Note: Descriptions are shown in the official language in which they were submitted.
3~
~ACKGROUND OF THE INVEN?~ON
5 The present invention relates to postage meters and .
more particularly to an electronic postage nletering system
inclllding zip code-to-zone convession.
The type of postal .scale which is in widespread
commercial use at present i5 a mechanical or electro-mechanical
d~vice for deriving postage as a func~ion of package weight
al~d destination zone.
While the Postal Service still uses zones for purposes
o~ calculating postage on packayes mailed from one part of
thse country to another, most people are not aware o~ which
15 ~one a particul.ar destination ~alls in. They ~re, however,
generally aware of the zip code at the destination. To permit
: a ~user to make a conversion from destination zip code to desti-
na~i~n ~one, the Postal Service publishes charts showing desti~
na~ion zones relative to a specific city of origin as a function
o~ the first thrèe digits or prefix of destination zip codes.
The Postal Service also publishes another chart tabulating
postage as a function of dif~erent weight-zone combinations. i
- A user consults one chart to determine the proper zone and
then, a~ter weighing the package to ~e mailed, consults the ;:
,
other chart to find the proper postage. The user employs the ,i
re~rieved postage entry to manually set a conventional postage
meter to imprint the postage on a tape which can be affixed
- . t~ the package.
s
U.S~ Patent 3,636,297 - Salava, discl~ses a 1 :
. . .
; co~æuter-type postage calculator in which the prefi~
. -2
` ' ' ` ' (
- - - _--, . - - - - --- - - ;
;~
,.. , , , - , - ::
,: . :, :
, : ,
. :~ , . , : : :' ' - -
.
~03~
of a destination zip code is converted to zone information
through the use of a look-up table in which zones are
stored as a function of zip code prefixes. The table
is scanned in numerically ascending order until a corre.s-
pondence is found between a user-entered destination zip
code prefix and one of the addresses in the table.
Signals representing the parcel weight, destination zone
and class of handling are apparently algebraically added.
The results would not appear to be consistently accurate.
The calculator apparently would establish the same postage
for a two-pound package being sent to zone 4 at parcel
post rates as it would for a four-pound package being
sent to zone 2 at the same rates. However, the Postal
Service has established different postages for these
two conditions. Moreover, the required memory or data
storage capacity for such a system would be large and
costly.
`'~'; , .
-,
,, ' ' .
-
- .. : ~, -: :
., , : :
:: ~: ,~
~ ~3~t~
SUMMARY OF T~IE INV_ION
'Lhe present invention is an improved postage
metering system for automating the task of computing
postage for packages being sent by any class of mail
selected from available classes.
The l.nvent:Lon relates to a postal conversion
apparat~ls for converting a Eirst postal d~signation to a
second postal designation, the second postal designation
being a function of the first postal designation~ the
postal conversion apparatus compris:lng: storage means
containing incremental first postal designation data
relating to first postal designation information and
second postal designation data being stored in a
predetermined storage sequence in relation to the
; incremental first postal designation data; means for
entering first postal designation inEormation for accessing
the incremental Eirst postal designation data of the
storage means; means for retrieving the incremental first
postal designation data ~rom the storage means in successive
increments in response to a first postal designation
information input; determination means responsive to ~
incrementally retrieved data for determining whether an :
accumulation of the incrementally retrieved data~equals or
exceeds a numerical value relating to at least a portion
- of the first postal designation information; and means
responsive to an output from the determination means for
retrieving a second postal designation from the storage
means that relates to the accumulation of the first postal
designation data increments.
.. ,,~
mb/\~ ~
' ' ' ' ,
:
1 ~ 3~4
The function of the system is to set a postage
printer in a postage meter. This system includes input
means for generating weight-representing signals and
input means for providlng signals representing the
destination zip code Eor the package. The system also
includes means for deter~lining the destination zone as a
function of a destination zip code and means for computing
the proper postage as a function of both the weight-
representing signal and a destination zone signal. Finally,
the system includes a meter setting means for translating
the postage-representing signals to sett:ings for the
postage printer.
The postage computing means includes means for
selecting a sequence consisting of a minimum postage
amount and incremental postage amoun~s. The weight-
representing signal is then successively decremented while
a postage-representing signal is synchronously incremented
by the minimum and incremental postage amounts. The
decrementing/incrementing operations cease when the
decremented weight-representing signal is found to be less
th~n or equal to a predetermined number. The incremented~
postage-representing signal is appIied to the postage
printer.
'- "':
,.~ ~.,.
mb/J~ - 4a -
.
~ ,
. ,: .
~1~3B~7~
DESCRIPTION OF THE DR~WIN~S
While the specification concludes with claims
particul~rly pointing out and distinctly claiming that
which is regarded as the present invention, :Eurther de-
tails Oe partieular embodiments of t~le invention may be
morc readily ~scertained from the following detalled
description when read in conjunction with the accompanying
drawings wherein:
FIGURE 1 is a functional block diagram of the
10 computerized postage meter system into which the present
invention could be incorporated; ~-
. PIGURE 2 is a perspective view of a housing
for the meter, including a scale mechanism;
~ IGURE 3 is an enlarged plan view of the key-
15 board display for the meter shown in FIGURE 2;
FIGURE ~ is a perspective view of a postage ~ -
printer~ .
FIGURE 5 is a block diagram of components of
t the postage meter system shown in functional form in
20 FIGURE l;
FIGURE 6 is a flow chart of the generaliæed
- overall operation of the system shown in FIGURE l;
FIGURE 7 is a flow chart of the æip-to-æone
conversion routine;
2~ FIGURE 8 is a flow chart of a routine for
calculating surface parcel pcst postage;
FIGURE 9 is a flow chart for a routine for
calculating library rates postage;
_5~
~ , :
. ~ .
.
:: ' ~ i . , ,
~ '
~3~3~3~
FIGURE 10 is a flow chart of a routine for
ealculating book rates postage;
Page 11 is a functional block diagram of a
: random loqie implementation of the invention;
S ~`IGURE 12 i9 a elass-of-handling selection
and loekout cireuit for the system shown in FIGURE 10;
FIGURE 13 is a block cliagram for the zip-to-
zone eonversion eireuits of the system of FIGURE 10;
FIGURE 14 illustrates a modificaition to the
eonverter of FIGURE 13 for providing an alternate
mode of operation; and
FIGURE lS is a sehematic diagram of the
postage eaileulating eircuit for the system shown in
F~GURE 11.
~: , ' ' ,
,
~`
' "'
, '
.-
:. '
.
.''; , , '
"
~;
.
7 ~ .
. ~,~ , .
~10
~ ;,, ~' : . .
' ~ : ' . ` ` '':~
.` ' ' " ', ' i ~., `
,
:~
~33~
DETAI LED DESCRI PTION
Refe~riny to FIGURE 1, th~ general functional
arrangement of ~ computerized postage meter incorporating
the present invention i9 shown. The system includes a
central processor unit ~CPU) 30 which operates on input
data ~nd controls th~ 10w of data between various
memory units. One type o~ memory unit employed with
the central processor unit 30 is a permanent, read only
memory 32 which stores the specific sequence of operations
to be performed in calculating postage and the sequence
of operations fox other routines employed within the
system. A second type of memory unit employed is a
read/write random access memory 34 which is used to hold
and forward working data needed by and generated within
the central processor unit 30.
~ n additional memory component coupled to the
central processor unit 30 i9 a non-volatile random
access memory 36 which operates on and stores certain
critical information employed in the postal system.
