CA1136167A - Method and apparatus for tracking creep and drift in a digital scale under full load - Google Patents
Method and apparatus for tracking creep and drift in a digital scale under full loadInfo
- Publication number
- CA1136167A CA1136167A CA000360669A CA360669A CA1136167A CA 1136167 A CA1136167 A CA 1136167A CA 000360669 A CA000360669 A CA 000360669A CA 360669 A CA360669 A CA 360669A CA 1136167 A CA1136167 A CA 1136167A
- Authority
- CA
- Canada
- Prior art keywords
- weight
- scale
- value
- load
- pan
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01G—WEIGHING
- G01G23/00—Auxiliary devices for weighing apparatus
- G01G23/14—Devices for determining tare weight or for cancelling out the tare by zeroising, e.g. mechanically operated
- G01G23/16—Devices for determining tare weight or for cancelling out the tare by zeroising, e.g. mechanically operated electrically or magnetically operated
- G01G23/163—Devices for determining tare weight or for cancelling out the tare by zeroising, e.g. mechanically operated electrically or magnetically operated involving digital counting
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01G—WEIGHING
- G01G23/00—Auxiliary devices for weighing apparatus
- G01G23/18—Indicating devices, e.g. for remote indication; Recording devices; Scales, e.g. graduated
- G01G23/36—Indicating the weight by electrical means, e.g. using photoelectric cells
- G01G23/37—Indicating the weight by electrical means, e.g. using photoelectric cells involving digital counting
- G01G23/3707—Indicating the weight by electrical means, e.g. using photoelectric cells involving digital counting using a microprocessor
Abstract
METHOD AND APPARATUS FOR TRACKING CREEP AND
DRIFT IN A DIGITAL SCALE UNDER FULL LOAD
Abstract of Disclosure Method for tracking creep and drift of a digital scale after a load has been placed upon a scale pan. The tare weight is determined at a time before a load is placed upon the pan and is stored. After the pan comes to equilibrium, the gross weight of the scale is determined and stored. Variance in the gross weight as a result of creep and drift is tracked and after an incremental amount is determined the value of such variance is added to the stored tare weight so that upon removal of the load from the pan the true tare weight is maintained.
DRIFT IN A DIGITAL SCALE UNDER FULL LOAD
Abstract of Disclosure Method for tracking creep and drift of a digital scale after a load has been placed upon a scale pan. The tare weight is determined at a time before a load is placed upon the pan and is stored. After the pan comes to equilibrium, the gross weight of the scale is determined and stored. Variance in the gross weight as a result of creep and drift is tracked and after an incremental amount is determined the value of such variance is added to the stored tare weight so that upon removal of the load from the pan the true tare weight is maintained.
Description
113~i~7 METHOD AND APPARATUS FOR T~ACKING C~EEP AND
DRIFT IN A DIGI.AL SCALE UNDER FULL LOAD
~ackground of Invention Digital scales of current design utilize automatic zero tracking techniques which eliminate the need for frequent operator intervention to readjust the tare weight. Such automatic zero tracking usually involves measuring sequential signals representa-tive of the tare weight and continually replacing older signals with newer signals until such time as a load is placed on the pan. When the load is placed on the pan, the latest siqnal or signals would then be stored to be subsequently subtracted from the gross weight.
Prior methods of tracking have been with an empty pan only; consequently when a weight is left on a pan for any sig-nificant duration, deviation from zero can taXe place either as a result of electronic or load cell drifts. With typical load cell scales, significant change in the reading takes place after a heavy load is applied due to the creep of the load cell; however, the change in indicated weight immediately after the load is applied is unaffected. Thus, when a load is left on a scale and creep takes place, immediately after the load has been removed the zero reading has been displaced by the same change as the creep. In time, the scale would uncreep in a manner symmetrical with the original creep ~ depenaing on the original creep's rate and size, but a significant amount of time may elapse before zero is recaptured.
It therefore would be advantageous to have a system whereby the tare weight, or zero, is continually updated whether a load , v r ' _ 113~7 is upon the pan or not. With such a system, not only would time be saved in waiting until the tare weight has reached equilibrium after a load is removed from the pan, but also at ~he time the load is upon the pan the gross weight would not be accurate if creep and drift are not taken into account.
Summary of the Invention A method and apparatus has been devised whereby the tare weight of a scale may be continually updated whether or not a load is received upon the. pan of the scale. Before a load to be weighed is placed on the pan, the tare weight, or zero, is continuously updated by storing newer values and eliminating older values of the tare weight. This will continue until such time as a load is placed upon the pan, at which time the latest tare weight will be stored. The scale is allowed to reach equilibrium at which time the gross weight of the load will be measured and stored. At this time, the tare weight will be subtracted from the gross weight automatically and the net weigh~
shown on some type of display. When the load is left upon the pan for an extended period, the creep and drift are continually -monitored so that the stored tare weight can be adjusted in response thereto. The storing of the adjusted tare weight when a load is on the pan will result in a constant net weigh~ being obtained.
Brief Des. iption of the Drawing FIG. 1 shows a block d agram of the components utilized in the method of this invention.
_ .. . . , . _ . . _; _ . , .
.
- ~3~167 FIG. 2 is a flow chart of the logic involved ln one entodiment of the instant invention.
FIG. 3 is a flow chart of the logic involved in another en~xodiment of the instant invention.
FIG. 4 is a graph of the measurement by a scale of gross weight and net weight as a function of time.
Detailed Description of the Preferred Embodiment Referring to FIG. 1, a microcomputer-driven computing scale system is shown. A weighing scale 10 capable of generating analog signals, such as Scale Model No. 5035 produced by Pitney-Bowes, Inc., is electrically connected by means of an output line 11 to an analog to digital (A/D) converter 12. Included in the A/D converter 12 is control logic, such as TTL count~rs and gates. The A/D converter and control logic 12 is electrically connected to a microoamputer 14, such as Model No. PPS 4/1, produced by Rcckwell International Ccmpany, by means of suitable input and output lines, shown generally at reference numeral 16. These input and output lines 16 comprise a data line and a control line.
Ihe microcamput~r 14 is, in turn, oonnected to a clcck 18, a power supply 20 and a PC~E3R ON reset 22, by means of suitable input lines, shown generally at reference numeral 24. ~he microoomputer 14 is connected to a display 26, and drives the display 26 over a oorresponding output line 28.
In operation, a weight or load is placed on the pan of the scale 10, which scale generates analog signals over the cutput line 11.
m e A/D oonverter 12 receives the analog signal and converts it to a digital number. Ihe digital number is transmitted over output lines 16 to the microcomputer 14. m e microcomputer 14 is powered by the power supply 20 and its associated Pr~D~R ON reset 22, by means of suitable lines 24. The clock 18, - csm/ ~
113~i167 generates timing signals for the microcomputer 14 over a suitable line 24. The microcomputer 14 is used to calculate the weight of a load on the scale 10 represented by the signal generated therefrom. The weight is then displayed on the display 26 as a-result of the signal generated from the microcomputer 14 over the output line 28.
Initial Operation Referring now also to FIG. 2, the microcomputer 14 performs the steps hereinafter described. Before a weighing operation is begun, initialization 30 takes place. During initialization, all input/output (I/O) ports are cleared, all flags are cleared, and all variables are set to default values or to zero. The digital number generated by the A/D converter 12 is sent over the data output line 16 to the microcomputer 14. The number is read 32 and stored 34 in a READING buffer in the microcomputer 14.
Digital values continue to be read into the READING buffer until the buffer is full 36. A comparison of these values is made to determine whether they vary significantly with respect to one another. This determination is shown at reference numeral 38.
The significance of this determination is that when consecutive values do not vary significantly from one another over a given time interval, the pan of the scale 10 can be presumed to be at rest, or not` in motion. Such is generally the case when the system first becomes operational with no load on the pan, or when the pan has had time to ettle after a load is applied.
In the case of initial operation with no load on the pan of the scale 10, the answer to the question asked at reference .,.,.. , .. ., , . .. ~ , .. . . .. ...
11361~7 `
umeral ~ (IS THE~E ANY MOTION?~, is negative. An average of the va~ues in the READING buffer, hereinafter called AVERAGE, is then computed ~0.
