 2) disposed below and along at least two sides of platen 35, is provided for illuminating the original 2 on platen 35. To control temperatures within the illumination space, the assembly is coupled through conduit 33 with a vacuum pump 38 which is adapted to withdraw overly heated air from the space. To retain the original 2 in place on platen 35 and prevent escape of extraneous light from the illumination assembly, a platen cover 35' may be provided.

The light image generated by the illumination system is projected via mirrors 39, 40 and a variable magnification lens assembly 41 onto the photoreceptive belt 20 at the exposure station 27. Reversible motor 43 is provided to move the main lens and add on lens elements that comprise the lens assembly 41 to different predetermined positions and combinations to provide the preselected image sizes corresponding to push button selectors 818, 819, 820 on operator module 800. (See FIG. 32) Sensors 116, 117, 118 signal the present disposition of lens assembly 41. Exposure of the previously charged belt 20 selectively discharges the photoconductive belt to produce on belt 20 an electrostatic latent image of the original 2. To prepare belt 20 for imaging, belt 20 is uniformly charged to a preselected level by charge corotron 42 upstream of the exposure station 27.

To prevent development of charged but unwanted image areas, erase lamps 44, 45 are provided. Lamp 44, which is referred to herein as the pitch fadeout lamp, is supported in transverse relationship to belt 20, lamp 44 extending across substantially the entire width of belt 20 to erase (i.e. discharge) areas of belt 20 before the first image, between successive images, and after the last image. Lamps 45, which are referred to herein as edge fadeout lamps, serve to erase areas bordering each side of the images. Referring particularly to FIG. 5, edge fadeout lamps 45, which extend transversely to belt 20, are disposed within a housing 46 having a pair of transversely extending openings 47, 47' of differing length adjacent each edge of belt 20. By selectively actuating one or the other of the lamps 45, the width of the area bordering the sides of the image that is erased can be controlled.

Referring to FIGS. 1, 6 and 7, magnetic brush rolls 50 are provided in a developer housing 51 at developing station 28. Housing 51 is pivotally supported adjacent the lower end thereof with interlock switch 52 to sense disposition of housing 51 in operative position adjacent belt 20. The bottom of housing 51 forms a sump within which a supply of developing material is contained. A rotatable auger 54 in the sump area serves to mix the developing material and bring the material into operative relationship with the lowermost of the magnetic brush rolls 50.

As will be understood by those skilled in the art, the electrostatically attractable developing material commonly used in magnetic brush developing apparatus of the type shown comprises a pigmented resinous powder, referred to as toner, and larger granular beads referred to as carrier. To provide the necessary magnetic properties, the carrier is comprised of a magnetizable material such as steel. By virtue of the magnetic fields established by developing rolls 50 and the interrelationship therebetween, a blanket of developing material is formed along the surfaces of developing rolls 50 adjacent the belt 20 and extending from one roll to another. Toner is attracted to the electrostatic latent image from the carrier bristles to produce a visible powder image on the surface of belt 20.

Magnetic brush rolls 50 each comprise a rotatable exterior sleeve 55 with relatively stationary magnet 56 inside. Sleeves 55 are rotated in unison and at substantially the same speed as belt 20 by a developer drive motor 57 through a belt and pulley arrangement 58. A second belt and pulley arrangement 59 drives auger 54.

To regulate development of the latent electrostatic images on belt 20, magnetic brush sleeves 55 are electrically biased. A suitable power supply 60 is provided for this purpose with the amount of bias being regulated by controller 18.

Developing material is returned to the upper portion of developer housing 51 for reuse and is accomplished by utilizing a photocell 62 which monitors the level of developing material in housing 51 and a photocell lamp 62' spaced opposite to the photocell 62 in cooperative relationship therewith. The disclosed machine is also provided with automatic developability control which maintains an optimum proportion of toner-to-carrier material by sensing toner concentration and replenishing toner, as needed. As shown in FIG. 8, the automatic developability control comprises a pair of transparent plates 64 mounted in spaced, parallel arrangement in developer housing 51 such that a portion of the returning developing material passes therebetween. A suitable circuit, not shown, alternately places a charge on the plates 64 to attract toner thereto. Photocell 65 on one side of the plate pair senses the developer material as the material passes therebetween. Lamp 65' on the opposite side of plate pair 64 provides reference illumination. In this arrangement, the returning developing material is alternately attracted and repelled to and from plate 64. The accumulation of toner, i.e. density determines the amount of light transmitted from lamp 65' to photocell 65. Photocell 65 monitors the density of the returning developing material with the signal output therefrom being used by controller 18 to control the amount of fresh or make-up toner to be added to developer housing 51 from toner supply container 67.

To discharge toner from container 67, rotatable dispensing roll 68 is provided in the inlet to developer housing 51. Motor 69 drives roll 68. When fresh toner is required, as determined by the signal from photocell 65, controller 18 actuates motor 69 to turn roll 68 for a timed interval. The rotating roll 68, which is comprised of a relatively porous sponge-like material, carries toner particles thereon into developer housing 51 where it is discharged. Pre-transfer corotron 70 and lamp 71 are provided downstream of magnetic brush rolls 50 to regulate developed image charges before transfer.

A magnetic pick-off roll 72 is rotatably supported opposite belt 20 downstream of pre-transfer lamp 71, roll 72 serving to scavenge leftover carrier from belt 20 preparatory to transfer of the developed image to the copy sheet 3. Motor 73 turns roll 72 in the same direction and at substantially the same speed as belt 20 to prevent scoring or scratching of belt 20. One type of magnetic pick-off roll is shown in U.S. Pat. No. 3,834,804, issued Oct. 10, 1974 to Bhagat et al.

Referring to FIGS. 4, 9 and 12, to transfer developed images from belt 20 to the copy sheets 3, a transfer roll 75 is provided. Transfer roll 75, which forms part of the copy sheet feed path, is rotatably supported within a transfer roll housing opposite belt support roll 21. Housing 76 is pivotally mounted at 76' to permit the transfer roll assembly to be moved into and out of operative relationship with belt 20. A transfer roll cleaning brush 77 is rotatably journalled in transfer roll housing 76 with the brush periphery in contact with transfer roll 90. Transfer roll 75 is driven through contact with belt 20 while cleaning brush 77 is coupled to main drive motor 34. To remove toner, housing 76 is connected through conduit 78 with vacuum pump 81. To facilitate and control transfer of the developed images from belt 20 to the copy sheets 3, a suitable electrical bias is applied to transfer roll 75.

To permit transfer roll 75 to be moved into and out of operative relationship with belt 20, cam 79 is provided in driving contact with transfer roll housing 76. Cam 79 is driven from motor 34 through an electromagnetically operated one revolution clutch 80. Spring means (not shown) serves to maintain housing 76 in driving engagement with cam 79.

To facilitate separation of the copy sheets 3 from belt 20 following transfer of developed images, a detack corotron 82 is provided. Corotron 82 generates a charge designed to neutralize or reduce the charges tending to retain the copy sheet on belt 20. Corotron 82 is supported on transfer roll housing 76 opposite belt 20 and downstream of transfer roll 75.

Referring to FIGS. 1, 2 and 10, to prepare belt 20 for cleaning, residual charges on belt 20 are removed by discharge lamp 84 and preclean corotron 94. A cleaning brush 85, rotatably supported within an evacuated semi-circular shaped brush housing 86 at cleaning station 29, serves to remove residual developer from belt 20. Motor 95 drives brush 85, brush 85 turning in a direction opposite that of belt 20.

Vacuum conduit 87 couples brush housing 86 through a centrifugal type separator 88 with the suction side of vacuum pump 93. A final filter 89 on the outlet of motor 93 traps particles that pass through separator 88. The heavier toner particles separated by separator 88 drop into and are collected in one or more collecting bottles 90. Pressure sensor 91 monitors the condition of final filter 89 while a sensor 92 monitors the level of toner particles in collecting bottles 90.

To obviate the danger of copy sheets remaining on belt 20 and becoming entangled with the belt cleaning mechanism, a deflector 96 is provided upstream of cleaning brush 85. Deflector 96, which is pivotally supported on the brush housing 86, is operated by solenoid 97. In the normal or off position, deflector 96 is spaced from belt 20 (the solid line position shown in the drawings). Energization of solenoid 97 pivots deflector 96 downwardly to bring the deflector leading edge into close proximity to belt 20.