The critical information includes working data
representing crucial accounting functions such as
the contents o~ descending register and an ascending
register. In one embodiment, the non-volatile memory
36 may be a CMOS random access memory with a hattery
back-up unit for holding the stored data in the event
of a loss o~ power to the sys~emO
Data and commands can be inputted to ~he CPU
30 through an input keyboard 3;3. Data can include
_ 7 _
:
': ' : .
' ~ : ' . ~ " `
~:
8~
directly entered postage values, destination zone values
~nd destin~tion zip code prefixes. Commands include a
display command which transfers the contents of memory
34 or memory 36 throu~h CPU 30 to an output display de-
- 5 vice 40. Input/output signals may be multiplexed by a
,' multiplexer 42 interposed between the central processor
30 a~d input/output components 3~ and 40. Central
proc~ssor unit 30 is also linked to scale interface
circuits 27 which provide binary weight-representing
signals to the system.
When appropriate postal data and commands
have been entered into the CPV 30 through keyboard 38
and weight-representing signals have been received
from scale interface circuits 27, CPU 30, under control
lS ~ programs stored in read-only memory 32, generates
postage-representing signals which are applied to a
postage printer 44 in a postage meter. When the meter
has been set to the appropriate values, a print command
. generated by a user-controlled input to CPU 30 causes
postage to be imprinted directly on an envelope or on
~- a tape to be affixed to an envelope or package.
One example of a system into which the present
invention may be incorporated can be found in co-pending ~-
United States Application S.N. 536,248 filed December 23,
. 25 1974, for a Uicro Computerized Electronic Postage Meter
~: .
5ystem, which application is assigned to the assignee of
the present invention.
- FIGURE 2 shows one type o~ housing for a
system incorporating the invention. The input keyboard
; 8
:~ .
,, ~,
,. .
:
: , . . . : :
38 and output display 40 are mounted in panel 46 on the
top surface of a housing 48. The postage printer may be
contained in a forward section 50 wherein the postage may
be imprinted either on envelopes, such as envelope 52
or on tapes (no~ shown) to be aPixed to packages. ~s
the p~esent inv~ntion pertains to calculation oE postage
for packages, the description Oe the operation of the
meter is llmited to calculation of package postage.
The meter includes a scale mechanism 54 on which a
package to be mailed can be deposited.
Scale 54 includes a suitable transducer mechanism
for converting displacement of the scale tray to an en-
coded binary signal which is supplied to the scale inter-
face circuits 27 within housing 48. The signals may or
may not be contemporaneously displayed on output display
The destination zone of a package, if known,
may be entered directly into the system through the 0-9
numerical pushbut~ons 56. If the destination zone is
;` 20 not known, the prefix (first three digits) of the
`~ destination zip code is entered into the system
through the pushbuttons 56. A ZIP-ZONE key 58 initiates
conversion of the destination zip code prefix to a
destination zone value.
The class of handling of the package (surface ;~
rates, book rates or library rates) is selected by means
o~ pushbuttons 60, 62, 64, respectively.~ Once the des-
:
tination zone, package weight and class of handling are
introduced into the system, the proper amount o postage
g_ , :
'
' '`'"'
.,
: .
' ' '
:.. : . , ~ ' :-
is automatically calculated and displayed on output display
40. A Set button 66 must be depressed before the postage
printer is set~ Once the meter has been set, an imprint
button 68 is depressed to start the actual print operation~
While the illustrated embodiment of the invention
permits calculation of postage ~or the three specified classes
of mail, the invention could easily be extended to performing
calculations for other classes such as priority mail or UPS
service.
FIGURE 3 is an enlarged view of panel 46 showing the
display 40 and the keyboard 38 in more detail. The keyboard 38
includes the numerical pushbuttons 56 used for entering the
amount of postage required (if known), a destination zone or a
destination zip code prefix into the system. Pushbuttons 70, 72,
74, 76, 78 and 80 control the display of register contents for
batch count, batch amount, piece count, control sum, ascending
register and descending register, respectively. When one of
these buttons is depressed, a numerical section 82 of display
40 is cleared. The content of the appropriate register are
loaded into and appear at numerical display 82 while the appro-
priate indicator lamp in a back-lighted legend display area 84
is energized.
The function of the various registers are described
briefly below. Batch count and batch amount registers contain
a running account of the total numher of pieces of mail processed
during a single run and of the total postage expended for this
mail. These
ob/ ~ - 10 -
': .
~3~
registers can be re~e~ to zero by the user, permitting
each of several departments in a large organization to
easily keep records of their postal activities. A piece
count re~ister indicates the total ~umber of postage
s printings (pioces Oe mail) th~ machino has performed.
.' The piece count register differs from the batch count
r~cJister in that the former is not resettable by the
user. Th~ piece count :information is used to deter-
mine when the system may require servicing and main-
10 tenance and for accounting purposes. The ascending
and descending registers serve standard functions.
The ascending register gives a running total o all
postage printed during the life of the meter and the
descending register informs the user of the amount
15 of postage funds still remaining in the postage system.
The control sum register provides a security check for the
descending and ascending registers. The control sum,
which must always correspond with the su~med readings
of the ascending and descending registers, is the total
- 20 amount of postage ever put into the machine.
A +- key 86 allows the user to add special
charges to the calculated postage such as special
deLivery or certificate charges and the like. A Clear
key 88 clears the numeri display 82. If the contents
25 of ona of the batch registers is displayed when~the
~ Clear key is actuated, that register is set to zero.
-~ ~he Set button 66 is depressed after the
required postaqe has been calculated and any special
charges added through numerical pushbuttons 56. .
. '~ . ',
` .
: ' :
8~)~
For one type of meter, depressing the set button causes the
print wheels in a print drum within the met~r to be set to
the desired postago.
A S unlock key 90 mllst be depressed by a user in
order to set postaye ~qual to, or in excess of, one dollar.
This extra physical step acts to prevent costly postage
printing mistakeq.
The postage meter housing 48 includes a hinged
security door or plate 92 having a latch 94. This latch
secures th~ door 92 to the housing 48 by means of a wired
lead seal 96. Postal authorities are the only ones em-
po~ered to open the seal 96. The door 92 protects switches
98 and 100 shown in phantom. Switch 98 enables the com-
puterized system to call into operation a routine which
provides for the entering of postage funds into the
system. Postage funds may be entered into the system by
first keying in the amount of postage to be added through
the numerical pushbuttons 56. This amount of postage
appears on the display and is added to the descending and
control sum registers of the postage meter system by
; opening security door 92 and pressing button 98. This
button initiates a jump in the postage meter program to
the above-mentioned routine. After the routine is executed,
the door 9Z i~ again secured by a seal 96.
Switch 100 is provided for removing funds from
the de:cending and control sum registers in the event a
- ~ ` mistake is adding funds has occurred. The nee~ for adding
funds to the system is signaled by an Insufficient Postage
indicator lamp 102.
-12-
,
- - : ., . . :
.
~ . .
7 . .
' " ' '
~ 1~3B~
-
A Check Date reminder is provided b~ in~icator
104 each time the postage meter system is turned on.
A Meter Enabled Indicator 106 lights when it is
established that ~a) the meter print drum has been set to
the proper posta~e; (b~ the postage to be imprinted appears
dt thQ numerical display 82; and (c) sufEicient funds are
availahle to imprint the posta~e desired.
Indicator lamp 108 signals the operator to call
a service man. This indicator is energized for certain
types of system errors, e.g., when the control sum is not
equal to the sum of the ascending and descending registers.