~ If no value has been loaded into a TARE register 42, as is the case in initial operation, the value calculated as AVERAGE is then loaded into the TARE register 44. A SAV~D
WEIG~T flag is set 46 and AVER~GE is also loaded into a SAVLD
WEIGHT register 48. A calculation is ~en performed, subtracting the value stored in the SAVFD WEIGHT register from AVERAGE. The resulting difference is then stored in a register called the R register 50. In the case of initial operation, since AVERAGE
had been previously loaded in the SAVED WEIG~T register, the value zero is loaded into the R register.
The system then determines whether the SAVED WEIGHT flag has been set 52. In the case of initial operation, the flag has been set. The system then determines whether the absolute value of the value loaded in the R register is within an arbitrary band called the CREEP THRESHOLD 54. In this embodiment, the CREEP T~RESHOLD is set at 1/40th of an ounce. In the case of initial operation, since tlle value stored in the R register is zero, the answer to tbe question asked at reference numeral 54 (IS ~R¦ LESS THAN THE CREEP l1HRISHOLD?), i~ affirmative. ~lhat is, zero is less than 1/40th of an ounce. ~rhe NET ~EIG~T is then computed by subtractiny the value in the TARE register from AVER~GE 56.
The system then determines whether the value calculated as ~ET WEIG~T is greater than an arbitrary band called the ZERO
T~IRESHOLD 58. In this embodiment, the ZE~O THRESHOLD is set at 1/15th of an ounce. In the case of initial operation, since AVERAGE was loaded into the TARE register, the computed NET
..
.... . . . - - .. ~ ; -113~167 ~IGHT is zero. Therefore, the answer to the question ~sked ~t reference numeral 5~ ( IS NET WEIGIIT GREAT~R'T~AN THE ZERO
TIIRESHOLD?), is ne~ative. The value in the TARE register is then updated 60, and the value in the TARE register is then loaded into the SAVED WEIGHT register 62. The NET WEIG~T is then converted to pounds and ounces 64. Since no load is on the pan, this would be zero.
Pre-Weighing Operation At this point, the system continues to process signals generated by the scale 10 and converted into digital numbers by the A/D converter 12. The microcomputer 14 reads the number generated by the A/D converter 12, shown at reference numeral 32.
This reading is stored 34 and the oldest reading is discarded from the READING buffer 36.
It is conceivable that, after initial operation but before a weight is placed on the scale 10, slightly different readings will be received by the microcomputer 14 from the A/D converte_ 12 as a result of drift or noise. The system will then determine that there is motion of the pan 38, and will then clear the SAVED
~JEIGHT flag 66 before continulng to read a value from the A/D
converter 12 shown at reference numeral 32. A cleared SAVED
WEIGHT flag thus indicates that the pan of the scale 10 is in motion. When the scale 10 damps out, and no motion is de-tected, as determined by the comparison hereinbefore described at reference numeral 38, he system then computes the AV~RAGE of the values in the READING buffer 40. At this point, the TARE
register had been previously loaded during initial operation.
Consequently, the system does not execute the step~ represented by reference numerals 44, 46 and 48, but rather con~putes the new _ . . . . . ... . . . . . . . .... ... . .
113~
AV~hAGE mir.us the value in the SAVED WEIGIIT-register. Tbe result of this computation is then stored in the a register 50.
A~ this point, the condition of the SAVED WEIGHT flag is again determined 52. Since this SAVED WEIGHT flag had previously been cleared, the system then determines whether the value in the R register is less than an arbitrary band called the CHANGE OF
WEIGHT T~ S~IOLD 6~. In tllis embodiment, the CHANGE OF WEIGHT
THRESIIOLD is set at 1/lSth of an ounce. If the answer to this question (IS ~R¦ LESS TffAN THE CHANGE OF WEI~HT THRESHOLD?), is negative, AVERAGE is then stored in the SAVED WEIGHT register 70.
If, however, the value ~n the R register is less than the C~ANGE
OF WEIGHT THRESHOLD, AVERAGE is not stored in the SAVED WEIGHT
register. ~t this point, regardless of whether the new AVERAGE
is stored in the SAVED WEIGHT register, the SAVED WEIGHT flag is set 72 and the ~T WEIGHT is computed by subtracting the value in the TA~E register from AVERAGE 56. Since no weight is on the pan of the scale 10, it is probable that the NET WEIGHT is not greater than the ZERO T~RESHOLD band 58. Consequently, the TARE
regis~er must be updated 60. The value in the TAXE register is stored also in the SAVED WEIGHT register 62. l'he NET W~IGHT is again converted to pounds and ounces 64.
Weighing Operation For purposes of illustration, it is now desirable to consider the operation of the system when a force .s applied ~o the pan of the scale 1C After a weight is placed on the pan, a digital number from the A/D converter 12 ia entered and stored 34 into the READI~G buffer 36. Initially, of course, the pan will be in motion 38, so the SAVED W~IG~T flag is cleared 66 and .,, ,,, . ~ . . .. . . . . . . . .. ... .
. , , ~, . ,, ~ , , .
- ` 113~67 reading of the digital numbers generated by tbe A~D converter ~2 continues to be made as shown at reference numeral 32. When the system determines that there is no significant difference among the values read, the answer to the question asked at reference numeral 38 (IS Ti~ERE ANY i~OTION?), is answered in the negative.
A new AVERAGi is computed.
A value had previously been loaded in the TARE register, so the answer to the question asked at reference numeral 42 (IS
TARE REGISTE~ LOADED?), is affirmative. rrhe value in the SAVED
WEIGHT register is now subtracted from the new AViRAGE and this result is stored in the R register 50. Because the SAVED WEIGHT
flag had previously been cleared, the answer to the question asked at reference numeral 52 (IS SAVED Wk~IGHT PLAG SET?), is negative. The system then determines whether the absolute value of the value stored in the R register is less than the CHANGE OF WEIGHT TilRESHOLD 68. If the answer to this question (IS ~R! LESS THAN THE CHANGE OE~ WEIGHT TIJRE~IIOLD?J, is negative, AVERA~E is then stored in the S~VED W~IGI~T
register 7~. If, however, the value in the R register is less than the CHANGE OF WEIGHT THRESHOLD, AVERAGE is not stored in the SAVEi~ W~IGHT reyister. At this point, regard-less of whether the new AVERAGE is stored in the SAVED
WEIGHT register, the SAVED WEIGHT flag is set 72 and the NET
WEIGHT is computed by subtrzcting the value in the TARE
register from AVERAGE 56 At this weighing, the N~T WEIGI~T is most probably greater than the ZERO THRESHOLD band 58, so the system immediately converts the NET WEIGHT into pounds and ounces 64.
--a-- !-;
~, .. . .;.... . ... , . _ _, . . .. . . . _ . ..
1136i67 Creep l`rackinq Uperation Now consider the situation that arises after a weight has been determined ~o be on the pan of the scale 10 for a significant length of time. In this case, a digital number is read from the A/D converter 12, shown at reference numeral 32. The value is stored 34 in the READING buffer 36. It can be assumed that there is no significant motion of the pan on the scale 10, as shown at reference numeral 38. An AVERAGE is computed 40.
A value had been previously loaded into the TARE register, so the answer to the question asked at reference numeral 42 (IS
l'ARE REGISTER LOADED?), is affirmative. The system then subtracts the value in the SAVED WEIGHT registe, from AVE~AGE and stores this result in the R register 50. Note that since there was no motion of the pan, the SAV~D WEIG~T flag is set. Consequently, the answer to the question asked at reference numeral 52 (IS
SAVED WEIGHT FLAG SET?~, is affirmative. The system then determines whether the absolute value of the value in the R
register is less than the cxrEp THRES~OLV 54. For purposes of illustration, it is assumed that the absolute value of the value in the R register is greater than the CREEP TliRES~VLD. The value in the TARE register is then updated with a value consisting of the old TARE register value plus the value in the R register 74. The AVERAGE is then stored in the SAVE~ WEIGHT register 70 and the SAVED WEIGHT flag is set 72 (although this is a redundant step). The NET WEI~iT is computed by subtracting the new value in the TARE reoister from AVERAGE 56.