Sensors 98, 99 are provided on each side of deflector 96 for sensing the presence of copy material on belt 20. A signal output from upstream sensor 98 triggers solenoid 97 to pivot deflector 96 into position to intercept the copy sheet on belt 20. The signal from sensor 98 also initiates a system shutdown cycle (mis-strip jam) wherein the various operating components are, within a prescribed interval, brought to a stop. The interval permits any copy sheet present in fuser 150 to be removed, sheet trap solenoid 158 (FIG. 12) having been actuated to prevent the next copy sheet from entering fuser 150 and becoming trapped therein. The signal from sensor 99, indicating failure of deflector 96 to intercept or remove the copy sheet from belt 20, triggers an immediate or hard stop (sheet on selenium jam) of the processor. In such instances the power to drive motor 34 is interrupted to bring belt 20 and the other components driven therefrom to an immediate stop.

Referring particularly to FIGS. 1 and 12, copy sheets 3 comprise precut paper sheets supplied from either main or auxiliary paper trays 100, 102. Each paper tray has a platform or base 103 for supporting in stack-like fashion a quantity of sheets. The tray platforms 103 are supported for vertical up and down movement by motors 105, 106. Side guide pairs 107, in each tray 100, 102 delimit the tray side boundaries, the guide pairs being adjustable toward and away from one another in accommodation of different size sheets. Sensors 108, 109 respond to the position of each side guide pair 107, the output of sensors 108, 109 serving to regulate operation of edge fadeout lamps 45 and fuser cooling valve 171 (FIG. 3). Lower limit switches 110 on each tray prevent overtravel of the tray platform in a downward direction.

A heater 112 is provided below the platform 103 of main tray 100 to warm the tray area and enhance feeding of sheets therefrom. Humidstat 113 and thermostat 114 control operation of heater 112 in response to the temperature/humidity conditions of main tray 100. Fan 115 is provided to circulate air within tray 100.

To advance the sheets 3 from either main or auxiliary tray 100, 102, main and auxiliary sheet feeders 120, 121 are provided. Feeders 120, 121 each include a nudger roll 123 to engage and advance the topmost sheet in the paper tray forward into the nip formed by a feed belt 124 and retard roll 125. Retard rolls 125, which are driven at an extremely low speed by motor 126, cooperate with feed belts 124 to restrict feeding of sheets from trays 100, 102 to one sheet at a time.

Feed belts 124 are driven by main and auxiliary sheet feed motors 127, 128 respectively. Nudger rolls 123 are supported for pivotal movement about the axis of feed belt drive shaft 129 with drive to the nudger rolls taken from drive shaft 129. Stack height sensors 133, 134 are provided for the main and auxiliary trays, the pivoting nudger rolls 123 serving to operate sensors 133, 134 in response to the sheet stack height. Main and auxiliary tray misfeed sensors 135, 136 are provided at the tray outlets.

Main transport 140 extends from main paper tray 100 to a point slightly upstream of the nip formed by photoconductive belt 20 and transfer roll 75. Transport 140 is driven from main motor 34. To register sheets 3 with the images developed on belt 20, sheet register fingers 141 are provided, fingers 141 being arranged to move into and out of the path of the sheets on transport 140 once each revolution (see also FIG. 4). Registration fingers 141 are driven from main motor 34 through electromagnetic clutch 145. A timing or reset switch 146 is set once on each revolution of sheet register fingers 141. Sensor 139 monitors transport 140 for jams. Further amplification of sheet register system may be found in U.S. Pat. No. 3,781,004, issued Dec. 25, 1973 to Buddendeck et al.

Pinch roll pair 142 is interspaced between transport belts that comprise main transport 140 on the downstream side of register fingers 141. Pinch roll pair 142 are driven from main motor 34.

Auxiliary transport 147 extends from auxiliary tray 102 to main transport 140 at a point upstream of sheet register fingers 141. Transport 147 is driven from motor 34.

To maintain the sheets in driving contact with the belts of transports 140, 147, suitable guides or retainers (not shown) may be provided along the belt runs.

The image bearing sheets leaving the nip formed by photoconductive belt 20 and transfer roll 75 are picked off by belts 155 of the leading edge of vacuum transport 149. Belts 155, which are perforated for the admission of vacuum therethrough, ride on forward roller pair 148 and rear roll 153. A pair of internal vacuum plenums 151, 154 are provided, the leading plenum 154 cooperating with belts 155 to pick up the sheets leaving the belt/transfer roll nip. Transport 149 conveys the image bearing sheets to fuser 150. Vacuum conduits 147, 156 communicate plenums 151, 154 with vacuum pumps 152, 152'. A pressure sensor 157 monitors operation of vacuum pump 152. Sensor 144 monitors transport 149 for jams.

To prevent the sheet on transport 149 from being carried into fuser 150 in the event of a jam or malfunction, a trap solenoid 158 is provided below transport 149. Energization of solenoid 158 raises the armature thereof into contact with the lower face of plenum 154 to intercept and stop the sheet moving therepast.

Referring particularly to FIGS. 4, 10 and 12, fuser 150 comprises a lower heated fusing roll 160 and upper pressure roll 161. Rolls 160, 161 are supported for rotation in fuser housing 162. The core of fusing roll 160 is hollow for receipt of heating rod 163 therewithin.

Housing 162 includes a sump 164 for holding a quantity of liquid release agent, herein termed oil. Dispensing belt 165, moves through sump 164 to pick up the oil, belt 165 being driven by motor 166. A blanket-like wick 167 carries the oil from belt 165 to the surface of fusing roll 160.

Pressure roll 161 is supported within an upper pivotal section 168 of housing 162. This enables pressure roll 161 to be moved into and out of operative contact fusing roll 160. Cam shaft 169 in the lower portion of fuser housing 162 serves to move housing section 168 and pressure roll 161 into operative relationship with fusing roll 160 against a suitable bias (not shown). Cam shaft 169 is coupled to main motor 34 through an electromagnetically operated one revolution clutch 159.

Fuser section 168 is evacuated, conduit 170 coupling housing section 168 with vacuum pump 152. The ends of housing section 168 are separated into vacuum compartments opposite the ends of pressure roll 161 thereunder to cool the roll ends where smaller size copy sheets 3 are being processed. Vacuum valve 171 (FIG. 3) in conduit 172 regulates communication of the vacuum compartments with vacuum pump 153' in response to the size sheets as sensed by side guide sensors 108, 109 in paper trays 100, 102.

Fuser roll 160 is driven from main motor 34. Pressure roll 161 is drivingly coupled to fuser roll 160 for rotation therewith.

Thermostat 174 (FIG. 12) in fuser housing 162 controls operation of heating rod 163 in response to temperature. Sensor 175 protects against fuser overtemperature. To protect against trapping of a sheet in fuser 150 in the event of a jam, sensor 176 is provided.

Following fuser 150, the sheet is carried by post fuser transport 180 to either discharge transport 181 or, where duplex or two sided copies are desired, to return transport 182. Sheet sensor 183 monitors passage of the sheets from fuser 150. Transports 180, 181 are driven from main motor 34. Sensor 181' monitors transport 181 for jams. Suitable retaining means may be provided to retain the sheets on transports 180, 181.

A deflector 184, when extended, directs sheets on transport 180 onto conveyor roll 185 and into chute 186 leading to return transport 182. Solenoid 179, when energized raises deflector 184 into the sheet path. Return transport 182 carries the sheets back to auxiliary tray 102. Sensor 189 monitors transport 182 for jams. The forward stop 187 of trays 102 is supported for oscillating movement. Motor 188 drives stop 187 back and forth tap sheets returned to auxiliary tray 102 into alignment for refeeding.

To invert duplex copy sheets following fusing of the second or duplex image, a displaceable sheet stop 190 is provided adjacent the discharge end of chute 186. Stop 190 is pivotally supported for swinging movement into and out of chute 186. Solenoid 191 is provided to move stop 190 selectively into or out of chute 186. Pinch roll pairs 192, 193 serve to draw the sheet trapped in chute 186 by stop 190 and carry the sheet forward onto discharge transport 181. Further description of the inverter mechanism may be found in U.S. Pat. No. 3,856,295, issued Dec. 24, 1974, to John H. Looney.

Output tray 195 receives unsorted copies. Transport 196 a portion of which is wrapped around a turn around roll 197, serves to carry the finished copies to tray 195. Sensor 194 monitors transport 196 for jams. To route copies into output tray 195, a deflector 198 is provided. Deflector solenoid 199, when energized, turns deflector 198 to intercept sheets on conveyor 181 and route the sheets onto conveyor 196.

When output tray 195 is not used, the sheets are carried by conveyor 181 to sorter 14.