When such errors are detected, the meter may be automatically
disabled to prevent further use. A service man would, of
course, be able to restore the meter to its normal operating
mode once the error is corrected.
` Indicator 110 signals the operator that the
postage to be set is equal to or more than $1.00 and that $
unlock button must be depressed before the set button 66
will function.
An indicator 112 is energized when the contents
of the ascending register are displayed in numerical dis- -
play 82. An indicator 114 is similarly energized when
the contents of the descending register appear on
` numerical display 82.
The batch amount indicator 115 and the batch
, ::
counter indicator 116 are energi~ed when the contents of
the respecti~e registers are belng displayed. The piece
count indicator 118 is enegized when piece count information
13- ~
. , .
.. .
,
'~
: . . ,
:,-
: ' ,
~3~3~
is displayed. Similarly, a control sum indicator 120 isenergi~ed when the control sum is displayed on numeric~l
display 82.
~ low postage ( ~ $100) indicator 123 informs
S an op~rator that th~ Eunds rem~ining in the descending
register are currently below ~100. 1'his should alert
the operator that he will need to recharge the system
with additional postage funds in the near future.
An embodiment of one meter setting mechanism is
illustrated in detail in FIGURE 4. The meter is a modified
Model 5300 postage meter manufactured by the assignee of
~his invention - Pitney Bowes, Inc., Stamford, Connecticut.
The modified meter includes a print drum 122 and print
wheel driving rack 43 from the Model 5300 postage meter.
Mechanical registers and actuator assemblies have been
removed. Print wheels (not shown) within print drum 122
of the modified meter are set by a mechanism driven by a
; stepping motor 124 and a pair of soleno.ids 126 and 123.
The stepping motor 124 drives an upper and lower set 43
20 of racks 43a, 43b, 43c, 43d through an upper pair of nested
shafts 130a, 130b, and a similar, lower pair of nested
shaf~s (not shown~.
The print drum 122 has four print ~heels (not
` shown) which provide a postage impression to a maximum
sum of $99.99. Each print wheel provides a separate
digit of this sum, a~d is settable from "0" to "S".
Each of the print wheels i~ set by means of one of the
four drive racks 43a, 43b, 43c, 43d. The drive racks
lide in the directions indicated by arrows ]31 within a
drum shaft 57.
.
`
;~`;; ':
::
The upper racks 43a, 43b are driven by pinion
gears 132a, 132b respectively. The lower racks 43c and
43d are controlled by a simil~r set of pinion gears. The
pinion gear 132a is mounted on shaEt 130a. The pinion
132b is mounted on shaft 130b. The pinion ge~rs for the
lower racks 43c, 43d are similarly mounted on the lower
set o~ nested shafts. These shafts are rotated in -the
directions indicated by double-headed arrow 59 by means
of spur gears, the upper pair 53a, 53b of which are shown.
A master gear 51 engages each of the spur gears
in succession to sequentially set the print wheels for
"tens of dollars", "dollars", "tens of cents'' and "units
cents" in the meter. The master gear 51 can be shifted
laterally (in the directions indicated by double-headed
arrow 65) into a meshing relationship with each of the
spur gears within a yoke 63 which slides on splined
shaft 134. The master gear 51 is mounted in a slot 136
in yoke 63 and can be rotated in either direction by the
stepping motor 124 through motor shaft 124a and splined
20 shaft 134. A sleeve bushing 138 separates yoke 63 from
the splined shaft 134. The yoke 63 and master gear 51
are guided and supported by an additional smooth shaft
61 which nests within a slot 67 of yoke 63 and prevents
rotation of the yoke due to any slight friction between
the mating surfaces of the yoke and sleeve bushing 138.
To assure alignment of the teeth of the master
gear 51 with the teeth of the several spur gears, a pair
of upper and lower ~ooth profiles are formed on the
- , ~ .
,
~338~ -
adjacent surfaces of the yoke 63. Only the upper profile
140 is shown. As the yoke 63 slides in the direction in-
dicated by the double-headed arrow 65, one tooth on each
of the spur gea~s is locked into place between the tooth
profiles. Each of the gears is free to turn only when the
m~ster gear is meshed with it. The tooth profiles also
s~rve the additional function of locking the spur gears
in place once the meter is sot to prevent anyone from
attempting to tamper with the meter by manually altering
the print wheel positions from the exterior of the meter.
Lateral movement of the yoke 63 is controlled
by a toggle pin 71 seated in a groove 142 of the yoke 63.
,~ - The toggle pin 71 pushes against the yoke 63 when a
pivotable link 73 to which it is attached is made to pivot
15 (arro~s 144) about a center shaft 75. Movement of link
73 is controlled by the two solenoids 126 and 1~8 acting
through pivot arms 146, 148 and 77, 154 respectively.
The solenoids 126 and 128 pull on their respective
pivot arms 146 and 77 through pull rods 150 and 79 which
are pinned to the pivot arms by pins 81 and lS2 respectively.
When the pull rod 79 pulls upon arm 77, the arm pivots about
a shaft 83. When this occurs, arm 154is caused to be
pivoted against the biasing action of a spring 156. This,
in turn allo~s a shaft 158 to pull pivot arm 73 forward
or in a direction indicated by arrow 89. The forward
movement of pivot arm 73 about center shaft 75 causes
the ~oggle pin 71 to move rearwardly or in the direction
indicated by arrow 91.
,
-16-
" '
~'
Thcre are four combined solenoid p~ll positions
corresponding to the four s~parate mating positions betwPen
the master gear 51 and the spur gears: neithcr solenoid
energiz~d: both solenoids energized; solenoi~ 126 only
s energized and solenoid 128 only energized. ~aster gear
51 is opposite a different one of the ~pur gears for each
different combination o~ ~nergized solenoids. When all of
the spur gears have been rotated to selected positions of
the master gear 51, causing the racks 43 and the print
wheels ~not shown) to assume postage value positions, the
drum 122 is ready to be rotated ~y shaft 57 in the direc-
tion indicated by arrow 97 to actually imprint the postage.
Tha home position of drum 122 is monitored by
a slotted disk 156 mounted on shaft 57. When a slot 158
` 15 on disk 156 moves into an optical detector 99, the print
cycle is completed.
All optical detectors in the setting mechanism
comprise a light emitting diode (LED) and a phototransistor
` for receiving light emitted by the LED. The latPral
position of master gear 51 in yoke 63 is monitored in-
- directly by monitoring the pivot positions of pivot arms
146 and 77 rQspectively. Pivot arm 148 has a finger 101
which pivots into and out of a detector 160 when the
- solenoid 128 is energized and de-energized.
The home positions of shafts 130a and 130b are
monitored by slotted~disks 105a, lOSb, respectively. When
a slot in disk 105a is ~ithin an associated optical detector,
; shaft 130a is at zero. Similarly, when a slo~ in disk 105b
.; i9 in well 107b, sha~t 52b is at zero. The lower pair of
.
` '~'
i~ '
`. . ~
, ~
::
3~
nested shafts are monitored through similar apparatus
(not shown).
Rotation of the steppinq motor shaft 124A, is
monitorcd through gears 162 and 16~, slotted monitoring
wlleel 109 ar)d monitoring well 166. When stepping motor
shaft 124a rotates splilled shaft 134 and master ge~r 51,
cJear 162 rotates through the same anyular increment.