Since there is a significant weight on the scale 10, it is assumed that the new computed NET WEIGHT is greater than the ~ERO THR~'S~i~L~ band ~8, sv the NET h~IGtiT is immediately converted to pounds and ounces 64. r _9_ 113fj~i7 Alternate Embodiment -- Initial Operation - Xeferring now to FIG. 3 in conjunction with FIG. 1, an .alternate embodiment is hereinbelow described. The microcomputer 14 performs steps as follows. ~efore a weighing operation is begun, initialization 80 takes place. During initialization all input/output (I/O) ports are cleared, all flags are cleared, and all variables are set to default values or to ze~o. The digltal number generated by the A/D converter 12 is sent over the data output line 16 to the microcomputer 14. The number is read 82 and stored 84 in a READING buffer in the microcomputer 14.
Digital numbers continue to be read into the READING buffer until the buffer is full 86. A comparison of these values is made to determine whether they vary significantly with respect to one another. This determination is shown at reference numeral 8B.
The significance of this determination is that when consecutive values do not vary significantly from one another over a given time interval, the pan of the scale 10 can be presumed to be at rest or not in motion. Such is generally the case when the system first becomes operational with no load on the pan, or when a load on the pan has time to settle (i.e., scale dampening is completed).
In the case of initial operation with no load on the pan of the scale 10, the answer to the question asked at referenced numeral ~ (IS ~HERE ANY MorrIoN?)~ is negative. An average of the values in the REA~ING buffer, hereinafter called AVERAGE, is then computed 90.
If no value has been loaded into the TAR register 92, as is the case in initial operation, the value calculated as AVERAGE is been loaded into the TARE register 94. The AVRA~E is ~ . , . .. , , . ~ _ , .
1~3~1~7 louded into a SA~ED WEIGHT register 96 and a SAVED WEI~HT fl~g is also set 9~
The system then determines whether the SAYED WEIGH~ flag has been set 100. In the case of initial operation, the flag has been set. The system then performs a calculation, subtracting the value stored in the SAVED WEIG~T register from AVERAGE 102.
The resulting difference is then stored in a register called the R register 102. In the case of initial operation, since AVERAGE
has been previously loaded in the SAV~D WEIGHT register, the value zero is loaded into the R register. The system then determines whether the absolute value of the value loaded in the R register is within the arbitrary band called the CREEP THR~SHOLD
104. In this embodiment, the CREEP THR~SHOLD is set at 1/40th of an ounce. In the case of initial operation, since the value stored in the R register is zero, the answer to the question asked àt reference numeral 104 (IS ~R¦ ~REATER THAN OR EQUAL TO THE
CRE~P THRESHOLD?), is negative. That is, zero is not greater than 1/4~th of an ounce. The AVERAGE is stored in a LAS'r VALI~
AVERAGE register 106. The NET WEIGHT is then computed by sub-tracting the value in the TARE register from AVE~GE 108.
The system then determines whether the value calculated as NET WEIGHT is greater than the arbitrary band called the ZERO
THRESflOLD 110. In this embodiment, the ZfRO THR~SHOLD is set at 1/15th of an ounce. In the case of initial operation, since AVERAGE was loaded into the TARE register, the computed NET
WEIGHT is zero. Therefore, the answer to the question asked at reference numeral 110 ~IS ~ET WEIGHT GX~ATfR TflAN THE ZERO
THRESHOLD?), is negative. The value in the TARE register is then updated if required 112, and the value in the TARE register is loaded into the SA~ED WEIGHT register 114. The NE~r WEI~HT is ~_ . , . . , . _ 113~i~7 .
ehe c~nverted to pounds and ounces 116. Since no load is on the pan, this would be zero.
Pre-Weiqhinq ~peration At this point the system continues to process signals generated by the scale 10 and converted into digital numbers by the A/D converter 12. The microcomputer 14 reads the signals generated by the A/D converter 12, shown at reerence numeral 82.
Once again, these readings are stored 84 in the READI~G buffer 86.
It is conceivable that, after initial operation but before a weight is placed on .the scale 10, slightly different readings will be received by the microcomputer 14 from the A/D converter 12 as a result of drift or noise. The system will then determine ~o~
that there is motion of the pan ~8 and will then clear the SAVED
WEIG~T flag 118 before continuing to read values from the A/D
converter shown a reference numeral 82. A cleared SAVED W~IGHl' flag thus indicates that the pan of the scale 1~ is in motion.
When the scale 10 damps out, and no motion is detected, as determined by the comparison hereinbefore described at reference numeral ~8, the system then computes the AVERAGE of the values in the READING buffer 90. At this point, the TARE register had been previously loaded during initial operation. Consequently, the system does not execute the steps represented by reference numerals 94, 96 and 98, but rather determines whether the SAVED
WElGH~ flag is set 100. Since this SAVPD WEIG~T flag had pre-viously been cleared, the system then suotracts the value stored in the LAST VA~ID AVERA~E register from the AVERAG~, and stores this value in ehe R register 120. The system then determines whether the value in the R register is greater than or equal to \~
... _ _ _ .......
~ . . . . ..
113i167 the CHANGE O~ IYEIGHT THRESIIOLD 122. In this embodiment, the CHANGE OF WEI~T TH~ESHOLD is set at 1/15th of an ounce. If the answer to this question (IS IR¦ G~EATER T~AN THE CHANGE OF WEIGHT
'rHRESHOLD?), is affirmative, A~RAGE is then stored in the SAVED
WEIGHT register and the SAVED WEIG~T flag is set 124. If, however, the value in the R register is less than the CHAN~E
OF WEIGHT THR~SHOLD, AVERAGE is not stored in the SAVED
WEIGHT register. However, note that the SAV~D hEIGHT flag is nevertheless set 126. At this point, a new value is computed for the R register, by subtracting the value in the SAVED WEIG~T
register from AVERAGE 102. The system then determines whether the absolute value of the value in the R register is greater than or equal to CREEP THRES~OLD 104. If this absolute value is greater, than the value of R is added to the ~rARE register 128, the average is stored in the SAVED weight register and the SAVED
weight~flag is set 124. If the absolute value of the value in the R register is less than the CREEP THRESH~LD, the system then performs the step immediately after reference numeral 124. That is, the AVERAGE is stored in the L~ST VALI~ AVER~G~ register 106 and the NET W~IGHT is computed 108. lhe systew then determines whether the NET WEIG~T is greater than the value of the Z~R~
THRESHOLD 110. The NET WEI~hT is eventually converted to pounds and ounces 116.
heiqhing O~erations For purposes of illustration, it is now desirable to consider the operation of the system when a force is applied to the pan of the scale 10. After a weight is placed on the pan, 113'~1~7 digital numbers ~rom the A/D converter 12 are entered and stored 84 into the R~ADING buffer ~6. Initially, of course, the pan will be in motion 8~, so the SAVED ~iIGHT FL~G is cleared 118 and reading of the digital pulses generated by the A/D converter 12 continues to be made as shown at reference numeral 82. When the system determines that there is no significant diference among the values read, the answer to the question asked at reference numeral 88 tIS THERE ANY MOTION?), is answered in the negative.
A new AVERAGE is computed.
A value had previously been loaded in the TZ~RE register, so the answer to the question asked at reference numeral 92 tIS
TARE REGISTEI~ LO~DEV?-), is affirmative. Once again, the system determines whether the SAVI~D WEIGHT flag is set l00. Since this SAV~D W~IGHT flag had previously been cleared, the answer to this guestion is negative. The system then subtracts the value in the LAST VALID AVERAGE register from the AV~RAGE 120. The system then determines whether the absolute value of the value stored in the R register is greater than or equal to the CitANGI~ OF WEI~HT
TIIRESHOLD 122. If the answer to this question (IS ¦XI
GREATER THAN OR E~UAL TO THE CHANGE OF WEIGHT THRESHOLD?), is affirmative, AVERAGE is then stored in the SAVED WEIGHT
register and the SAVED WEIGH1~ flag is set 124. If, however, the value in the R register is less than the CHANGE OF
h`EIGHT T~3RESI~OLD, AVERAGE is not stored in the SAVE~ WEIGHT
register, but the SAVED WEIGHT flag is nevertheless set 1~o .
At this weighing the NET WEIGltT is most ~robably greater than the ZERO THRESHOL~ band t1~, so this system immediately converts the NET WEIGHT Into pounds and ounces 116.