Referring particularly to FIG. 13, sorter 14 comprises upper and lower bin arrays 210, 211. Each bin array 210, 211 consists of series of spaced downwardly inclined trays 212, forming a series of individual bins 213 for receipt of finished copies 3'. Conveyors 214 along the top of each bin array, cooperate with idler rolls 215 adjacent the inlet to each bin to transport the copies into juxtaposition with the bins. Individual defelctors 216 at each bin cooperate, when depressed, with the adjoining idler roll 215 to turn the copies into the bin associated therewith. An operating solenoid 217 is provided for each deflector.

A driven roll pair 218 is provided at the inlet to sorter 14. A generally vertical conveyor 219 serves to bring copies 3' to the upper bin array 210. Entrance deflector 220 routes the copies selectively to either the upper or lower bin array 210, 211 respectively. Solenoid 221 operates deflector 220.

Motor 222 is provided for each bin array to drive the conveyors 214 and 219 of upper bin array 210 and conveyor 214 of lower bin array 211. Roll pair 218 is drivingly coupled to both motors.

To detect entry of copies 3' in the individual bins 213, a photoelectric type sensor 225, 226 is provided at one end of each bin array 210, 211 respectively. Sensor lamps 225', 226' are disposed adjacent the other end of the bin array. To detect the presence of copies in the bins 213, a second set of photoelectric type sensors 227, 228 is provided for each bin array, on a level with a tray cutout (not shown). Reference lamps 227', 228' are disposed opposite sensors 227, 228.

Referring particularly to FIGS. 14 and 15, document handler 16 includes a tray 233 into which originals or documents 2 to be copied are placed by the operator following which a cover (not shown) is closed. A movable bail bar or separator 235, driven in an oscillatory path from motor 236 through a solenoid operated one revolution clutch 238, is provided to maintain document separation.

A document feed belt 239 is supported on drive and idler rolls 240, 241 and kicker roll 242 under tray 233, tray 233 being suitably apertured to permit the belt surface to project therewithin. Feedbelt 239 is driven by motor 236 through electromagnetic clutch 244. Guide 245, disposed near the discharge end of feed belt 239, cooperates with belt 239 to form a nip between which the documents pass.

A photoelectric type sensor 246 is disposed adjacent the discharge end of belt 239. Sensor 246 responds on failure of a document to feed within a predetermined interval to actuate solenoid operated clutch 248 which raises kicker roll 242 and increases the surface area of feed belt 239 in contact with the documents. Another sensor 259 located underneath tray 233 provides an output signal when the last document 2 of each set has left the tray 233.

Document guides 250 route the document fed from tray 233 via roll pair 251, 252 to platen 35. Roll 251 is drivingly coupled to motor 236 through electromagnetic clutch 244. Contact of roll 251 with roll 252 turns roll 252.

Roll pair 260, 261 at the entrance to platen 35 advance the document onto platen 35, roll 260 being driven through electromagnetic clutch 262 in the forward direction. Contact of roll 260 with roll 261 turns roll 261 in the document feeding direction. Roll 260 is selectively coupled through gearset 268 with motor 236 through electromagnetic clutch 265 so that on engagement of clutch 265 and disengagement of clutch 262, roll 260 and roll 261 therewith turn in the reverse direction to carry the document back to tray 233 via return chute 276. One way clutches 266, 267 permit free wheeling of the roll drive shafts.

The document leaving roll pair 260, 261 is carried by platen feed belt 270 onto platen 35, belt 270 being comprised of a suitable flexible material having an exterior surface of xerographic white. Belt 270 is carried about drive and idler rolls 271, 272. Roll 271 is drivingly coupled to motor 236 for rotation in either a forward or reverse direction through clutches 262, 265. Engagement of clutch 262 operates through belt and pulley drive 279 to drive belt in the forward direction, engagement of clutch 265 operates through drive 279 to drive belt 270 in the reverse direction.

To locate the document in predetermined position on platen 35, a register 273 is provided at the platen inlet for engagement with the document trailing edge. For this purpose, control of platen belt 270 is such that the following transporting of the document onto plate 35 and beyond register 273, belt 270 is reversed to carry the document backwards against register 273.

To remove the document from platen 35 following copying, register 273 is retracted to an inoperative position. Solenoid 274 is provided for moving register 273.

A document deflector 275, is provided to route the document leaving platen 35 into return chute 276. For this purpose, platen belt 270 and pinch roll pair 260, 261 are reversed through engagement of clutch 265. Discharge roll pair 278, driven by motor 236, carry the returning document into tray 233.

To monitor movement of the documents in document handler 16 and detect jams and other malfunctions, photoelectric type sensors 246 and 280, 281 and 282 are disposed along the document routes.

To align documents 2 returned to tray 233, a document patter 284 is provided adjacent one end of tray 233. Patter 284 is oscillated by motor 285.

To provide the requisite operational synchronization between host machine 10 and controller 18 as will appear, processor or machine clock 202 is provided. Referring particularly to FIG. 1, clock 202 comprises a toothed disc 203 drivingly supported on the output shaft of main drive motor 34. A photoelectric type signal generator 204 is disposed astride the path followed by the toothed rim of disc 203, generator 204 producing, whenever drive motor 34 is energized, a pulse like signal output at a frequency correlated with the speed of motor 34, and the machine components driven therefrom.

As described, a second machine driven clock, termed a pitch reset clock 138 herein, and comprising timing switch 146 is provided. Switch 146 cooperates with sheet register fingers 141 to generate an output pulse once each revolution of fingers 141. As will appear, the pulse like output of the pitch reset clock is used to reset or resynchronize controller 18 with host machine 10.

Referring to FIG. 15, a document handler clock 286 consisting of apertured disc 287 on the output shaft of document handler drive motor 236 and cooperating photoelectric type signal generator 288 is provided. As in the case of machine clock 202, document handler clock 286 produces an output pulse train from which components of the document handler may be synchronized. A real time clock derived from clock 552 of FIG. 17, is utilized to control internal operations of the controller 18 as is known in the art, as well as the timing of some of the machine components.

Referring to FIG. 16, controller 18 includes a Central Processor Unit (CPU) Module 500, Input/Output (I/O) Module 502, and Interface 504. Address, Data and Control Buses 507, 508, 509 respectively operatively couple CPU Module 500 and I/O Module 502. CPU Module 500 I/O Module 502 are disposed within a shield 518 to prevent noise interference.

Interface 504 couples I/O Module 502 with special circuits module 522, input matrix module 524, and main panel interface module 526. Module 504 also couples I/O Module 502 to operating sections of the machine, namely, document handler section 530, input section 532, sorter section 534 and processor sections 536, 538. A spare section 540, which may be used for monitoring operation of the host machine, or which may be later utilized to control other devices, is provided.

Referring to FIGS. 17, 18, CPU module 500 comprises a processor 542 such as an Intel 8080 microprocessor manufactured by Intel Corporation, Santa Clara, Calif. 16K Read Only Memory (herein ROM) and 2K Random Access Memory (herein RAM) sections 545, 546, Memory Ready section 548, power regulator section 550, and onboard clock 552. Bipolar tri-state buffers 510, 511 in Address and Data buses 507, 508 disable the bus on a Direct Memory access (DMA) signal (HOLDA) as will appear. While the capacity of memory sections 545, 546 are indicated throughout as being 16K and 2K respectively, other memory sizes may be readily contemplated.

Referring particularly to FIG. 19, clock 552 comprises a suitable clock oscillator 553 feeding a multi-bite (Qa-Qn) shift register 554. Register 554 includes an internal feedback path from one bit to the serial input of register 554. Output signal waveforms .phi..sub.1, .phi..sub.2, .phi..sub.1-1 and .phi..sub.2-1 are produced for use by the system.

Referring to FIG. 20, the memory bytes in ROM section 545 are implemented by address signals (A0-A15) from processor 542, selection being effected by 3 to 8 decode chip 560 controlling chip select 1 (CS-1) and a 1 bit selection (A13) controlling chip select 2 (CS-2). The most significant address bits (A14, A15) select the first 16K of the total 64 bytes of the addressing space. The memory bytes in RAM section 546 are implemented by Address signals (A0-A15) through selector circuit 561. Address bit A10 serves to select the memory bank while the remaining five most significant bits (A11-A15) select the last 2K bytes out of the 64K bytes of addressing space. RAM memory section 546 includes a 40 bit output buffer the output of which is tied together with the output from ROM memory section 545 and goes to tri-state buffer 562 to drive Data bus 508. Buffer 562 is enabled when either memory section 545 or 546 is being addressed and either a (MEM READ) or DMA (HOLD A) memory request exists. An enabling signal (MEMEN) is provided from the machine control or service panel (not shown) which is used to permit disabling of buffer 562 during servicing of CPU Module 500. Write control comes from either processor 542 (MEM WRITE) or from DMA (HOLD A) control. Tri-state buffers 563 permit Refresh Control 605 of I/O Module 502 to access MEM READ and MEM WRITE control channels directly on a DMA signal (HOLD A) from processor 542 as will appear.