Gear 162 intermeshes with gear 163 which is attached to
the slotted monitorir~g wheel 109, This train of gears
causes wheel 109 to turn through the same angles as
shaft lZ4a. Every fifth slot 111 on the monitoring wheel
109 is extra long to provide a check on the setting
mechanism. Each slot in wheel 109 corresponds to a change
of one unit of postage value. The slotted wheel 109 is
optically monitored by detector 166. Detector 166 has
two photosensors. The first photosensor is located near
the periphery of slotted wheel 109 and senses every step
of the stepping motor 124. The other photosensor is
located near the center of the slotted wheel 109 and senses
every fifth step. By counting the number of single step
movements and determining whether a count of five exists
when slot 111 i5 aligned with detector 116, it can be
determined whether all single step movements have been
properly sensed.
~5 Referring now to ~IGURE 5, a block diagram of
a suitable computer control is shown. The system is
m~de up of components generally included in a MCS-4
microcomputer component set which is a product of Intel
Corporation, Santa Clara, California. This set of
-18-
j
.
.
., . ~ .. . . .
. ~: ::
':~ , ' :~ .. .
~3~
components includes a central processor unit ~CPU) 10
which is connected to a number of read only memory (ROM)
units 11, 12, 13, 14 and 15 and to a number of random
access memory ~R~M~ units 16, 17, 18 and 19. Random
access memory unit 17 is made to op~rate as the non
volatil~ memory unit through the use of the battery
backup ~Init. As discussed earlier, uni~ 17 is used to
store critical accounting data. A ~umber of shift
registers (S/R) 20, 21 and 22 are connected into the
system through output port 25. In one commercially
available device, output port 25 would be physically
located on the same chip as random access memory unit
16 but would function independently. Each output port
has four binary-value output lines as shown. The read
15 only memories 11, 12 and 13 also are associated with
input/output ports ~I/O) 429, 430 and 431 respectively,
each of which has a four-bit capacity. Although the
input/output ports are physically located on the chips,
they are logically independent of the read only memories.
The shift registers 20, 21 and 22 provide
port expansion for the postage meter system. In
addition shift register 20 provides a multiplexing
capability digit selest drivers 436 of numerical
display 82 and for a keyboard and meter setting
25 detector matrix 23. Shift registers 21 and 22 are
serially connected to provide an extended length register
for controlling driver circuits 21a and 22a for indicator
~ lamps 21b and solenoids 22b, respectively. Solenoids
: ~ -19-
- .
. ~ ~
i
!~
.
`~ ' ` '` ' `
'. `
. ~ . : .
22b are the meter settin~ solenoids lZ6 and 128 described
earlier.
The numerical display 82 is controlled by decoder/
driver 446 which is connected into the s~stem through shift
i register 20. One lead of output port 25 provides a blanking
control signal to the decoder/driver 946 to elimlnate
leading z.eros in the numerical display 82.
Tha inputs from the keyboard detectors matrix
23 are fad to the system through input/output 429.
Postage requests and multiplexing select signals
are applied to scale interface circuit 432 from output
port 433. The scale signals are applied to the system
through input/output port 431.
Stepping motors in the meter setting mechanism
are controlled by driver circuits 434 connected to the
system through output port 435.
A computer system of the particular type des-
cribed employs logic level voltage power supply 438. A
power sense/reset circuit 439 is intèrconnected into the
system to detect power failures, When a power failure
or unacceptable low voltage is detected by power sense/
reset circuit 439, the system updates the contents of
the non-volatile memory 17 as part of a shutdown routine.
A clock 441 serves to corractly phase the :~
i operations of the system. Two non-overlapping clock
signals ~1 and 02 are supplied to the system by clock
441.
` -20-
- , .
, . :
- f
The CP~ 10 generates a SYNC signal every eight
clock periods as shown in the Users Manual for the MCS-4
microcomputer set, copyright 1972, Figure 2 on page 6
thereof. The SYNC signal marks the beginning of each
instmction cycle. The ROMs and RA~s ge~erate internal
timing usincJ SYNC, 0L and 02. The shiEt registers ~re
statia devices and do not employ these clock pulses in
their operation.
Referring to FIGURE 6, the overall operation
of the system is represented in simplified flow chart
form. When power is first applied to the system as
~shown in operation 300, a general reset system pulse
initializes the total system. This system reset pulse
causes the CPU registers, RAM memories and input/output
ports to be cleared and initiates execution of a postage
meter program. The print wheels of the mechanism
shown in FIGURE 4 are set to zero if they are not
zeroed already.
Once the system has been initialized, a scan
routine begins. This routine is shown generally as
operations 302, 303 and 308. The scan routine searches
for a depressed key on keyboard 38 and multiplexes the
numerical display 82. When a validly depressed key lS
detected at block 308, the scan routine branches to the
appropriate subroutine corresponding to the function
; called for by that particular key. The scan routine
retrieves an address of a subroutine called by the key
rom storage in a "look up table". The stored address
-21-
~ ' ' . .
.,
o
., .
.
.
~' ' '' ~ " ' ' ' ., .
~U3~
is transferrcd to a register in the CPU 10. A subroutine
is then executed to provide a ~u~p to the address stored
in the register. After a particular key is serviced as
indicated at 310, the scan routine is reentered to look
S for new inputs erom the keyboard 38.
~ lring the co~rso o~ a scan routine, the power
status of the system is periodically checked as indicated
at 303 In case of a power failure, the posta~e meter
system must be able to complete any current operations
including the loading of critical accounting data
into non-volatile random access memory unit 17. When
the current operations have been completed, the system
enters a trap at block 306. The program cannot reenter
the scan routine except by the in.itiation of a complete
"power-up" sequence.
The system described with reference to
FIGURES 1, 5 and 6 utilizes a number of programs
which, together with explanatory appendiums are printed
as an Appendix to this specification. The programs in-
clude a number of routines and subroutines which aredescribed briefly below.
A SCAN routine, described as part of FIGURE
6, multiplexes the display and searches for keyboard
inputs. The SCAN routine is entered upon completion of
an INITIALIZATION routine which clears the CPU registers,
RA~ registers, and I/O ports. The SCAN routine
periodically calls up a FCTN subroutine when a depressed
key is sensed. The FCTN subroutine services the key;
-22-
.~ ' ` .
, , ;
7 '
. . . , -
',~; , '` ~ ~ ' ; '
~3~f~
i.e., perform~ the function requested by the key. Periodic
checks of the power condition of the system are made and
a DOWN suhroutine entered if operations must be wound up.
Th~ initialization procedure includ~s a C~{CK routine used
to detect ~hether the ascending register plus the descending
register minus the control sum eq~lals zero. If not, the
CHCK routine energi~es "call service" indicator 108 alld
disables the meter~
A number of subroutines are used to control
the operation of the meter setting mechanism. A HOME
subroutine is employed to set the four print wheels of
the meter to their home or zero position. The photocells
used to monitor the home position are read in the course
- of a ZEROB subroutine called up when the print wheels are
being set to ~ero. A STPB subroutine, which selects the
print wheel to be set, is included as part of a meter MAIN
; routine. Another included subroutine is a S~EP routine
used to change the setting of a selected print wheel by
one unit. The solenoids which control the lateral position
of the yoke for the master gear are under the control of
the STP~ subroutine.
; There are, of course, a number of "housekeeping"
subroutines. The CLEAR subroutine is multifunction in that
it (1) clears the display, (2) recalls the contents of ~n
addition register into the display, (3~ clears the
addition register of the second successive clear, and (4)
clears both the batch count ar.d batch amount registexs if
the contents of either are displayed when the subroutine
is called. An ADDD subroutine is used to increment or
23
.
. ' ~ -
.