.. - \~
.
.~ , .
. . .
113~
~ree~
~ inally, consider the situation that arises when a weight is placed on the pan of the scale 10 for a significant length of time. In this case, digital numbers are read from the A/D con-verter 12 shown at reference numeral B2. Values are stored 84 in the READING buffer until the READING buffer is full 86. It can be assumed that there is no significant motion of the pan on the scale 10, as shown at reference numeral 88. An AVERAGE is computed 90.
A value had been previously loaded into the TARE register, so the answer to the question asked at reference numeral 92 (IS
TA~E RE~I~TE~ LOADED?), is affirmative. The system then determines whether the SAVED ~7EIGIIT flag is set 100. Since there was no motion of the pan, the SAVED h~EIGHT flag is set. Consequently, the answer to the ~uestion asked at reference numeral 100 (IS
SAVED ~-EIG~T FLAG SET?), is affirmative. The system then sub-tracts the value in the SAVED WEIGHT register from AV~RAGE and stores this result in the R register 102. The system then determines whether the absolute value of the value stored in the R register is greater than or equal to the C~EEP THRESHOLD 104.
If the answer to this question is affirmative, the value in the R
register is added to the existing value in the TARE register 128.
The AVERAGE is then stored in tbe SAVED WEIGHT register and the SAVED WEIGHT flag is set 124. The AV~RAGE is then also stored in the LAST VALID AVERAGE register 106. The NET ~lEIGHT is then computed 108.
Since there is a significant weight on the scale l0, it is assumed that the new computed NET WEIGHT is greater than . _ . .. . ...
, .. .... .. ...... . .. . ... ..
~ i , . . .. .. .. . , -- . . .. , . , , , ~
1~3fjl~7 the Z~O THRESHOLD AND 110, so the NET WEIGHT is immediately converted to pounds and ounces 116.
From the foregoing description of both embodiments of the presene invention, it can be seen that a new method of tracking creep and drift of a digital scale has been disclosed. If, as a result of obtaining values over a period of time, the microcomputer system determines that the scale has crept or drifted substantially ~i.e., beyond an arbitrary predetermined band); an adjustment is made to nullify the effects of such creep and drift. Reference is made to FIG. 4 wherein A represents the tare weight and B
represents the gross weiqht after a load is placed on the pan.
Obviously, ~-A would yield the net weight. In the example re-ported by FIG. 4, the qross weight indication is shown increasing as a result of creep and drift in an exagerated manner for purposes of illustration by an amount C. When C exceeds a pre-determined value, as previously described, the value of B+C
will be stored as the new gross weight ~ and the value C will be added to the stored tare weight A to give a new tare weight A' which is stored in place of A. A' will be subtrated from B' to yield the true net weight. Additionally, when the load is removed from the pan the true tare weight A' will be stored so that a zero reading is displayed immediately.
~ .
L~
SlIIAI I3 CL~IMEL .3.
,
DRIFT IN A DIGI.AL SCALE UNDER FULL LOAD
~ackground of Invention Digital scales of current design utilize automatic zero tracking techniques which eliminate the need for frequent operator intervention to readjust the tare weight. Such automatic zero tracking usually involves measuring sequential signals representa-tive of the tare weight and continually replacing older signals with newer signals until such time as a load is placed on the pan. When the load is placed on the pan, the latest siqnal or signals would then be stored to be subsequently subtracted from the gross weight.
Prior methods of tracking have been with an empty pan only; consequently when a weight is left on a pan for any sig-nificant duration, deviation from zero can taXe place either as a result of electronic or load cell drifts. With typical load cell scales, significant change in the reading takes place after a heavy load is applied due to the creep of the load cell; however, the change in indicated weight immediately after the load is applied is unaffected. Thus, when a load is left on a scale and creep takes place, immediately after the load has been removed the zero reading has been displaced by the same change as the creep. In time, the scale would uncreep in a manner symmetrical with the original creep ~ depenaing on the original creep's rate and size, but a significant amount of time may elapse before zero is recaptured.
It therefore would be advantageous to have a system whereby the tare weight, or zero, is continually updated whether a load , v r ' _ 113~7 is upon the pan or not. With such a system, not only would time be saved in waiting until the tare weight has reached equilibrium after a load is removed from the pan, but also at ~he time the load is upon the pan the gross weight would not be accurate if creep and drift are not taken into account.
Summary of the Invention A method and apparatus has been devised whereby the tare weight of a scale may be continually updated whether or not a load is received upon the. pan of the scale. Before a load to be weighed is placed on the pan, the tare weight, or zero, is continuously updated by storing newer values and eliminating older values of the tare weight. This will continue until such time as a load is placed upon the pan, at which time the latest tare weight will be stored. The scale is allowed to reach equilibrium at which time the gross weight of the load will be measured and stored. At this time, the tare weight will be subtracted from the gross weight automatically and the net weigh~
shown on some type of display. When the load is left upon the pan for an extended period, the creep and drift are continually -monitored so that the stored tare weight can be adjusted in response thereto. The storing of the adjusted tare weight when a load is on the pan will result in a constant net weigh~ being obtained.
Brief Des. iption of the Drawing FIG. 1 shows a block d agram of the components utilized in the method of this invention.
_ .. . . , . _ . . _; _ . , .
.
- ~3~167 FIG. 2 is a flow chart of the logic involved ln one entodiment of the instant invention.
FIG. 3 is a flow chart of the logic involved in another en~xodiment of the instant invention.
FIG. 4 is a graph of the measurement by a scale of gross weight and net weight as a function of time.
Detailed Description of the Preferred Embodiment Referring to FIG. 1, a microcomputer-driven computing scale system is shown. A weighing scale 10 capable of generating analog signals, such as Scale Model No. 5035 produced by Pitney-Bowes, Inc., is electrically connected by means of an output line 11 to an analog to digital (A/D) converter 12. Included in the A/D converter 12 is control logic, such as TTL count~rs and gates. The A/D converter and control logic 12 is electrically connected to a microoamputer 14, such as Model No. PPS 4/1, produced by Rcckwell International Ccmpany, by means of suitable input and output lines, shown generally at reference numeral 16. These input and output lines 16 comprise a data line and a control line.
Ihe microcamput~r 14 is, in turn, oonnected to a clcck 18, a power supply 20 and a PC~E3R ON reset 22, by means of suitable input lines, shown generally at reference numeral 24. ~he microoomputer 14 is connected to a display 26, and drives the display 26 over a oorresponding output line 28.
In operation, a weight or load is placed on the pan of the scale 10, which scale generates analog signals over the cutput line 11.
m e A/D oonverter 12 receives the analog signal and converts it to a digital number. Ihe digital number is transmitted over output lines 16 to the microcomputer 14. m e microcomputer 14 is powered by the power supply 20 and its associated Pr~D~R ON reset 22, by means of suitable lines 24. The clock 18, - csm/ ~
113~i167 generates timing signals for the microcomputer 14 over a suitable line 24. The microcomputer 14 is used to calculate the weight of a load on the scale 10 represented by the signal generated therefrom. The weight is then displayed on the display 26 as a-result of the signal generated from the microcomputer 14 over the output line 28.
Initial Operation Referring now also to FIG. 2, the microcomputer 14 performs the steps hereinafter described. Before a weighing operation is begun, initialization 30 takes place. During initialization, all input/output (I/O) ports are cleared, all flags are cleared, and all variables are set to default values or to zero. The digital number generated by the A/D converter 12 is sent over the data output line 16 to the microcomputer 14. The number is read 32 and stored 34 in a READING buffer in the microcomputer 14.
Digital values continue to be read into the READING buffer until the buffer is full 36. A comparison of these values is made to determine whether they vary significantly with respect to one another. This determination is shown at reference numeral 38.
The significance of this determination is that when consecutive values do not vary significantly from one another over a given time interval, the pan of the scale 10 can be presumed to be at rest, or not` in motion. Such is generally the case when the system first becomes operational with no load on the pan, or when the pan has had time to ettle after a load is applied.
In the case of initial operation with no load on the pan of the scale 10, the answer to the question asked at reference .,.,.. , .. ., , . .. ~ , .. . . .. ...
11361~7 `
umeral ~ (IS THE~E ANY MOTION?~, is negative. An average of the va~ues in the READING buffer, hereinafter called AVERAGE, is then computed ~0.