Referring to FIG. 21, memory ready section 548 provides a READY signal to processor 542. A binary counter 566, which is initialized by a SYNC signal (.phi.,) to a prewired count as determined by input circuitry 567, counts up at a predetermined rate. At the maximum count, the output at gate 568 comes true stopping the counter 566. If the cycle is a memory request (MEM REQ) and the memory location is on board as determined by the signal (MEM HERE) to tri-state buffer 569, a READY signal is sent to processor 542. Tri-state buffer 570 in MEM REQ line permits Refresh Control 605 of I/O Module 502 to access the MEM REQ channel directly on a DMA signal (HOLD A) from processor 542 as will appear.

Referring to FIG. 22, power regulators 550, 551, 552 provide the various voltage levels, i.e. +5v, +12v, and -5v D.C. required by the module 500. Each of the three on board regulators 550, 551, 552 employ filtered D.C. inputs. Power Not Normal (PNN) detection circuitry 571 is provided to reset processor 542 during the power up time. Panel reset is also provided via PNN. An enabling signal (INHIBIT RESET) allows completion of a write cycle in Non Volatile (N.V.) Memory 610 of I/O Module 502.

Referring to FIGS. 18, 20, 21 and the DMA timing chart (FIG. 18a) data transfer from RAM section 546 to host machine 10 is effected through Direct Memory Access (DMA), as will appear. To initiate DMA, a signal (HOLD) is generated by Refresh Control 605 (FIG. 23a). On acceptance, processor 542 generates a signal HOLD ACKNOWLEDGE (HOLD A) which works through tri-state buffers 510, 511 and through buffers 563 and 570 to release Address bus 507, Data bus 508 and MEM READ, MEM WRITE, and MEM REQ channels (FIGS. 20, 21) to Refresh Control 605 of I/O Module 502.

Referring to FIG. 23, I/O Module 502 interfaces with CPU module 500 through bi-directional Address, Data and Control buses 507, 508, 509. I/O Module 502 appears to CPU module 500 as a memory portion. Data transfers between CPU and I/O modules 500, 502, and commands to I/O module 502 except for output refresh are controlled by memory reference instructions executed by CPU module 500. Output refresh which is initiated by one of several uniquely decoded memory reference commands, enables Direct Memory access (DMA) by I/O module 502 to RAM section 546.

I/O module 502 includes Matrix Input select 604 (through which inputs from the host machine 10, are received), Refresh Control 605, Nonvolatile (NV) memory 610, Interrupt Control 612, Watch dog Timer and failure Flag 614 and clock 570.

A Function Decode Section 601 receives and interprets commands from CPU section 500 by decoding information on address bus 507 along with control signals from processor 542 on control bus 509. On command, decode section 601 generates control signals to perform the function indicated. These functions include (a) controlling tri-state buffers 620 to establish the direction of data flow in Data bus 508; (b) strobing data from Data bus 508 into buffer latches 622; (c) controlling multiplexer 624 to put data from Interrupt Control 612, Real Time clock register 621, Matrix Input Select 604 or N.V. memory 610 onto data bus 508; (d) actuating refresh control 605 to initiate a DMA operation; (e) actuating buffers 634 to enable address bits A0-A7 to be sent to the host machine 10 for input matrix read operations; (f) commanding operation of Matrix Input Select 604; (g) initiating read or write operation of N.V. memory 610 through Memory Control 638; (h) loading Real Time clock register 621 from data bus 508; and (i) resetting the Watch Dog timer or setting the Fault Failure flag 614. In addition, section 601 includes logic to control and synchronize the READY control line to CPU module 500, the READY line being used to advise module 500 when data placed on the Data bus by I/O module 502 is valid.

Watch dog timer and failure flag 614, which serves to detect certain hardwired and software malfunctions, comprises a free running counter which under normal circumstances is periodically reset by an output refresh command (REFRESH) from Function Decode Section 601. If an output refresh command is not received within a preset time interval, (i.e. 25m sec) a fault flip flop is set and a signal (FAULT) sent to the host machine 10. The signal (FAULT) also raises the HOLD line to disable CPU Module 500. Clearing of the fault flip flop may be by cycling power or generating a signal (RESET). A selector (not shown) may be provided to disable (DISABLE) the watch dog timer when desired. The fault flip flop may also be set by a command from the CPU Module to indicate that the operating program detected a fault.

Matrix Input select 604 has capacity to read up to 32 groups of 8 discrete inputs from host machine 10. Lines A.sub.3 through A.sub.7 of Address bus 507 are routed to host machine 10 via CPU Interface Module 504 to select the desired group of 8 inputs. The selected inputs from machine 10 are received via Input Matrix Module 524 (FIG. 28) and are placed by matrix 604 onto data bus 508 and sent to CPU Module 500 via multiplexer 624. Bit selection is effected by lines A.sub.0 through A.sub.2 of Address bus 507.

Output refresh control 605, when initiated, transfers either 16 or 32 sequential words from RAM memory output buffer 546' to host machine 10 at the predetermined clock rate in line 574. Direct Memory access (DMA) is used to facilitate transfer of the data at a relatively high rate. On a Refresh signal from Function Decode Section 601, Refresh Control 605 generates a HOLD signal to processor 542. On acknowledgement (HOLD A) processor 542 enters a hold condition. In this mode, CPU Module 500 releases address and data buses 507, 508 to the high impedance state giving I/O module 502 control thereover. I/O module 502 then sequentially accesses the 32 memory words from output buffer 546' (REFRESH ADDRESS) and transfers the contents to the host machine 10. CPU Module 500 is dormant during this period.

A control signal (LOAD) in line 607 along with the predetermined clock rate determined by the clock signal (CLOCK) in line 574 is utilized to generate eight 32 bit serial words which are transmitted serially via CPU Interface Module 504 to the host machine remote locations where serial to parallel transformation is performed. Alternatively, the data may be stored in addessable latches and distributed in parallel directly to the required destinations.

N.V. memory 610 comprises a predetermined number of bits of nonvolatile memory stored in I/O module 502 under Memory Control 638. N.V. memory 610 appears to CPU module 500 as part of the CPU module memory complement and therefore may be accessed by the standard CPU memory reference instruction set. Referring particularly to FIG. 24, to sustain the contents of N.V. memory 610 should system power be interrupted, one or more rechargeable batteries 635 are provided exterior to I/O module 502. CMOS protective circuitry 636 couples batteries 635 to memory 610 to preserve memory 610 on a failure of the system power. A logic signal (INHIBIT RESET) prevents the CPU Module 500 from being reset during the N.V. memory write cycle interval so that any write operation in progress will be completed before the system is shut down.

For tasks that require frequent servicing, high speed response to external events, or synchronization with the operation of host machine 10, a multiple interrupt system is provided. These comprise machine based interrupts, herein referred to as Pitch Reset interrupt and the Machine interrupt, as well as a third clock driven interrupt, the Real Time interrupt.

Referring particularly to FIGS. 23(a) and 34, the highest priority interrupt signal, Pitch reset signal 640, is generated by the signal output of pitch reset clock 138. The clock signal is fed via optical isolator 645 and digital filter 646 to edge trigger flip flop 647.

The second highest priority interrupt signal, machine clock signal 641, is sent directly from machine clock 202 through isolation transformer 648 to a phase locked loop 649. Loop 649, which serves as bandpath filter and signal conditioner, sends a square wave signal to edge trigger flip flop 651. The second signal output (LOCK) serves to indicate whether loop 649 is locked onto a valid signal input or not.

The lowest priority interrupt signal, Real Time Clock signal 643, is generated by register 261. Register 621 which is loaded and stored by memory reference instructions from CPU module 500 is decremented by a clock signal in line 543 which may be derived from I/O Module clock 570. On the register count reaching zero, register 621 sends an interrupt signal to edge trigger flip flop 656. A spare interrupt 642 is also provided.

Setting of one or more of the edge trigger flip flops 647, 651, 654, 656 by the interrupt signals 640, 641, 642, 643 generates a signal (INT) via priority chip 659 to processor 542 of CPU Module 500. On acknowledgement, processor 542, issues a signal (INTA) transferring the status of the edge trigger flip flops 647, 651, 654, 656 to a four bit latch 660 to generate an interrupt instruction code (RESTART) onto the data bus 508.