'
,~
: . - : ,
' :. ' :, ', ' ~ . ' ' ` '
: ~ .
- ,,
. : :
': `
~ \
3~
decrement a selected meter register. A PLUS subroutine adds
the contents of the display to the addition register and
stores the result in both the addition register and the
display recJister. A CLDSP subroutine writes "0's" into the
display while a CI,F:ER subroutine writes "0's" into an area
,' specifiecl by a preset index register. A FETC~I subroutine
initializes an index register to specify the metex register
being called into operation.
The programs also include a CM*AR subroutine for
comparing the contents of a meter setting register
against the contents of the descending register to
determine if sufficient funds are available for the
proposed printing of postage. An UNLOCK subroutine
sets a $ UNLOCK flag to enable the printer if the
- 15 requested postage exceeds one dollar. The POST subroutine
updates meter registers each time postage is printed while
an ENBLE subroutine determines whether the printer may be
eslabled for a subsequent imprint for the same amount.
In the display routine, a LDLMP subrcutine
transfers data ~n an indicator register to a shift
register which drives the selected lamp display. An ~,
OUTPT subroutine is employed to enter a parallel- - -
presented 4 bit word into a display register in serial
fashion. ~`
The su~routine employed in adding funds to the
meter is identified as the ADP subroutine while SUBP sub- ^~
routine is employed for substracting funds from the meter.
As indicated earlier, a user may enter a des-
tination zone for a package to be mailed directly through
-24-
'
` '
,~v
..... . .
~ , . . .
;' ~
3~ 3~
numerical keys 56 or, in the alternative, may enter a
destination zip code prefix throuqh keyboard S6. If the
zip code pr~fix is entered, tha user depresses the Zip-
%one key 58 to initiate a zip-to-zone conversion.
The zip-to-zone conversion utilizes a AZIP
approach. When a meter is installed at a specific
location, conversion tables are generated with reference
to the Postal Service Official Zone Chart for that
location. A small portion of the Official Zone Chart
for use by mailers having originating Zip Codes 06801-
06999; i.e., the Stamford-Danbury area of Connectic-~t,
is reproduced in Table I below. This information was
taken from Zone Chart No. 068-069 issued by the U.S.
Post Office in May, 1969.
TAaLB I
Zip Code
Prefixes 2One
.
006 - 009...................... ..7
010 - 018.. ,................... ..2
019........ ~................... ..3 -
020 - 025..,...,.,, .,...,,2
026............................ ..3
027 - 031.,.. ,...... ,....... ,.2
032 - 033......... ,.,....... ..3
034................... ...... ..2
035.......... ,.............. .~3
036.......... ,.... .,,....... ..2
037 - 043.... ................ 3
-25-
:' ~ : . : ' ` -
3~
To establish the ~ZIP values employed in the
conversion process, the zip code prefixes in the Official
Zone Chart ~re isolated into groups of consecutively-
numbered zip code pre~ixes falling within the same zone.
ReferrincJ to Table II below, the zip code prefixes 006-
nos are consecutively-numbered pre~ixes which Eall within
zone 7. These preEixes comprise the first group of valid
codes. Zip code prefixes 000-005 are as yet, unassigned
by the Postal Service. For purposes of this invention,
these prefixes are grouped together and assigned a Zone "F"
to indicate no valid zip code exists. Zip code prefixes
010-018 are also consecutively-numbered but are within
zone 2. The change in Zone numbers requires that these
prefixes be grouped separately into another group.
The QZIP values represent the difference between
the numerical value of the highest zip code prefix in
one group and the highest zip code prefix in the
preceding group; i.e., the number of consecutive zip
code prefixes in a group. By way of example, the highest
zip code prefix in group 11 of Table 11 is 043 while the
highest zip code prefix for the preceding group, group 10
is Q36. The difference between 043 and 036 provides a
~ZIP value of seven for group No. 11. QZIP value and -
associated zone values are stored in memory in the same
sequence in which they appear in Table II and the
necessary continua~ion of Table II for the remainder
o~ the Official Zone Chart.
-
-26-
'
1,, - .
:; : .
TABLE II
Zip Code
Prefixes _one Gro~ No, ~ZIP
000 - 005 F 0 6
,' 5 006 - 009 7
010 - 01~ 2 2
019 3 3
020 - 025 2 4 6
026 3 5
027 - 031 2 6 5
032 - 033 3 7 2
039 2 8
035 3 9
036 2 ` 10
037 ~ 043 3 11 7
Referring to FIGURES 3 and 7, the zip-zone conversion
routine is called by depressing Zip~Zone key 5a in key-
board 38. The first step in the routine is to determine
whether a valid ~ip code prefix has been enetered fro~
the keyboard. The validity checks consists of deter-
mining that the entered prefix has ~ 3 digits and was
entered through the keyboard. I~ the entry did not
originate at the keyboard, th~ system is directed to an
error routine 172 in which an error sign is loaded into -
the display. The error routine then returns the system
to the normal scan routine, I~ a keyboard entry is confirmed, ~ -
:
-2?~
~, .
,~ .
. , .
.' .... ~
, ' .
~. '
,~
: :
3~-f ~
a check is made at block 174 as to whether the entered
prefix contained more than three digits. If more than
three digits were entered, the system enters the error
routine 17~.
If only t:hree digits were entered into the
,' keyboard, the address of the Group 0 of the L~ZIP locations
i5 loaded into CPIJ 10 at operation 176. CPU 10 retrieves
the L\ZIP valùe; i.e., a value of 6 for Group 1. The
entered zip code is decremented by the ~ZIP quantity in
operation 17B, and the result checked in operation 180
to determine whether the decremented result is less than
zero. If the result is greater than or equal to zero,
~ the Group address is incremented and the processes iterated.
!` The result of operation 178 is iteratively decremented by
15 ~ZIP values in consecutively numbered groups until operation .
180 shows that the result is less than ~ero. The zone -
value stored at the address of the las~ utilized ~ZIP Group ;~
is retrieved and transferred to the display. ;
', , .
::
- ,.
.~ :
' ' ' ~ ~.
:
,, ',
~ 2B-
:~ .
.
.,
., .
.. . .
:
:,
~33~
__ ~
~ ................ ~
_ ~. . __
. ,,
K V o o O O ~ O
. .~
~ ~ ~ o a~ ~ ~ ~
~ N O o o o o o o o
? H ..... ._
-1 _
O O ~i N
~ . ',,'
'.
_ --_
R ~1 ~ ~ .
.a
~Q . ~ : -
~: :
~' 29
' .
g~
TABLE III is a tabular summary of a zip-to-zone
conversion which occurs when a 029 2ip code prefix, selected
arbitrarily, is entered into the keyboard. This zip code
prefix is decremented in operation 178 by the ~ZIP values
associated with consecutively-numbered groups. For Groups
0-5, the result of operation 178 is greater than zero.
When the ~ZIP stored at the Group 6 address is used to
further decrement the zip value resulting from the
previous iteration, however, the result is a negative
value. The system responds to this negative value by
` retrieving the zone value stored at the Group 6 address.
The zip-to-zone conversion subroutine is used
in conjunction with a postage calculation routine shcwn
in flow chart form in ~IGURE 8. The routine automatically
calculates the postage required to mail a package at sur-
face parcel post rates~ The routine requires that a user
enter either the destination zip code or the destination
æone and utilizes weight-representing signals generated
by scale 54.
~he first operation 315 in the routine is to
initialize the registers used for working storage. The
scale input is read in operation 316. In one embodiment
of the invention, the scale input is read in one pound
increments. Package weights falling between pound
increments are read as the next higher pound lncrement.