~ If no value has been loaded into a TARE register 42, as is the case in initial operation, the value calculated as AVERAGE is then loaded into the TARE register 44. A SAV~D
WEIG~T flag is set 46 and AVER~GE is also loaded into a SAVLD
WEIGHT register 48. A calculation is ~en performed, subtracting the value stored in the SAVFD WEIGHT register from AVERAGE. The resulting difference is then stored in a register called the R register 50. In the case of initial operation, since AVERAGE
had been previously loaded in the SAVED WEIG~T register, the value zero is loaded into the R register.
The system then determines whether the SAVED WEIGHT flag has been set 52. In the case of initial operation, the flag has been set. The system then determines whether the absolute value of the value loaded in the R register is within an arbitrary band called the CREEP THRESHOLD 54. In this embodiment, the CREEP T~RESHOLD is set at 1/40th of an ounce. In the case of initial operation, since tlle value stored in the R register is zero, the answer to tbe question asked at reference numeral 54 (IS ~R¦ LESS THAN THE CREEP l1HRISHOLD?), i~ affirmative. ~lhat is, zero is less than 1/40th of an ounce. ~rhe NET ~EIG~T is then computed by subtractiny the value in the TARE register from AVER~GE 56.
The system then determines whether the value calculated as ~ET WEIG~T is greater than an arbitrary band called the ZERO
T~IRESHOLD 58. In this embodiment, the ZE~O THRESHOLD is set at 1/15th of an ounce. In the case of initial operation, since AVERAGE was loaded into the TARE register, the computed NET
..
.... . . . - - .. ~ ; -113~167 ~IGHT is zero. Therefore, the answer to the question ~sked ~t reference numeral 5~ ( IS NET WEIGIIT GREAT~R'T~AN THE ZERO
TIIRESHOLD?), is ne~ative. The value in the TARE register is then updated 60, and the value in the TARE register is then loaded into the SAVED WEIGHT register 62. The NET WEIG~T is then converted to pounds and ounces 64. Since no load is on the pan, this would be zero.
Pre-Weighing Operation At this point, the system continues to process signals generated by the scale 10 and converted into digital numbers by the A/D converter 12. The microcomputer 14 reads the number generated by the A/D converter 12, shown at reference numeral 32.
This reading is stored 34 and the oldest reading is discarded from the READING buffer 36.
It is conceivable that, after initial operation but before a weight is placed on the scale 10, slightly different readings will be received by the microcomputer 14 from the A/D converte_ 12 as a result of drift or noise. The system will then determine that there is motion of the pan 38, and will then clear the SAVED
~JEIGHT flag 66 before continulng to read a value from the A/D
converter 12 shown at reference numeral 32. A cleared SAVED
WEIGHT flag thus indicates that the pan of the scale 10 is in motion. When the scale 10 damps out, and no motion is de-tected, as determined by the comparison hereinbefore described at reference numeral 38, he system then computes the AV~RAGE of the values in the READING buffer 40. At this point, the TARE
register had been previously loaded during initial operation.
Consequently, the system does not execute the step~ represented by reference numerals 44, 46 and 48, but rather con~putes the new _ . . . . . ... . . . . . . . .... ... . .
113~
AV~hAGE mir.us the value in the SAVED WEIGIIT-register. Tbe result of this computation is then stored in the a register 50.
A~ this point, the condition of the SAVED WEIGHT flag is again determined 52. Since this SAVED WEIGHT flag had previously been cleared, the system then determines whether the value in the R register is less than an arbitrary band called the CHANGE OF
WEIGHT T~ S~IOLD 6~. In tllis embodiment, the CHANGE OF WEIGHT
THRESIIOLD is set at 1/lSth of an ounce. If the answer to this question (IS ~R¦ LESS TffAN THE CHANGE OF WEI~HT THRESHOLD?), is negative, AVERAGE is then stored in the SAVED WEIGHT register 70.
If, however, the value ~n the R register is less than the C~ANGE
OF WEIGHT THRESHOLD, AVERAGE is not stored in the SAVED WEIGHT
register. ~t this point, regardless of whether the new AVERAGE
is stored in the SAVED WEIGHT register, the SAVED WEIGHT flag is set 72 and the ~T WEIGHT is computed by subtracting the value in the TA~E register from AVERAGE 56. Since no weight is on the pan of the scale 10, it is probable that the NET WEIGHT is not greater than the ZERO T~RESHOLD band 58. Consequently, the TARE
regis~er must be updated 60. The value in the TAXE register is stored also in the SAVED WEIGHT register 62. l'he NET W~IGHT is again converted to pounds and ounces 64.
Weighing Operation For purposes of illustration, it is now desirable to consider the operation of the system when a force .s applied ~o the pan of the scale 1C After a weight is placed on the pan, a digital number from the A/D converter 12 ia entered and stored 34 into the READI~G buffer 36. Initially, of course, the pan will be in motion 38, so the SAVED W~IG~T flag is cleared 66 and .,, ,,, . ~ . . .. . . . . . . . .. ... .
. , , ~, . ,, ~ , , .
- ` 113~67 reading of the digital numbers generated by tbe A~D converter ~2 continues to be made as shown at reference numeral 32. When the system determines that there is no significant difference among the values read, the answer to the question asked at reference numeral 38 (IS Ti~ERE ANY i~OTION?), is answered in the negative.
A new AVERAGi is computed.
A value had previously been loaded in the TARE register, so the answer to the question asked at reference numeral 42 (IS
TARE REGISTE~ LOADED?), is affirmative. rrhe value in the SAVED
WEIGHT register is now subtracted from the new AViRAGE and this result is stored in the R register 50. Because the SAVED WEIGHT
flag had previously been cleared, the answer to the question asked at reference numeral 52 (IS SAVED Wk~IGHT PLAG SET?), is negative. The system then determines whether the absolute value of the value stored in the R register is less than the CHANGE OF WEIGHT TilRESHOLD 68. If the answer to this question (IS ~R! LESS THAN THE CHANGE OE~ WEIGHT TIJRE~IIOLD?J, is negative, AVERA~E is then stored in the S~VED W~IGI~T
register 7~. If, however, the value in the R register is less than the CHANGE OF WEIGHT THRESHOLD, AVERAGE is not stored in the SAVEi~ W~IGHT reyister. At this point, regard-less of whether the new AVERAGE is stored in the SAVED
WEIGHT register, the SAVED WEIGHT flag is set 72 and the NET
WEIGHT is computed by subtrzcting the value in the TARE
register from AVERAGE 56 At this weighing, the N~T WEIGI~T is most probably greater than the ZERO THRESHOLD band 58, so the system immediately converts the NET WEIGHT into pounds and ounces 64.
--a-- !-;
~, .. . .;.... . ... , . _ _, . . .. . . . _ . ..
1136i67 Creep l`rackinq Uperation Now consider the situation that arises after a weight has been determined ~o be on the pan of the scale 10 for a significant length of time. In this case, a digital number is read from the A/D converter 12, shown at reference numeral 32. The value is stored 34 in the READING buffer 36. It can be assumed that there is no significant motion of the pan on the scale 10, as shown at reference numeral 38. An AVERAGE is computed 40.
A value had been previously loaded into the TARE register, so the answer to the question asked at reference numeral 42 (IS
l'ARE REGISTER LOADED?), is affirmative. The system then subtracts the value in the SAVED WEIGHT registe, from AVE~AGE and stores this result in the R register 50. Note that since there was no motion of the pan, the SAV~D WEIG~T flag is set. Consequently, the answer to the question asked at reference numeral 52 (IS
SAVED WEIGHT FLAG SET?~, is affirmative. The system then determines whether the absolute value of the value in the R
register is less than the cxrEp THRES~OLV 54. For purposes of illustration, it is assumed that the absolute value of the value in the R register is greater than the CREEP TliRES~VLD. The value in the TARE register is then updated with a value consisting of the old TARE register value plus the value in the R register 74. The AVERAGE is then stored in the SAVE~ WEIGHT register 70 and the SAVED WEIGHT flag is set 72 (although this is a redundant step). The NET WEI~iT is computed by subtracting the new value in the TARE reoister from AVERAGE 56.