Each interrupt is assigned a unique RESTART instruction code. Should an interrupt of higher priority be triggered, a new interrupt signal (INT) and RESTART instruction code are generated resulting in a nesting of interrupt software routines whenever the interrupt recognition circuitry is enabled within the CPU 500.

Priority chip 659 serves to establish a handling priority in the event of simultaneous interrupt signals in accordance with the priority schedule described.

Once triggered, the edge trigger flip flop 647, 651, 654 or 656 must be reset in order to capture the next occurrence of the interrupt associated therewith. Each interrupt subroutine serves, in addition to performing the functions programmed, to reset the flip flops (through the writing of a coded byte in a uniquely selected address) and to re-enable the interrupt (through execution of a re-enabling instruction). Until re-enabled, initiation of a second interrupt is precluded while the first interrupt is in progress.

Lines 658 permit interrupt status to be interrogated by CPU module 500 on a memory reference instruction.

I/O Module 502 includes a suitable pulse generator or clock 570 for generating the various timing signals required by module 502. Clock 570 is driven by the pulse-like output .phi..sub.1-1, .phi..sub.2-1 of processor clock 552 (FIG. 19a). As described, clock 570 provides a reference clock pulse (in line 574) for synchronizing the output refresh data and is the source of clock pulses (in line 643) for driving Real Time register 621.

CPU interface module 504 interfaces I/O module 502 with the host machine 10 and transmits operating data stored in RAM section 546 to the machine. Referring particularly to FIGS. 25 and 26, data and address information are inputted to module 504 through suitable means such as optical type couplers 700 which convert the information to single ended logic levels. Data in bus 508 on a signal from Refresh Control 605 in line 607 (LOAD), is clocked into module 546 at the reference clock rate in line 574 parallel by bit, serial by byte for a preset byte length, with each data bit of each successive byte being clocked into a separate data channel D0-D7. As best seen in FIG. 25, each data channel D0-D7 has an assigned output function with data channel D0 being used for operating the front panel lamps 830 in the digital display, (see FIG. 32), data channel D1 for special circuits module 522, and remaining data channels D2-D7 allocated to the host machine operating sections 530, 532, 534, 536, 538 and 540. Portions of data channels D1-D7 have bits reserved for front panel lamps and digital display.

Since the bit capacity of the data channels D2-D7 is limited, a bit buffer 703 is preferably provided to catch any bit overflow in data channels D2-D7.

Inasmuch as the machine output sections 530, 532, 534, 536, 538 and 540 are electrically a long distance away, i.e. remote, from CPU interface module 504, and the environment is electrically "noisy", the data stream in channels D2-D7 is transmitted to remote sections 530, 532, 534, 536, 538 and 540 via a shielded twisted pair 704. By this arrangement, induced noise appears as a differential input to both lines and is rejected. The associated clock signal for the data is also transmitted over line 704 with the line shielded carrying the return signal currents for both data and clock signals.

Data in channel D1 destined for special circuits module 522 is inputted to shift register type storage circuitry 705 for transmittal to module 522. Data is also inputted to main panel interface module 526. Address information in bus 507 is converted to single ended output by couplers 700 and transmitted to Input Matrix Module 524 to address host machine inputs.

CPU interface module 504 includes fault detector circuitry 706 for monitoring both faults occurring in host machine 10 and faults or failures along the buses, the latter normally comprising a low voltage level or failure in one of the system power lines. Machine faults may comprise a fault in CPU module 500, a belt mistrack signal from sensor 27 (see FIG. 2), opening one of the machine doors or covers as responded to by conventional cover interlock sensors (not shown), a fuser over temperature as detected by sensor 175, etc. In the event of a bus fault, a reset signal (RESET) is generated automatically in line 709 to CPU module 500 (see FIGS. 17 and 18) until the fault is removed. In the event of a machine fault, a signal is generated by the CPU in line 710 to actuate a suitably relay (not shown) controlling power to all or a portion of host machine 10. A load disabling signal (LOAD DISBL) is inputted to optical couplers 700 via line 708 in the event of a fault in CPU module 500 to terminate input of data to host machine 10. Other fault conditions are monitored by the software background program. In the event of a fault, a signal is generated in line 711 to the digital display on control console 800 (via main panel interface module 526) signifying a fault.

Referring particularly to FIGS. 25 and 27, special circuits module 522 comprises a collection of relatively independent circuits for either monitoring operation of and/or driving various elements of host machine 10. Module 522 incorporates suitable circuitry 712 for amplifying the output of sensors 225, 226, 227, 228 and 280, 281, 282 of sorter 14 and document handler 16 respectively; circuitry 713 for operating fuser release clutch 159; and circuitry 714 for operating main and auxiliary paper tray feed roll clutches 130, 131 and document handler feed clutch 244.

Additionally, fuser detection circuitry 715 monitors temperature conditions of fuser 150 as responded to by sensor 174. On overheating of fuser 150, a signal (FUS-OT) is generated to turn heater 163 off, actuate clutch 159 to separate fusing and pressure rolls 160, 161; trigger trap solenoid 158 to prevent entrance of the next copy sheet into fuser 150, and initiate a shutdown of host machine 10. Circuitry 715 also cycles fuser heater 163 to maintain fuser 150 at proper operating temperatures and signals (FUS-RDUT) host machine 10 when fuser 150 is ready for operation.

Circuitry 716 provides closed loop control over sensor 98 which responds to the presence of a copy sheet 3 on belt 20. On a signal from sensor 98, solenoid 97 is triggered to bring deflector 96 into intercepting position adjacent belt 20. At the same time, a backup timer (not shown) is actuated. If the sheet is lifted from the belt 20 by deflector 96 within the time alloted, a signal from sensor 99 disables the timer and a misstrip type jam condition of host machine 10 is declared and the machine is stopped. If the signal from sensor 99 is not received within the allotted time, a sheet on selenium (SOS) type jam is declared and an immediate machine stop is effected.

Circuitry 718 controls the position (and hence the image reduction effected) by the various optical elements that comprise main lens 41 in response to the reduction mode selected by the operator and the signal inputs from lens position responsive sensors 116, 117, 118. The signal output of circuitry 718 serves to operate lens drive motor 43 as required to place the optical elements of lens 41 in proper position to effect the image reduction programmed by the operator.

Referring to FIG. 28, input matrix module 524 provides analog gates 719 for receiving data from the various host machine sensors and inputs (i.e. sheet sensors 135, 136; pressure sensor 157; etc), module 524 serving to convert the signal input to a byte oriented output for transmittal to I/O module 502 under control of Input Matrix Select 604. The byte output to module 524 is selected by address information inputted on bus 507 and decoded on module 524. Conversion matrix 720, which may comprise a diode array, converts the input logic signals of "0" to logic "1" true. Data from input matrix module 524 is transmitted-via optical isolators 721 and Input Matrix Select 604 of I/O module 502 to CPU Module 500.

Referring particularly to FIG. 29, main panel interface module 526 serves as interface between CPU interface module 504 and operator control console 800 for display purposes and as interface between input matrix module 524 and the console switches. As described, data channels D0-D7 have data bits in each channel associated with the control console digital display or lamps. This data is clocked into buffer circuitry 723 and from there, for digital display, data in channels D1-D7 is inputted to multiplexer 724. Multiplexer 724 selectively multiplexes the data to HEX to 7 segment converter 725. Software controlled output drivers 726 are provided for each digit which enable the proper display digit in response to the data output of converter 725. This also provides blanking control for leading zero suppression or inter digit suppression.

Buffer circuitry 723 also enables through anode logic 728 the common digit anode drive. The signal (LOAD) to latch and lamp driver control circuit 729 regulates the length of the display cycle.

For console lamps 830, data in channel D0 is clocked to shift register 727 whose output is connected by drivers to the console lamps. Access by input matrix module 524 to the console switches and keyboard is through main panel interface module 526.

The machine output sections 530, 532, 534, 536, 538, 540 are interfaced with I/O module 502 by CPU interface module 504. At each interrupt/refresh cycle, data is outputted to sections 530, 532, 534, 536, 538, 540 at the clock signal rate in line 574 over data channels D2, D3, D4, D5, D6, D7 respectively.

Referring to FIG. 30, wherein a typical output section i.e. document handler section 530 is shown, data inputted to section 530 is stored in shift register/latch circuit combination 740, 741 pending output to the individual drivers 742 associated with each machine component. Preferably d.c. isolation between the output sections is maintained by the use of transformer coupled differential outputs and inputs for both data and clock signals and a shielded twisted conductor pair. Due to transformer coupling, the data must be restored to a d.c. waveform. For this purpose, control recovery circuitry 744, which may comprise an inverting/non-inverting digital comparator pair and output latch is provided.