For example, the scale reading for a 5.4 lb. package would
be 6 lbs.
A series of error checks are made following
the reading of the scale input. A weight check 317 determines
.,
~ ~30-
:,
., ,; .
.~-
~ "., .. ~ ~ .
,
, . . . . .
:'. . ' . - '- ' ~, ' ''' ' ' , ~
.
~ - \ .
:
whether the scale reads 0, indicating an error s~lch as a
disconn~cted or malfunctioning scale. Subsequcntly, a
weight check 318 is made to determine whether the scale
reading exceeds 70 lbs. A scale reading in excess o~
70 lbs. may indicate a scale malfunction where the scale
upper measuring limit is 70 lbs. In any event, a scale
reading in excess of 70 lbs. indicates that the package
cannot be mailed at parcel post rates under current
postal service regulations. Finally, a chec~ 319 is
made as to whether the zone value is less than or equal
to 8. Since there are only B ~ones in use according to
current postal regulations, a zone reading of 9 or above
must be an error.
If the weight signal and the zone signal are
~alid, the "cents" and "tens of cents" digits of a base
postage for the selected zone are retrieved from storage
in operation 320. A further check 321 is made as to
whether the zone signal is less than or equal to 3. If
; the selected zone is in the 4-8 range, the base postage
is eyual to $1 plus the retrieved digits. If the selected~
zone is in the Local-3 range, the base postage is equal
to the retrieved digits only. This arrangement conserves
data storage space in a preferred embodiment of the
; invention since the B bit registers which are used can
store 2 digits in binary-coded decimal format. The use ~--
of extra registers for storing the dollar digits is avoided
by the described arrangement.
The surface parcel post rates for each of the
-31-
' .
,
~3~
eight different zones increases from the base postage
for the zone either by uniform incremental amounts or
by different alternating incremental amounts~ For
example, in zone 1-2, seven cents is added to the
postaqe for each additional pound in weight above 2 lbs.
while in ~one 3, the postage increases either by ~ cents
or by 9 cents for each additional pound over 2 lbs.
Table IV below shows the pattern of increments
for each of the eight different zones, zone 1-2, being
considered a single zone. The patterns continue through-
out the entire weight range. The Table defines the
current base postage for each of the zones, an increment
A for each of the zones and an alternating increment B
for each of the zones.
'
: .
~ -32-
_ -
r ' .
.. : . -
: ~ . : ' ' , . ~ . ''
: . :
O ~ +: + n ~
~ I ~ ~O ~ ~I N ~1
_ r~~ _ _.. __ _ __
' r o LL L~L~ ~ I~ `
~o ~ ,1 r ,~ r ~ ,1 ,1 - .
_ _ _ . . ._ _
~ L L~ ~ ~ ~ - ~
: ~ o ~ ~ ~, o~ ~ ~ o~
~ . _ _ _ _
o
~ : j ~,
~ + + t + ,~ + ~
~ , :
. . .
, ~ '
'
,; ` ~ .
,' ~ ,
''' '
:i-
, ~ .. .. . .
~' ' ~ ~, ' : :
3~
When the base postage, increment A and increment
~ for the selected zone have been retrieved, a weight
which is other than an exact pound is a~justed downwardly
to the next lower unit pound in operation 322. The postal
service regulations provide that a minimum or base surface
pa~cel post rate applies to any parcel weighincJ 1-2 lbsA
~s will b~ made clearer later, adjusting the weight-
representing signal downwardly to the next lower unit ~ -
pound assures that the proper number of incrementing
operations are performed during a calculation. After
the weight is adjusted, a weight check 323 is made to
determine whether the weight-representing signal equals 0.
An affirmative answer at this point indicates the package
- weighs exactly 1 lb. The retrieved base postage would be
1 15 loaded into the display and control of the meter would be
returnèd to a main program.
If the weight check 323 indicates the weight is
not equal to 0, the weight-representing signal is decremented
by a pound and another weight check 324 performed. If the
second weight check shows the once-decremented signal is
equal to 0, indicating the package weighed between 1 and 2 lbs.,
the base postage amount is loaded into the display and
; control returned to the main program. If the seccnd
weight check indicates the decremented weight signal is
not equal to 0, the postage is incremented in operation 325
by the amount of increment A for the particular zone and
the weight-representing signal is again decremented. A
third weiqht check 326 is then performed. An affirmative `
'
, .
.,' : .
,
.
3~
answer to the weight check 326 causes the incremented postal
amount to be loaded into the display. A negative answer calls
for the postage amount to be incremented by the amount of
increment ~ for the zone. When increment has been added to
S the postage, the weight-representing signal is again decre-
mented and a fourth weight check 327 is performed. If the
fourth weight check indicates the decremented weight is not
equal to 0, the routine loops back to incrementing operation
325.
The increment postage/decrement weight process
is reiterated until one of the weight checks 326 or 327
inclicates the weight-representing signal has been decremented
to 0. When this occurs, the incremented postage is loaded
into the display and control returned to a main program.
An example of the calculations performed is set out in
Table V wherein it is assumed that a package weighing
5.5 lbs. is to be mailed to zone 6 at surface parcel post
rat~s.
~,
~3~
~ ~ r~
~ o ; ,, ,, ,i ,i
u~ r ~ ~ ~ o
. 3 u~
U~
o o o ~ o o o o o
æ z ~ z; æ æ z æ
.
. ~
.. ~1
u~ _, ,~ " ,~ . a
r~
'~ s ~s ~s ,~s
z
hl ~`. O ~ `. 3 ~`- ~ 3 ~ ' ~ t`. cr 3 ~ 3
X Q) O ~ O O 0 O 0 O 0 O 111 0 ~
: ~ t~ ~ ~a O C 11 0 C 11 0 ~ 11 0 ~ 11 e
æ u, ~ ~ v l a~ v ~ u
.~ o
~ ~ rl U~rl U U r~ U U r~ U U ~ U ~
c~ ~ o Q~ O ~ c a~ ~ o
1. ~ K 3 3 ~ ~S C:l 3 H Q 3 H C~ 3 H a 3 H a a O ~
. ., ~ '~
~ :
.~ .
., ~ ~ Gl O ~1 ~3 1'') ~ In v~
O ~ ~1 ~1 ~ ~ ~ ~ ~ ~`J ~ N
.. ~``
- ~6-
.
. ~ ,
: ' .
,.
:
;
~3~
It would be possible to perform the postage
calcl~lations using a postage lookup table storing the
proper or postage for each package weight in the 1-70
lb. welqht range For each of the ~ones. The arrangement
just described clearly requires much less memory capa-
bility.
The same decrement weight/increment postage
concept is used to calculate postage for parcels being
mailed at library rates. FIG~RE 9 is the flow chart
for the calculation of library rate postage. The first
four steps for the library rate routine are identical to
the corresponding steps of the parcel post rate routine.
That is, registers are initialized, the scale input is
read and weight checks are performed to determine whether the
lS scale reading is either equal to 0 or greater than 70.
If the scale input is greater than 0 and less than or
equal to 70, the base postage and postage increments for
the library rate class of handling are retrieved in operation
328. Currently, the library rates are 8 cents for the first
pound of a parcel plus 4 cents Eor each additional pound.
Thus, the base postage would be 8 cents and the increment
4 cents. The weight-representing signal is decremented
by 1 lb. in operation 329 and a weight check 330 performed
to determine whether the weight equals 0. If the decremented
weight-representing signal equals 0, the postage is loaded
into the display and control of the meter return to the
main program. If ~he signal is not equal to zero, the
postage is incremented and the routine is re-entered at
-37-
,
,
.: , . .