Since there is a significant weight on the scale 10, it is assumed that the new computed NET WEIGHT is greater than the ~ERO THR~'S~i~L~ band ~8, sv the NET h~IGtiT is immediately converted to pounds and ounces 64. r _9_ 113fj~i7 Alternate Embodiment -- Initial Operation - Xeferring now to FIG. 3 in conjunction with FIG. 1, an .alternate embodiment is hereinbelow described. The microcomputer 14 performs steps as follows. ~efore a weighing operation is begun, initialization 80 takes place. During initialization all input/output (I/O) ports are cleared, all flags are cleared, and all variables are set to default values or to ze~o. The digltal number generated by the A/D converter 12 is sent over the data output line 16 to the microcomputer 14. The number is read 82 and stored 84 in a READING buffer in the microcomputer 14.
Digital numbers continue to be read into the READING buffer until the buffer is full 86. A comparison of these values is made to determine whether they vary significantly with respect to one another. This determination is shown at reference numeral 8B.
The significance of this determination is that when consecutive values do not vary significantly from one another over a given time interval, the pan of the scale 10 can be presumed to be at rest or not in motion. Such is generally the case when the system first becomes operational with no load on the pan, or when a load on the pan has time to settle (i.e., scale dampening is completed).
In the case of initial operation with no load on the pan of the scale 10, the answer to the question asked at referenced numeral ~ (IS ~HERE ANY MorrIoN?)~ is negative. An average of the values in the REA~ING buffer, hereinafter called AVERAGE, is then computed 90.
If no value has been loaded into the TAR register 92, as is the case in initial operation, the value calculated as AVERAGE is been loaded into the TARE register 94. The AVRA~E is ~ . , . .. , , . ~ _ , .
1~3~1~7 louded into a SA~ED WEIGHT register 96 and a SAVED WEI~HT fl~g is also set 9~
The system then determines whether the SAYED WEIGH~ flag has been set 100. In the case of initial operation, the flag has been set. The system then performs a calculation, subtracting the value stored in the SAVED WEIG~T register from AVERAGE 102.
The resulting difference is then stored in a register called the R register 102. In the case of initial operation, since AVERAGE
has been previously loaded in the SAV~D WEIGHT register, the value zero is loaded into the R register. The system then determines whether the absolute value of the value loaded in the R register is within the arbitrary band called the CREEP THR~SHOLD
104. In this embodiment, the CREEP THR~SHOLD is set at 1/40th of an ounce. In the case of initial operation, since the value stored in the R register is zero, the answer to the question asked àt reference numeral 104 (IS ~R¦ ~REATER THAN OR EQUAL TO THE
CRE~P THRESHOLD?), is negative. That is, zero is not greater than 1/4~th of an ounce. The AVERAGE is stored in a LAS'r VALI~
AVERAGE register 106. The NET WEIGHT is then computed by sub-tracting the value in the TARE register from AVE~GE 108.
The system then determines whether the value calculated as NET WEIGHT is greater than the arbitrary band called the ZERO
THRESflOLD 110. In this embodiment, the ZfRO THR~SHOLD is set at 1/15th of an ounce. In the case of initial operation, since AVERAGE was loaded into the TARE register, the computed NET
WEIGHT is zero. Therefore, the answer to the question asked at reference numeral 110 ~IS ~ET WEIGHT GX~ATfR TflAN THE ZERO
THRESHOLD?), is negative. The value in the TARE register is then updated if required 112, and the value in the TARE register is loaded into the SA~ED WEIGHT register 114. The NE~r WEI~HT is ~_ . , . . , . _ 113~i~7 .
ehe c~nverted to pounds and ounces 116. Since no load is on the pan, this would be zero.
Pre-Weiqhinq ~peration At this point the system continues to process signals generated by the scale 10 and converted into digital numbers by the A/D converter 12. The microcomputer 14 reads the signals generated by the A/D converter 12, shown at reerence numeral 82.
Once again, these readings are stored 84 in the READI~G buffer 86.
It is conceivable that, after initial operation but before a weight is placed on .the scale 10, slightly different readings will be received by the microcomputer 14 from the A/D converter 12 as a result of drift or noise. The system will then determine ~o~
that there is motion of the pan ~8 and will then clear the SAVED
WEIG~T flag 118 before continuing to read values from the A/D
converter shown a reference numeral 82. A cleared SAVED W~IGHl' flag thus indicates that the pan of the scale 1~ is in motion.
When the scale 10 damps out, and no motion is detected, as determined by the comparison hereinbefore described at reference numeral ~8, the system then computes the AVERAGE of the values in the READING buffer 90. At this point, the TARE register had been previously loaded during initial operation. Consequently, the system does not execute the steps represented by reference numerals 94, 96 and 98, but rather determines whether the SAVED
WElGH~ flag is set 100. Since this SAVPD WEIG~T flag had pre-viously been cleared, the system then suotracts the value stored in the LAST VA~ID AVERA~E register from the AVERAG~, and stores this value in ehe R register 120. The system then determines whether the value in the R register is greater than or equal to \~
... _ _ _ .......
~ . . . . ..
113i167 the CHANGE O~ IYEIGHT THRESIIOLD 122. In this embodiment, the CHANGE OF WEI~T TH~ESHOLD is set at 1/15th of an ounce. If the answer to this question (IS IR¦ G~EATER T~AN THE CHANGE OF WEIGHT
'rHRESHOLD?), is affirmative, A~RAGE is then stored in the SAVED
WEIGHT register and the SAVED WEIG~T flag is set 124. If, however, the value in the R register is less than the CHAN~E
OF WEIGHT THR~SHOLD, AVERAGE is not stored in the SAVED
WEIGHT register. However, note that the SAV~D hEIGHT flag is nevertheless set 126. At this point, a new value is computed for the R register, by subtracting the value in the SAVED WEIG~T
register from AVERAGE 102. The system then determines whether the absolute value of the value in the R register is greater than or equal to CREEP THRES~OLD 104. If this absolute value is greater, than the value of R is added to the ~rARE register 128, the average is stored in the SAVED weight register and the SAVED
weight~flag is set 124. If the absolute value of the value in the R register is less than the CREEP THRESH~LD, the system then performs the step immediately after reference numeral 124. That is, the AVERAGE is stored in the L~ST VALI~ AVER~G~ register 106 and the NET W~IGHT is computed 108. lhe systew then determines whether the NET WEIG~T is greater than the value of the Z~R~
THRESHOLD 110. The NET WEI~hT is eventually converted to pounds and ounces 116.
heiqhing O~erations For purposes of illustration, it is now desirable to consider the operation of the system when a force is applied to the pan of the scale 10. After a weight is placed on the pan, 113'~1~7 digital numbers ~rom the A/D converter 12 are entered and stored 84 into the R~ADING buffer ~6. Initially, of course, the pan will be in motion 8~, so the SAVED ~iIGHT FL~G is cleared 118 and reading of the digital pulses generated by the A/D converter 12 continues to be made as shown at reference numeral 82. When the system determines that there is no significant diference among the values read, the answer to the question asked at reference numeral 88 tIS THERE ANY MOTION?), is answered in the negative.
A new AVERAGE is computed.
A value had previously been loaded in the TZ~RE register, so the answer to the question asked at reference numeral 92 tIS
TARE REGISTEI~ LO~DEV?-), is affirmative. Once again, the system determines whether the SAVI~D WEIGHT flag is set l00. Since this SAV~D W~IGHT flag had previously been cleared, the answer to this guestion is negative. The system then subtracts the value in the LAST VALID AVERAGE register from the AV~RAGE 120. The system then determines whether the absolute value of the value stored in the R register is greater than or equal to the CitANGI~ OF WEI~HT
TIIRESHOLD 122. If the answer to this question (IS ¦XI
GREATER THAN OR E~UAL TO THE CHANGE OF WEIGHT THRESHOLD?), is affirmative, AVERAGE is then stored in the SAVED WEIGHT
register and the SAVED WEIGH1~ flag is set 124. If, however, the value in the R register is less than the CHANGE OF
h`EIGHT T~3RESI~OLD, AVERAGE is not stored in the SAVE~ WEIGHT
register, but the SAVED WEIGHT flag is nevertheless set 1~o .