The LOAD signal serves to lockout input of data to latches 741 while new data is being clocked into shift register 740. Removal of the LOAD signal enables commutation of the fresh data to latches 741. The LOAD signal also serves to start timer 745 which imposes a maximum time limit within which a refresh period (initiated by Refresh Control 605) must occur. If refresh does not occur within the prescribed time limit, timer 745 generates a signal (RESET) which sets shift register 740 to zero.

With the exception of sorter section 534 discussed below, output sections 532, 536, 538 and 540 are substantially identical to document handler section 530.

Referring to FIG. 31 wherein like numbers refer to like parts, to provide capacity for driving the sorter deflector solenoids 217, a decode matrix arrangement consisting of a Prom encoder 750 controlling a pair of decoders 751, 752 is provided. The output of decoders 751, 752 drive the sorter solenoids 217 of upper and lower bin arrays 210, 211 respectively. Data is inputted to encoder 750 by means of shift register 754.

Referring now to FIG. 32, control console 800 serves to enable the operator to program host machine 10 to perform the copy run or runs desired. At the same time, various indicators on console 800 reflect the operational condition of machine 10. Console 800 includes a bezel housing 802 suitably supported on host machine 10 at a convenient point with decorative front or face panel 803 on which the various machine programming buttons and indicators appear. Programming buttons include power on/off buttons 804, start print (PRINT) buttons 805, stop print (STOP) button 806 and keyboard copy quantity selector 808. A series of feature select buttons consisting of auxiliary paper tray button 810, two sided copy button 811, copy lighter button 814, and copy darker button 815, are provided.

Additionally, image size selector buttons 818, 819, 820; multiple or single document select buttons 822, 823 for operation of document handler 16; and sorter sets or stacks buttons 825, 826 are provided. An on/off service selector 828 is also provided for activation during machine servicing.

Indicators comprise program display lamps 830 and displays such as READY, WAIT, SIDE 1, SIDE 2, ADD PAPER, CHECK STATUS PANEL, PRESS FAULT CODE, QUANTITY COMPLETED, CHECK DOORS, UNLOAD AUX TRAY, CHECK DOCUMENT PATH, CHECK PAPER PATH, JOB INCOMPLETE AND UNLOAD SORTER. Other display information may be envisioned.

As will appear, host machine 10 is conveniently divided into a number of operational states. The machine control program is divided into background routines and Foreground routines with operational control normally residing in the Background routine or routines appropriate to the particular machine state then in effect. The output buffer 546' of RAM memory section 546 is used to transfer/refresh control data to the various remote locations in host machine 10, control data from both Background and Foreground routines being inputted to buffer 546' for subsequent transmittal to host machine 10. Transmittal/refresh of control data presently in output buffer 546' is effected through Direct Memory access (DMA) under the aegis of a Machine Clock interrupt routine.

Foreground routine control data which includes a Run Event Table built in response to the particular copy run or runs programmed, is transferred to output buffer 546' by means of a multiple prioritized interrupt system wherein the Background routine in process is temporarily interrupted while fresh Foreground routine control data is inputted to buffer 546' following which the interrupted Background routine is resumed.

The operating program for host machine 10 is divided into a collection of foreground tasks, some of which are driven by the several interrupt routines and background or non-interrupt routines. Foreground tasks are tasks that generally require frequent servicing, high speed response, or synchronization with the host machine 10. Background routines are related to the state of host machine 10, different background routines being performed with different machine states. A single background software control program (STCK) composed of specific sub-programs associated with the principal operating states of host machine 10 is provided. A byte called STATE contains a number indicative of the current operating state of host machine 10. The machine STATES are as follows:

Referring to FIG. 33, each STATE is normally divided into PROLOGUE, LOOP and EPILOGUE sections. As will be evident from the exemplary program STCK reproduced in TABLE I, entry into a given STATE (PROLOGUE) normally causes a group of operations to be performed, these consisting of operations that are performed once only at the entry into the STATE. For complex operations, a CALL is made to an applications subroutine therefor. Relatively simpler operations (i.e. turning devices on or off, clearing memory, presetting memory, etc.) are done directly.

Once the STATE PROLOGUE is completed, the main body (LOOP) is entered. The program (STCK) remains in this LOOP until a change of STATE request is received and honored. On a change of STATE request, the STATE EPILOGUE is entered wherein a group of operations are performed, following which the STATE moves into the PROLOGUE of the next STATE to be entered.

Referring to FIG. 34 and the exemplary program (STCK) in TABLE I. On actuation of the machine POWER-ON botton 804, the software Initialize STATE (INIT) is entered. In this STATE, the controller is initialized and a software controlled self test subroutine is entered. If the self test of the controller is successfully passed, the System Not Ready STATE (NRDY) is entered. If not, a fault condition is signaled.

In the System Not Ready STATE (NRDY), background subroutines are entered. These include setting of Ready flags, control registers, timers, and the like; turning on power supplies, the fuser, etc., initializing the Fault Handler, checking for paper jams (left over from a previous run), door and cover interlocks, fuser temperatures, etc. During this period, the WAIT lamp on console 800 is lit and operation of host machine 10 precluded.

When all ready conditions have been checked and found acceptable, the controller moves to the system ready state (RDY). The READY lamp on console 800 is lit and final ready checks made. Host Machine 10 is now ready for operation upon completion of input of a copy run program, loading of one or more originals 2 into document handler 16 (if selected by the operator), and actuation of START PRINT button 805. As will appear hereinafter, the next state is PRINT wherein the particular copy run programmed is carried out.

While the machine is completing a copy run, the controller normally enters the Run Not Print state (RUNNPRT) where the controller calculates the number of copies delivered, resets various flags, stores certain machine event information in the memory, as well as generally conditioning the machine for another copy run, if desired. The controller then returns to the System Not Ready state (NRDY) to recheck for ready conditions prepatory for another copy run, with the same state sequence being repeated until the machine is turned off by actuation of POWER OFF button 804 or a malfunction inspired shutdown is triggered. The last state (TECHREP) is a machine servicing state wherein certain service routines are made available to the machine/repair personnel, i.e. Tech Reps.

Referring particularly to FIG. 32 and Tables II, III, IV, V, VI and VII, the machine operator uses control console 800 to program the machine for the copy run desired. Programming may be done during either the System Not Ready (NRDY) or System Ready (RDY) states, although the machine will not operate during the System Not ready state should START PRINT button 805 be pushed. The copy run includes selecting (using keyboard 808) the number of copies to be made, and such other ancillary program features as may be desired, i.e. use of auxiliary paper tray 102, (push button 810), image size selection (push buttons 818, 819, 820), document handler/sorter selection (push buttons 822, 823, 825, 826), copy density (push buttons 814, 815), duplex or two sided copy button 811, etc. On completion of the copy run program, START PRINT button 805 is actuated to start the copy run programmed (presuming the READY lamp is on and an original or originals 2 have been placed in tray 233 of document handler 16 if the document handler has been selected).

With programming of the copy run instructions, controller 18 enters a Digit Input routine in which the program information is transferred to RAM section 546. The copy run program data passes via Main Panel Interface Module 526 to Input Matrix Module 524 and from there is addressed through Matrix Input Select 604, Multiplexer 624, and Buffers 620 of I/O Module 502 to RAM section 546 of CPU Module 500.

On entering PRINT STATE, a Run Event Table (FIG. 35) comprised of Foreground tasks is built for operating in cooperation with the background tasks the various components of host machine 10 in an integrated manner to produce the copies programmed. The run Event Table is formed by controller 18 through merger of a Fixed Pitch Event Table (TABLE II) (stored in ROM 545 and Non Volatile Memory 610) and a Variable Pitch Event Table (TABLE III) in a fashion appropriate to the parameters of the job selected.

The Fixed Pitch Event Table (TABLE II) is comprised of machine events whose operational timing is fixed during each pitch cycle such as the timing of bias to transfer roll 75, (TRN 2 CURR), actuating toner concentration sensor 65 (ADC ACT), loading roll 161 of fuser 150 (FUS*LOAD), and so forth, irrespective of the particular copy run programmed. The Variable Pitch Table (TABLE III) is comprised of machine events whose operational timing varies with the individual copy run programmed, i.e. timing of pitch fadeout lamp 44 (FO*ONBSE) and timing of flash illumination lamps 37 (FLSH BSE). The variable Pitch Table is built by the Pitch Table Builder (TABLE IV) from the copy run information programmed in by controller 18 (using the machine control program stored in ROM section 545 and Non-Volatile Memory 610), coupled with event address information from ROM section 545, sorted by absolute clock count (via the routine shown in TABLE V), and stored in RAM section 546 (via the routine shown in TABLE VI). The Fixed Pitch Event Table and Variable Pitch Table are merged with the relative clock count differences between Pitch events calculated to form a Run Event Table (TABLE VII).