., ... ~ '. '
,
~ `
~ 33~
operation 329. The process is iteratively performed until
the weiyht check 330 reveals the weight-representincJ signal
ha~ been decremented to 0.
The routine Eor calculating book r~te postage
amounts uses the same decrement weight/increment postage
conc~pt as the routines just descri~ed. The details of the
book rate calculation routine differ because of the
difference in current postal service rate structures.
Postal service regulations establish current
book rate postage at 21 cents for the first pound of a
parcel plus 9 cents for each additional pound up to 7 lbs.
The postage increases by 8 cents for each pound above 7 lbs.
Referring to FIGU~E lO for a flow cha.~t of the
routine far calculating book rate postage, the first four
steps of the routine are the same as the corresponding
: steps for the parcel post calculation routine and the
library rate calculation routine. ~hat is, registers are
initialized, the scale input is read, and weight checks
are performed to determine whether the weight~representing
slgnal either equals 0 or is greater than 70. Assuming
the weight checks are satisf~ed, a further weight check 331
is made to determine whether the weight of the parcel is
`~ more than 7 lbs. If the parcel weighs 7 lbs. or less, the ~:
base amount of 21 cents and an .incremental a~nount of 9 cents
per lb. ~at current rates) is retrieved. The weight-repre-
senting signal is decremented by l lb. in operation 332
and another weight check 333 performed to determine whether
the decremented signal equal 0. If not, the base amo~n~ is
38-
'
' ,
- ;
~3~
incremented and the routine re-entered at operation 332.
The process iterations continue until the weight che~k 333
indic~tes the weight-representin~ signal does equal 0.
If the weight check 331 indicates the package
weighs more th~n 7 lbs., the scale input signal is re-
defined in operation 334 by decrementing it by 7 lbs.
a psuedobase and rate is established in operation 335. The
psuedo base amount is actually the amount of postage required
for a 7 lh. package while the incremental rate is the 8 cent
increment required for parcels weighing more than 7 lbs.
Thus, according to this subroutine, a 10 lb. parcel would
be redefined as a 3 lb. parcel having a base postage of 75
cents. The decrement weight/increment postage iterative
routine would decrement the 3 lb. signal while incrementing
the 75 cent base postage by 8 cents for each of the remaining
3 lbs.
When the proper value has been calculated and any
special charges added by the user through numerical key-
board 56t depressing the set button and the unlock but~on,
if necessary, causes the system to initiate setting of the
meter. In very general terms, the master gear 51 is shifted
a print wheel bank at a time by selec*ive energization of
solenoids 126 and 128. The rotation of master gear Sl in
each of the print wheel bank positions is controlled as a
function of the set postage. Motor control signals are
p-ovided through ou~put port 435.
While a postal calculator embodying the present
invention has been described in the context of a special
-39-
... . . .
3~
puxpose computer system, such a calculator can be implemented
in th~ ~orm of discrete or hard-wired logic circuits.
FIGURE 11 is a block diagram of such an implementation and
includes a keyboard 500 for entering postage directly, for
S ~ntering destination ~ip code prefixes or destination zones
,~ and for selecting the mailing class to be utilized. A
class selection lockout circuit 502 accepts the keyboard
input and provides an energizing output to one of three
circuits; a surface rate data circuit 504, a library rate
data circuit 506 and a book rate data circuit 508. Class
selection lockout circuit 502 simultaneously inhibits inputs
representing the other two classes of handling. A
destination zip code prefix entered through keyboard 500
is applied directly to surface rate data circuit 504 which
determines the proper zone value and accesses data
storage elements containing the minimum postage and the
postage increments for that zone. The minimum postage and
postaqe increments for the selected class of handling and
zone ~for surface rate class) are inputted to a postage
calculation circuit 510 which also accepts a weight-
representing signal generated in the scale mechanism 512,
provided the input weight falls within the proper limits
as determined by an interposed weiqht limit check
circuit 154. If the weight is outside the appropriate
limits, an error signal is generated.
The class seiection lockout circuit is shown
in PIGURE 12. The lockout circuit includes inputs fxom
the surface rate select key 516, a library rate select
.
.
.
3~
key 518, and a book rate select key 520, each o~ which
~enerates a positive going pulse when depressed. The
o~tputs of thc keys 516, 518 and 520 are applied to trigger
inputs of con~entional J-K flip flops 522, 524, S26, res-
pectively. ~he J input terminal for each flip flop is
provided by a dual-input AND gate having its inputs
connected to the Q output of the other two flip flops.
The K input terminal for each flip flop is also provided
from the dual-input AND gate having its inputs provided by
the Q output terminals of the other two flip flops. As an
example, AND gate 528 connected to the J input terminal of
flip flop 522 has its first input from the Q output terminal
of flip flop 524 and its second input from the Q output
; terminal of flip flop 526. An AND gate 530 connected to
the K input terminal of flip flop 522 has one input from
the Q output terminal of flip flop 524 and another input
from the Q output terminal of flip flop 5~6.
When the system is initialized prior to use,
each o~ the flip flops 522, 524, 526 i9 driven to a rese~
state by a clear pulse applied to a clear input terminal
~; (not shown). In the reset state, the Q output for each
flip flop is at a binary 1 level while the Q output is at
a binary zero level. In this initializea condition, the
J input terminal for each flip flop carries a binasy 1
signal and the K input terminal carries a binary 0 signal.
Whcn one o~ the keys 516, 518. 520 is depressed, the flip
flop connected to that key is driven to a set state wherein
its Q output rises to a binary 1 while its Q output falls
to a binary zero.
-41-
'; ' ~ '. ' ' ~ ' . '' '-
', ~
' . ' , ' -:
. ' ' '
\
3~
For purposes of illustration, assume key 516 has
been depressed to select the surface rates class. At the
trailir.g edge o~ a pulse gene~lted by depression of the
k~y, Clip flop 522 is driven to the set state. A binary
zero signal on the Q output terminal is applied to the
AND gdtes at the J input terminals of the other two flip
flops 524 and 526. The AND gates providc binary zero
signals to the J and K input terminals of the flip Elops
524 and 526 to inhibit any change in state of those flip
flops if either the library rate select key 518 or the book
rate select key 520 is subsequently depressed. A binary 1
signal on the Q output terminal of flip flop 522 initializes
and energizes a zip to zone conversion circuit shown in
block diagram form in FIGURE 13.
The zip to ~OnR conversion circuit includes the
Xeyboard 500. In one embodiment of the invention, a five
bit word representing a 0-3 numher is entered in parallel
into the system. Four of the bits identify the numeral.
The fifth bit, a fixed binary one, is used for signal
shifting and error checking purposes.
When any numerical key on keyboard 500 is
depressed, a five bit word is a~pplied in parallel to an
OR gate 532 and to a delay circuit 534. The OR gate 532
always responds to an entry regardless of the numerical
value since each word contains at least one binary 1 in
the fifth or control position. OR gate 532 transmits a
pulse to a pulse generator 536. Pulse genera~or circuit
-42-
'
.
, ~ ; .
: ' ' :~' . , ,' : ' ~ ,
~,. . . . .
.
. .
f,~ ,
536 provides a shaped pulse suitable for controlling the
shifting ofparallel data through serially connected shift
registers 538, 540, 542, 544. A delay circuit 534 trans-
fers each S bit word into the first of these registers
53B shortly after the shift pulse occurs. Upon entry
of subsequent words, the pulse generated by pulse
generator 536 causes the words to be propagated through
successi~e registers.