At this weighing the NET WEIGltT is most ~robably greater than the ZERO THRESHOL~ band t1~, so this system immediately converts the NET WEIGHT Into pounds and ounces 116.
.. - \~
.
.~ , .
. . .
113~
~ree~
~ inally, consider the situation that arises when a weight is placed on the pan of the scale 10 for a significant length of time. In this case, digital numbers are read from the A/D con-verter 12 shown at reference numeral B2. Values are stored 84 in the READING buffer until the READING buffer is full 86. It can be assumed that there is no significant motion of the pan on the scale 10, as shown at reference numeral 88. An AVERAGE is computed 90.
A value had been previously loaded into the TARE register, so the answer to the question asked at reference numeral 92 (IS
TA~E RE~I~TE~ LOADED?), is affirmative. The system then determines whether the SAVED ~7EIGIIT flag is set 100. Since there was no motion of the pan, the SAVED h~EIGHT flag is set. Consequently, the answer to the ~uestion asked at reference numeral 100 (IS
SAVED ~-EIG~T FLAG SET?), is affirmative. The system then sub-tracts the value in the SAVED WEIGHT register from AV~RAGE and stores this result in the R register 102. The system then determines whether the absolute value of the value stored in the R register is greater than or equal to the C~EEP THRESHOLD 104.
If the answer to this question is affirmative, the value in the R
register is added to the existing value in the TARE register 128.
The AVERAGE is then stored in tbe SAVED WEIGHT register and the SAVED WEIGHT flag is set 124. The AV~RAGE is then also stored in the LAST VALID AVERAGE register 106. The NET ~lEIGHT is then computed 108.
Since there is a significant weight on the scale l0, it is assumed that the new computed NET WEIGHT is greater than . _ . .. . ...
, .. .... .. ...... . .. . ... ..
~ i , . . .. .. .. . , -- . . .. , . , , , ~
1~3fjl~7 the Z~O THRESHOLD AND 110, so the NET WEIGHT is immediately converted to pounds and ounces 116.
From the foregoing description of both embodiments of the presene invention, it can be seen that a new method of tracking creep and drift of a digital scale has been disclosed. If, as a result of obtaining values over a period of time, the microcomputer system determines that the scale has crept or drifted substantially ~i.e., beyond an arbitrary predetermined band); an adjustment is made to nullify the effects of such creep and drift. Reference is made to FIG. 4 wherein A represents the tare weight and B
represents the gross weiqht after a load is placed on the pan.
Obviously, ~-A would yield the net weight. In the example re-ported by FIG. 4, the qross weight indication is shown increasing as a result of creep and drift in an exagerated manner for purposes of illustration by an amount C. When C exceeds a pre-determined value, as previously described, the value of B+C
will be stored as the new gross weight ~ and the value C will be added to the stored tare weight A to give a new tare weight A' which is stored in place of A. A' will be subtrated from B' to yield the true net weight. Additionally, when the load is removed from the pan the true tare weight A' will be stored so that a zero reading is displayed immediately.
~ .
L~
SlIIAI I3 CL~IMEL .3.
,
Claims (15)
PROPERTY OR PRIVILEGE IS CLAIMED ARE DEFINED AS FOLLOWS:
1. A method for tracking creep and drift in an electronic scale during the period when a load remains on the scale, comprising:
a) continuously updating the tare weight of the scale prior to placing said load on the scale;
b) placing a load on the scale;
c) storing the current tare weight recorded at the time the load is placed upon the scale;
d) storing the gross weight of the load;
e) tracking the variance in the gross weight as a result of creep and drift; and f) updating the stored tare weight by adding thereto the variance in the gross weight while the load remains on the scale.
a) continuously updating the tare weight of the scale prior to placing said load on the scale;
b) placing a load on the scale;
c) storing the current tare weight recorded at the time the load is placed upon the scale;
d) storing the gross weight of the load;
e) tracking the variance in the gross weight as a result of creep and drift; and f) updating the stored tare weight by adding thereto the variance in the gross weight while the load remains on the scale.
2. The method of claim 1 wherein the step of storing the gross weight of the load, comprises:
a) storing a plurality of digital numbers generated by the scale representative of a weight;
b) continually sampling the digital numbers until none of the digital numbers-vary significantly with respect to one another; and c) computing an average of the digital numbers.
a) storing a plurality of digital numbers generated by the scale representative of a weight;
b) continually sampling the digital numbers until none of the digital numbers-vary significantly with respect to one another; and c) computing an average of the digital numbers.
3. The method of claim 1 wherein the step of tracking the variance in the gross weight as a result of creep and drift, comprises:
a) storing a plurality of digital numbers generated by the scale representative of a weight;
b) continually sampling the digital numbers until none of the digital numbers vary significantly with respect to one another;
c) computing an average of the digital numbers;
d) repeating the steps of a) through c) to obtain a second average;
e) computing the difference between the first and second averages.
a) storing a plurality of digital numbers generated by the scale representative of a weight;
b) continually sampling the digital numbers until none of the digital numbers vary significantly with respect to one another;
c) computing an average of the digital numbers;
d) repeating the steps of a) through c) to obtain a second average;
e) computing the difference between the first and second averages.
4. The method of claim 1 wherein the step of updating the stored tare weight by adding thereto the variance in the gross weight while the load remains on the scale, comprises:
a) storing a plurality of digital numbers generated by the scale representative of a weight;
b) continually sampling the digital numbers until none of the digital numbers vary significantly with respect to one another;
c) computing an average of the digital numbers;
d) establishing a creep threshold;
e) repeating the steps of a) through c) to obtain a second average;
f) computing the difference between the first and second averages; and g) adjusting the tare weight to compensate for creep thereof by adding thereto the difference between the first and second averages if said difference exceeds the creep threshold.
a) storing a plurality of digital numbers generated by the scale representative of a weight;
b) continually sampling the digital numbers until none of the digital numbers vary significantly with respect to one another;
c) computing an average of the digital numbers;
d) establishing a creep threshold;
e) repeating the steps of a) through c) to obtain a second average;
f) computing the difference between the first and second averages; and g) adjusting the tare weight to compensate for creep thereof by adding thereto the difference between the first and second averages if said difference exceeds the creep threshold.
5. The method of claim 4 wherein the first average is stored as a tare weight.
6. A method of compensating for creep and drift in an electronic scale after a load is placed on the pan of the scale, comprising:
a) continuously updating a value representative of a tare weight;
b) sensing whether a load is placed on the pan;
c) discontinuing the tare weight value updating when a load is placed on the pan and storing the updated tare weight;
d) obtaining a first value representative of the gross weight on the scale;
e) storing the first value;
f) obtaining a second value representative of the gross weight on the scale;
g) storing the second value;
h) computing the difference between the first and second values;
i) establishing a creep threshold; and j) updating the tare weight value to compensate for creep thereof by adding thereto the difference when the difference between the first and second values is greater than the creep threshold and no load has been placed on the pan during the interval between the time at which the first value has been obtained and the time at which the second value has been obtained.
a) continuously updating a value representative of a tare weight;
b) sensing whether a load is placed on the pan;
c) discontinuing the tare weight value updating when a load is placed on the pan and storing the updated tare weight;
d) obtaining a first value representative of the gross weight on the scale;
e) storing the first value;
f) obtaining a second value representative of the gross weight on the scale;
g) storing the second value;
h) computing the difference between the first and second values;
i) establishing a creep threshold; and j) updating the tare weight value to compensate for creep thereof by adding thereto the difference when the difference between the first and second values is greater than the creep threshold and no load has been placed on the pan during the interval between the time at which the first value has been obtained and the time at which the second value has been obtained.
7. The method of claim 6, including calculating the net weight of the load on the pan after each value of the gross weight on the scale has been obtained by subtracting the updated tare weight value from the value representative of the gross weight.
8. The method of claim 6, further comprising:
k) establishing a change of weight treshold;
and l) discontinuing the tare weight value updating as recited in step (j) when the difference between the first and second values exceeds the change of weight threshold.
k) establishing a change of weight treshold;
and l) discontinuing the tare weight value updating as recited in step (j) when the difference between the first and second values exceeds the change of weight threshold.
9. The method of claim 8 including calculating the net weight of the load on the pan after each value of the gross weight on the scale has been obtained, by subtracting the updated tare weight value from the value representative of the gross weight.