Referring particularly to FIG. 35, the Run Event Table consists of successive groups of individual events 851. Each event 851 is comprised of four data blocks, data block 852 containing the number of clock pulses (from machine clock 202) to the next scheduled pitch event (REL DIFF), data block 853 containing the shift register position associated with the event (REL SR), and data blocks 854, 855 (EVENT LO) (EVENT HI) containing the address of the event subroutine.

In machine states other than PRINT, data blocks 852, 853 (REL DIFF) (REL SR) are set to zero. Data blocks 854, 855 hold the address information for the Non-Print state event.

Control Data in the Run Event Table represents a portion of the foregound tasks and is transferred to the output buffer 546' of RAM memory section 546 by the Pitch Reset and Machine Clock interrupt routines. Other control data, representing foreground tasks not in the Run Event Table is transferred to RAM output buffer 546' by the Real Time Clock interrupt routine. Transfer of the remainder of the control data to output buffer 546' is by means of background (non-interrupt) routines.

Transfer of control data from ouput buffer 546' of RAM memory section 546 to the various locations in host machine 10 is through output Refresh via Direct Memory access (DMA) In response to machine clock interrupt signals as will appear. The interrupt routines are initiated by the respective interrupt signals.

Referring particularly to FIG. 23 and 35-37 and TABLES VII, VIII the interrupt having the highest priority, the Pitch Reset interrupt (signal 640), is operable only during the PRINT state, and occurs once each revolution of sheet register fingers 141 as responded to by sensor 146 of pitch reset clock 138. At each pitch reset interrupt signal, after a determination of priority by Priority Chip 659 in the event of multiple interrupt signals, an interrupt signal (INT) is generated. The acknowledgement signal (INTA) from processor 542 initiates the pitch reset interrupt routine.

On entering the pitch reset routine, the interrupt is re-enabled and the contents of the program working registers stored. A check is made to determine if building of the Run Event Table is finished. Also checks are made to insure that a new shift register schedules have been built and at least 910 clock counts since the last pitch reset have elapsed. If not, an immediate machine shutdown is initiated.

Presuming that the above checks are satisfactory, the shift register pointer (SR PTR), which is the byte variable containing the address of a pre-selected shift register position (SR O), is decremented by one and adjusted for overflow and the shift register contents are updated with a byte variable (SR+VALUV) containing the new shift register value to be shifted in following the pitch reset interrupt. The event pointer (EV*PTR), a two byte variable containing the full address of the next scheduled event, is reset to Event #1. The count in the C register equals the time to the first event.

Machine Cycle Down, Normal Down, and Side One Delay checks are made, and if negative, the count on a cycle up counter (CYC UP CT) is checked. If the count is less than a predetermined control count (i.e. 5), the counter (CYC UP CT) is incremented by one. When the count on the cycle up counter equals the control count, an Image Made Flag is set.

If a Normal Down, Cycle Down, or Side One Delay has been initiated, the cycle up counter (CYC UP CT) is reset to a preset starting count (i.e. 2). The pitch reset interrupt routine is exited with restoration of the working registers and resetting of pitch reset flip flop 647.

The Machine Clock Interrupt routine, which is second in priority, is operative in all operational states of host machine 10. Although nominally driven by machine clock 202, which is operative only during Print state when processor main drive motor 34 is energized, machine clock pulses are also provided by phase locked loop 649 when motor 34 is stopped.

Referring particularly to FIG. 38 and TABLE IX, entry to the Machine Clock interrupt routine there shown is by a signal (INTA) from processor 542 following a machine clock interrupt signal 642 as described earlier. On entry, the event control register (C REG) is obtained and the working register contents stored. The C REG is decremented by one, the register having been previously set to a count corresponding to the next event in the Event Run Table.

The control register (C REG) is checked for zero. If the count is not zero and is an odd number, an output refresh cycle is initiated to effect transfer/refresh of data in RAM output buffer 546' to host machine 10. If the number is even, or following an output refresh, the interrupt system is re-enabled, the machine clock interrupt flip flop 651 is reset and the working registers are restored. Return is then made to the interrupted routine.

If the control register (C REG) count is zero, the Event Pointer (EV*PTR), which identifies the clock count (in data block 852) for the next scheduled event (REL DIFF), is loaded and the control register (C REG) reset to a new count equal to the time to the next event. The Event Pointer (EV*PTR) is incremented to the relative shift register address for the event (REL SR, data block 853), and the shift register address information is set in appropriate shift registers (B, D, E, A registers).

The event Pointer (EV*PTR) is incremented successively to the event subroutine address information (EVENT LO) (EVENT HI) in the Event Run Table, and the address information therefrom loaded into a register pair (D and E registers). The Event Pointer (EV PTR) is incremented to the first data block (REL DIFF) of the next succeeding event in the Run Event Table, saved, and the register pair (H and L registers) that comprise the Event Pointer are loaded with the event subroutine address from the register pair (D and E registers) holding the information. The register pair (D and E registers) are set to the return address for the Event Subroutine. Using the address information, the Event Subroutine is called and the subroutine data transferred to RAM output buffer 546' for transfer to the host machine on the next Output Refresh.

Following this, the Machine Clock interrupt routine is exited as described earlier.

The Output Refresh cycle alluded to earlier functions, when entered, to transfer/refresh data from the output buffer of 546' RAM section 546 to host machine 10. Direct Memory Access (DMA) is used to insure a high data transfer rate.

On a refresh, Refresh Control 605 (see FIG. 23) raises the HOLD line to processor 542, which on completion of the operation then in progress, acknowledges by a HOLD A signal. With processor 542 in a hold mode and Address and Data buses 507, 508 released to I/O Module 502 (through operation of tri-state buffers 510, 511, 563, 570), the I/O module then sequentially accesses the output buffer 546' of RAM section 546 and transfers the contents thereof to host machine 10. Data previously transferred is refreshed.

The Real Time Interrupt, which carries the lowest priority, is active in all machine states. Primarily, the interrupt acts as an interval timer by decrementing a series of timers which in turn serve to control initiation of specialized subroutines used for control and error checking purposes.

Referring particularly to FIG. 39 and TABLE X, the Real Time interrupt routine is entered in the same manner as the interrupt routines previously described, entry being in response to a specific RESTART instruction code assigned to the Real Time interrupt. On entry, the interrupt is re-enabled and the register contents stored. The timer pointer (PNTR) for the first class of timers (i.e. 10 msec TIMERS) is loaded, and a loop counter identifying the number of timers of this class (i.e. 10 msec TIMERS) preset. A control register (E REG) is loaded and a timer decrementing loop is entered for the first timer. The loop decrements the particular timer, increments the timer pointer (PNTR) to the location of the next timer in this class, checks the timer count, and decrements the loop counter. The decrementing loop routine is repeated for each timer in the class (i.e. 10 msec TIMERS) following which a control counter (CNTR) for the second group of timers (i.e. 100 msec TIMERS) is decremented by one and the count checked.

The control counter (CNTR) is initially set to a count equal to the number of times the first timer interval is divisible into the second timer interval. For example, if the first class of timers are 10 msec timers and the second timer class are 100 msec timers, the control counter (CNTR) is set at 10 initially and decremented on each Real Time interrupt by one down to zero.

If the count on the control counter (CNTR) is not zero, the registers are restored, Real Time interrupt flip flop 856 reset, and the routine exited. If the count on the control counter is zero, the counter is reloaded to the original maximum count (i.e. 10) and a loop is entered decrementing individually the second group of timers (i.e. 100 msec TIMERS). On completion, the routine is exited as described previously.

In the following TABLES:

" " -- is used to indicate flags, counters and subroutine names.

"#" -- is used to indicate input signals.

"$" -- is used to indicate output signals.

":" -- is used to indicate macro instructions, system subroutines, system flags, and data, etc.