A validly-entered zip code prefix includes only
three numerical digits. Therefore, iE register 544 contains
any fourth numerical digit, including a ~ero, at least one
binary 1 signal will be stored in register 544 since the
fifth bit of each word is always a binary 1. A binary 1
output from OR gate 546 is construed as an error signal.
To determine whether at least three digits have
been entered into the system, the control bit position in
each of the shift registers 538, 540, 542 is connected to
one input of a quad input AND gate 548. The fourth input to
AND gate 548 is provided by a zip-~one conversion key
on keyboard 500. If the necessary three digits have been
entered into the registers when the zip-zone conversion
key is depressed, all inputs to AND gate 548 will be at
binary 1 levels, producing a binary 1 signal on the output ~ ~-
of that AND gate. The output of AND gate 548 is the control
input for a high impedance logic buffer circuit 550 such a
a DM 7094/DM8094 T~I-STATE quad buffer available from National
Semiconductor Corporation. With a binary 1 on its coptrol
lnput, buffer circuit 550 transmits the contents of the
-43-
.
. .
~3~
registers 538, 540, 542 to an unassigned ~ip code detector
circuit 552 and to the inputs oE a second logic buffer
circuit 558. Since the fifth bit of each input word is
no longer needed ~nd since the buffer circuits 550 and 558
~rQ made up of parallel-connected quad input devices the
necessary number of devices/buffer circuit is held at
three by only using the four numeral-identifying inputs of
each word. Unassigned zip code detector 552 may comprise
a BCD to decimal converter for applying decimal represen-
tations of the digits of the zip code prefix to a numberof AND gates each having normal and inverted inputs which
allow the AND gate output to rise to a binary 1 level when
a signal pattern for an unassigned zip code is detected at
its inputs. If the output of any ~ND gate in the unassigned
zip code detector 552 goes to a binary 1 level, indicating
the entered zip code is not in use, an error signal is
generated.
; Assuming the zip code prefix is a valid one,
the output of the unassigned zip code detector circuit
552 remains at a binary 0 level. This binary zero signal
is inverted by an inverter 556 to provide a signal-passing
binary 1 control input to the second logic buffer circuit
~ 558. Logic bufer 558 transmits the information to an
-~ OR gate 560 and to an input register 554. The OR gate
25 560 initiates operation of a counter/decoder circuit 562
which provides sequential decimal-coded outputs to AZIP
registers 564 containing ~ZIP values isola~ed into groups,
defined earlier. The ~ZIP values are transferred
44-
.~ , .
,~, .
,: - , . . . - . : - .
.,
~, .
.
3~J~
sequentially into an arithmetic unit 566 which has the
second input from the input register 554. The function
of ~rithmetic unit 566 is to decrement the contents of
register 554 by the ~ZIP value currently applied to the
arithmetic unit 566. This decrelnented signal is applied
to an output re~ister 568 from which it is fed back to
the input reqister 554 for use in the next iteration.
The output register 568 also provides an input to a result
check circuit 570 which operates to determine when the
contents of register 568 have become less than or equal
to zero.
If the result check shows ~hat the decremented
signal in register 568 has become less than or equal to
æero, a binary 1 signal is applied to one input of each
of a number of dual input AND gates in an array 572. A
second input to each AND gate is energiz`ed by the output
of counter/decoder 562 only at a particular decimal
count. When an AND gate in the array 572 is energized,
that AND gate causes the contents of an associated zone
value register to be shifted out for further use in the
postal calculations. The same count from counter/decoder
circuit 562 which transfers the final ~ZIP value into the
arithmetic unit 566 is used to establish access to the
~one register. Therefore, the last ~ZIP value retrieved
and the zone value might be considered to have the same
storage address notwithstanding they are stored in
separate registers.
The zip to zone conversion process described
-45-
: ' '
''-.
..
: ~ ,,- ' ~ ,, : ;
- .
:, ",
.
above involves decremcnti~gof a zip ocde related signal.
It is possible, as an alternative mode of computation,
to provide the same ~ip to zone conversion through the
use of an incrementinq process wherein iteratively
S accumulated ~ZIP va.lues are checked against the originally
entered zip code prefix. Referring to FIGURE 14, the
destination zlp code prefix transmitted by logic buffer
circuit 558 is stored in a register 574 providing one input
to a digital comparator circuit 576. ~ZIP values retrieved
from the ~ZIP registers 564 are totalled in an accumulator
circuit 578 providing a second input to the digital comparator
circuit 576. The digital comparator circuit 576 continually
compares the contents of prefix register 574 and accumulator
.~ .
circuit 57B until it is determined that the contents of
circuit 578 are equal to or greater than the contents of
;: register 574. Comparator circuit 576 provides the control
input to AND gate array 572, which accesses the contents
of the zone registers 573.
~eferring to FIGURE 15, the postage calculator
circuitry has inputs rom a scale mechanism 582, from
the zip to zone conversion circuit as represented by
surface rates select block 584, from a library rates
select block 586 and from a book rates select block 588.
The outpu-ts of blocks 586 and 588 are the outputs of flip
flops 524 and 526, respectively, in the class selection
- loakout circuit 502 of FIGURE 11. The output of scale
: mechanism 582 is applied to weight checking logic gates
590, 592, 594, 596 and 598. To simplify the drawings,
~- -46-
- ' , ,
~' ~ . ' ' ~ . .
, - : . .
,'` ' ' . :,
-
. .~ -.
~lq33~
the decimal value for the binary inputs to each of these
gates is shown. NOR gate 590 produces a binary 1 signal
ifan~ only if the weight of the package is less than 1 lb.
AND qate 594 produces a binary 1 output if and only if
the weight of the package is greater than 70 lbs;
specifically, 72, 80 or 96 lbs. AND gate 598 produces
a binary 1 output signal if and only if tne weight of
the packa~e is 70+ lbs. OR gate 600 provides an error
signal if either of the AND gates 594, 598 is enabled.
This error signal indicates that the package weight is
greater than 70 lbs.
Assuming the weight checks indicate the package
weight is within the range of 1 to 70 lbs., the 6 bit
binary-coded weight-representing signal is employed in
addressing programmable read only memories 614, 616 and
618. Programmable read only memory 614 is also addressed
by the 3 bit binary coded zone signal. The binary signals
representing a particular weight and a particular zone are
used as; address signals to select a unique memory location
within programmable read only memory 614 for each zone-
weight combination. The postage amount corresponding
to the zone-weight combination is stored in that location.
Using 16 bit words allows postage values ranging from
$00.00 to S99.99 to be stored in each location in binary- ,~
coded decimal format. The postage value stored in a location
addressed by the weight-zone signal is retrieved, provided
the surface rate select block 584 has enabled programmable
read only memory 614. A retrieved value is outputed to an
output register 620.
-47-
' ~ ' ' ' ' ,' .
~3~
The same 6 bit weight-represcnting signal is
applied to programmable read only memories 616 and 618
which are enabled when library rates and book rates,
respectively, have becn selected. In each of these
memories, the weiyht siynal is used to retrieve a
postaye amount from a lacation uniquely related to the
weiyht.
While there have been described what are
considered to be preferred embodiments of the invention,
variations and modifications therein will occur to those
skilled in the art once they become familiar with the
invention. Therefore, it is intended that the appended
claims shall be construed to include all such variations
and modifications as fall within the true spiri-t and
lS scope of the invention.
48-
... .
`
':