For further explanation of the mnemonics and particular instructions utilized by the following routines, the reader is directed to Intel Corporation's Programming Manual for the 8080 Microcomputer System. ##SPC1## ##SPC2## ##SPC3## ##SPC4## ##SPC5##

Referring particularly to the timing chart shown in FIG. 40, an exemplary copy run wherein three copies of each of two simplex or one-sided originals in duplex mode is made. Referring to FIG. 32, the appropriate button of copy selector 808 is set for the number of copies desired, i.e. 3 and document handler button 822, sorter select button 825 and two sided (duplex) button 811 depressed. The originals, in this case, two simplex or one-sided originals are loaded into tray 233 of document handler 16 (FIG. 14) and the Print button 805 depressed. On depression of button 805, the host machine 10 enters the PRINT state and the Run Event Table for the exemplary copy run programmed is built by controller 18 and stored in RAM section 546. As described, the Run Event Table together with Background routines serve, via the multiple interrupt system and output refresh (through D.M.A.) to operate the various components of host machine 10 in integrated timed relationship to produce the copies programmed.

During the run, the first original is advanced onto platen 35 by document handler 16 where, as seen in FIG. 41, three exposures (1ST FLASH SIDE 1) are made producing three latent electrostatic images on belt 20 in succession. As described earlier, the images are developed at developing station 28 and transferred to individual copy sheets fed forward (1ST FEED SIDE 1) from main paper tray 100. The sheets bearing the images are carried from the transfer roll/belt nip by vacuum transport 155 to fuser 150 where the images are fixed. Following fusing, the copy sheets are routed by deflector 184 (referred to as an inverter gate in the tables) to return transport 182 and carried to auxiliary tray 102. The image bearing sheets entering tray 102 are aligned by edge patters 187 in preparation for refeeding thereof.

Following delivery of the last copy sheet to auxiliary tray 102, the document handler 16 is activated to remove the first original from platen 35 and bring the second original into registered position on platen 35. The second original is exposed three times (FLASH SIDE 2), the resulting images being developed on belt 20 at developing station 28 and transferred to the opposite or second side of the previously processed copy sheets which are now advanced (FEED SIDE 2) in timed relationship from auxiliary tray 102. Following transfer, the side two images are fused by fuser 150 and routed, by gate 184 toward stop 190, the latter being raised for this purpose. Abutment of the leading edge of the copy sheet with stop 190 causes the sheet trailing edge to be guided into discharge chute 186, effectively inverting the sheet, now bearing images on both sides. The inverted sheet is fed onto transport 181 and into an output receptacle such as sorter 14 where, in this example, the sheets are placed in successive ones of the first three trays 212 of either the upper of lower arrays 210, 211 respectively depending on the disposition of deflector 220.

The present invention is especially concerned with the case where an odd number of single sided or simplex original documents 2 are desired to be copied in a duplex mode, as compared with an even number as just previously described. It can be realized that in such case the machine need only copy an image of the last original onto the front or first side of the copy sheet 3. For example, where three original documents are to be copied in a duplex mode, both sides of the first copy sheet will contain images, while only the front side of the second copy sheet need contain an image. The machine could cycle the second copy sheet through the normal duplex mode by routing the second copy sheet back to the auxiliary tray 102 and then completing the cycle previously described with a blank image being transferred to its backside. However, this would cause unnecessary machine operations resulting in degradation of the machine speed throughput capability since there is no informational image being transferred to the backside of the last copy sheet. To optimize the throughput of the machine this invention provides a control system, under the command of controller 18, which integrally controls various machine components so that the copy sheets are routed directly to the output receptacles when the number of original documents to be copied in the duplex mode is odd in number.

For purpose of illustration assume that there are three originals 2 placed in the document handler 16 (see FIG. 14) with the separator or bail bar 235 being placed on top of the last original. Again for purposes of illustration assume that only one duplex copy of this set is desired to be made. The appropriate buttons of operator console 800 are then pushed as previously described, with specified routines, e.g. the Duplex Select routines 2SD CPY and MIN DUPL of Table XIV, setting flags in the computer memory indicating that the machine is in the duplex mode. Similary, another flag is set indicating on which side of the copy sheet the image is to be made. For example, a side 1 flag is set for the first original, a side 2 flag for the second original, with the side 1 flag being reset for the third original, etc. The controller 18 then progresses through its various states in the Background or State Checker routine (see Table I).

Referring now also to FIG. 41, when the controller 18 reaches the PRINT state, the Automatic Document Handler (ADH) Control routine (ADH CTRL of Table XI) is periodically called, here every 10 milliseconds. Under the command of this routine, the document handler 16 advances the first original 2 onto platen 35 where a latent electrostatic image is formed on belt 20 which is developed and transferred to side one of the first copy sheet 3 fed from main paper tray 100. Turning to FIG. 12, the first copy sheet proceeds through fuser 150, is directed by deflector 184 to return transport 182 and is carried back to auxiliary tray 102 with the image bearing side (side 1) facing upwardly. Under control of the ADH Control routine, the first original 2 is removed from platen 35 via auxiliary transport 276 of document handler 16 and placed on top of bail bar 235. The second original which has been previously fed to a "wait" station underneath sensor 280, is then fed onto platen 35. Now the image of the second original is transferred to the backside or side 2 of the first copy sheet fed from auxiliary paper tray 102 via transport 147. The first copy sheet, which now contains images on both sides, is finished by permanently affixing the transferred image by fuser 150. Deflector 184 which remains in its operative state when sorter 14 is selected now directs the copy sheet via inverter roll 185 towards return transport 182. However, now the stop 190 is activated to intercept the sheet wherein abutment of the leading edge of the copy sheet with stop 190 causes the sheet trailing edge to be guided into discharge chute 186, thus effectively inverting the sheet. The inverted sheet is fed onto transport 181 and into sorter 14 with the side one image facing the bin or tray.

Referring back again to FIG. 14, the second original 2 is fed back to copy tray 233 on top of the first original, which in turn, rests on bail bar 235. The last original, here, original number three which is odd in number, is then fed onto platen 35. However, previous to this, the last original has been fed to the "wait" station underneath sensor 280. When this last document left tray 233, bail bar 235 activates a switch 259 thereby signaling (by setting a flag, LST ORG) that this is the last original in the set to be copied.

The activation of switch 259 is sensed by the ADH Control routine which causes it to call another subroutine, the Input Empty routine (LEDGIEMP) shown in Table XII. The Input Empty routine checks flags in specified registers in the computer memory to determine whether the machine is in the duplex mode and whether the image of the last document is to be made on side one of the copy sheets. If so, a specified coded bit or flag, herein termed the odd last flag (ODD LAST), is stored in a predetermined location in the memory. In such a manner, the controller realizes that the last document to be placed on platen 35 by the document handler 16 is to be duplex copied and is odd in number. It should be realized that the entire set of original documents need not be odd in number for this to occur. For example, if there were four originals in the set, but only the last three originals were desired to be duplex copied, the last original would still be an odd number to be duplex copied.

The State Checker routine of Table I while in the main body of the PRINT state subsequently calls a Shift Register Scheduler routine shown in Table XIII. As previously described, the Shift Register Schedule routine controls the timing of the activation of various machine components, of which primary concern here is the activation of deflector 184. The shift register schedule routine reads the specified location in the memory which now contains the odd last flag. Accordingly, this routine provides a signal which indicates that the inverter gate or deflector 184 should be removed from the paper path and schedules at which time the event should occur, with the routine again storing this information is a specified location in the Run Event Table of FIG. 35.

The Inverter Gate Control routine of Table XV controls the operation of deflector 184. This routine is a fixed pitch event (see Table II) which is entered through the Machine Clock Interrupt routine (Table IX) at a predetermined clock count. The Inverter Gate routine interrogates the information for this copy run which has been stored by the Shift Register Scheduler routine. If that information indicates that the inverter gate should be removed from the paper path, it calls another routine (HOLDSOFF of Table XVI) which removes the deflector 184 from the paper path.

Hence, when the image of the last original, here, original number three is transferred to the second copy sheet fed from main paper tray 100, the deflector 184 is removed from the paper path so that the finished copy sheet proceeds directly to the sorter 14. In such manner, the last copy sheet is not returned through the duplex cycle, i.e. to the auxiliary tray, for subsequently forming an unneeded copy on its backside when the number of originals to be duplex copied is odd. Accordingly, the throughput of the machine is optimized by the limiting unnecessary machine operations.

The various features of the present invention have been illustrated according to the patent statutes by describing a reproduction machine, a programmable digital computer, and programs for instructing the computer to carry out the claimed functions. However, it should be understood that the spirit of this invention can also be performed by hardwired circuitry if it is desired to do so, for example, by integrated circuit devices which contain the same basic elements which are only temporarily utilized by the computer when instructed by the software programs. Therefore, while this invention has been described in connection with particular examples thereof, no limitation is intended thereby except as defined in the appended claims.