10. The method of claim 6 wherein said first value representative of the gross weight on the scale is an average of a plurality of values and the second value representative of the gross weight on the scale is an average of a plurality of values.
11. Apparatus for compensating for creep and drift in an elec-tronic scale after a load is placed on a pan of the scale, com-prising:
a) means for continously updating a tare weight register;
b) means for sensing motion of the pan as a result of a load being placed on the pan;
c) means for discontinuing the tare weight register updating;
d) means for obtaining a first value representative of the gross weight on the scale;
e) a first storage register for storing the first value;
f) means for obtaining a second value representative of the gross weight on the scale;
g) a second storage register for storing the second value;
h) means for computing the difference between the first and second values; and i) means for updating the tare weight register whereby an adjustment of the value in the tare weight re-gister for creep and drift is made when no motion of the pan has occurred between obtaining the first and second values.
a) means for continously updating a tare weight register;
b) means for sensing motion of the pan as a result of a load being placed on the pan;
c) means for discontinuing the tare weight register updating;
d) means for obtaining a first value representative of the gross weight on the scale;
e) a first storage register for storing the first value;
f) means for obtaining a second value representative of the gross weight on the scale;
g) a second storage register for storing the second value;
h) means for computing the difference between the first and second values; and i) means for updating the tare weight register whereby an adjustment of the value in the tare weight re-gister for creep and drift is made when no motion of the pan has occurred between obtaining the first and second values.
12. The apparatus of claim 11 including means for calculating the net weight of the load on the pan after each value of the gross weight on the scale has been obtained.
13. The apparatus of claim 11, wherein the means for sensing motion of the pan is a settable flag register.
14. The apparatus of claim 13, further comprising:
j) a creep threshold; and k) means for updating the tare weight value in ac-cordance with paragraph (i) when the difference be-tween the first and second values is greater than the creep threshold.
j) a creep threshold; and k) means for updating the tare weight value in ac-cordance with paragraph (i) when the difference be-tween the first and second values is greater than the creep threshold.
15. The apparatus of claim 14, further comprising:
1) a change of weight threshold; and m) means for discontinuing the tare weight value up-dating as recited in paragraph (i) when the dif-ference between the first and second values exceeds the change of weight threshold.
1) a change of weight threshold; and m) means for discontinuing the tare weight value up-dating as recited in paragraph (i) when the dif-ference between the first and second values exceeds the change of weight threshold.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| US06/077,468 US4412298A (en) | 1979-09-20 | 1979-09-20 | Method for tracking creep and drift in a digital scale under full load |
| US077,468 | 1979-09-20 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CA1136167A true CA1136167A (en) | 1982-11-23 |
Family
ID=22138233
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CA000360669A Expired CA1136167A (en) | 1979-09-20 | 1980-09-19 | Method and apparatus for tracking creep and drift in a digital scale under full load |
Country Status (6)
| Country | Link |
|---|---|
| US (1) | US4412298A (en) |
| JP (1) | JPS5696225A (en) |
| CA (1) | CA1136167A (en) |
| DE (1) | DE3035126A1 (en) |
| FR (1) | FR2466005A1 (en) |
| GB (1) | GB2060172B (en) |
Families Citing this family (38)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CH639484A5 (en) * | 1980-07-10 | 1983-11-15 | Mefina Sa | WEIGHING APPARATUS. |
| JPS5862525A (en) * | 1981-10-09 | 1983-04-14 | Shimadzu Corp | Electronic balance |
| JPS58153332U (en) * | 1982-03-31 | 1983-10-13 | 株式会社島津製作所 | electronic balance |
| JPS593322A (en) * | 1982-06-30 | 1984-01-10 | Tokyo Electric Co Ltd | Zero tracking device of load cell balance |
| US4535854A (en) * | 1982-11-16 | 1985-08-20 | Pennsylvania Scale Company | Calibration method and apparatus for an electronic weight indicator |
| US4531599A (en) * | 1983-08-01 | 1985-07-30 | Langdon Wales R | Method and apparatus for weighing waste signatures from a printing press |
| US4691290A (en) * | 1984-08-06 | 1987-09-01 | Reliance Electric Company | Creep-compensated weighing apparatus |
| US4787046A (en) * | 1985-11-12 | 1988-11-22 | Pitney Bowes Inc. | Mailing system having a capability for one-step postage metering |
| CH670507A5 (en) * | 1986-05-23 | 1989-06-15 | Mettler Instrumente Ag | |
| US4760539A (en) * | 1986-05-27 | 1988-07-26 | Ncr Corporation | Method of calibrating a weighing apparatus |
| US4660663A (en) * | 1986-05-27 | 1987-04-28 | Ncr Corporation | Method of calibrating a weighing apparatus within an enclosure |
| US4831527A (en) * | 1986-08-11 | 1989-05-16 | Clark Lloyd D | Dynamic measuring system for determining the amount of excess body fat |
| US4787048A (en) * | 1986-09-19 | 1988-11-22 | Pitney Bowes Inc. | Postal weighing apparatus and method |
| US4849918A (en) * | 1987-12-30 | 1989-07-18 | Pitney Bowes Inc. | Weighing instrument having adaptive breakpoints |
| US5463192A (en) * | 1988-12-14 | 1995-10-31 | Wirth Gallo Messtechnik Ag | Process for operating a measurement instrument, and measurement instrument suitable for use with this process |
| EP0454916A1 (en) * | 1990-04-30 | 1991-11-06 | Yamato Scale Company, Limited | Device for compensating for time-dependent error due to creep and like of measuring apparatus |
| US5166892A (en) * | 1990-04-30 | 1992-11-24 | Yamato Scale Company, Limited | Device for compensating for time-dependent error due to creep and like of measuring apparatus |
| DE4022512A1 (en) * | 1990-07-14 | 1992-01-23 | Sartorius Gmbh | ELECTRIC SCALE WITH CORRECTION OR CORRECTION PROGRAM |
| US9477638B2 (en) * | 1995-06-12 | 2016-10-25 | Circuits & Systems, Inc. | Surface acoustic wave scale that automatically updates calibration information |
| DE19523034C1 (en) * | 1995-06-24 | 1996-09-05 | Sartorius Gmbh | Drift elimination system for electronic weighing device |
| US6080938A (en) * | 1997-08-13 | 2000-06-27 | Ncr Corporation | Method and apparatus for resetting a product scale of a retail checkout terminal |
| US5987426A (en) | 1997-10-14 | 1999-11-16 | Ncr Corporation | Point-of-sale system including isolation layer between client and server software |
| US6215078B1 (en) * | 1998-12-22 | 2001-04-10 | Ncr Corporation | Method and apparatus for determining a stable weight measurement for use in a security software application of a self-service checkout terminal |
| US6856964B1 (en) * | 1999-03-22 | 2005-02-15 | Ncr Corporation | System and methods for integrating a self-checkout system into an existing store system |
| DE10041251B4 (en) * | 2000-08-23 | 2010-03-11 | HBM Wägetechnik GmbH | Verifiable weighing system and method for the determination of legally required measured value data |
| ITVR20020028A1 (en) | 2002-03-28 | 2003-09-29 | Moretto Plastics Automation Sr | GRAVIMETRIC DOSING EQUIPMENT AND MIXING OF GRANULAR PRODUCTS |
| US6886746B1 (en) | 2002-12-23 | 2005-05-03 | Ncr Corporation | Method and apparatus for separating noise from a weight signal of a self-service checkout terminal |
| JPWO2006132235A1 (en) * | 2005-06-07 | 2009-01-08 | 株式会社島津製作所 | Load cell electronic balance |
| PL1736746T3 (en) * | 2005-06-21 | 2012-12-31 | Mettler Toledo Gmbh | Method for optimizing the behavior of a force measuring device and force measuring device for carrying out the method. |
| US8599025B2 (en) * | 2011-01-04 | 2013-12-03 | General Electric Company | System and method to determine an infant weight while minimizing disturbance to the infant |
| CN103196528B (en) * | 2013-03-07 | 2015-02-11 | 郑州恒科实业有限公司 | Period compensation type weighing methodof electronic hoist scale |
| EP2910914B1 (en) * | 2014-02-21 | 2018-01-31 | Multipond Wägetechnik GmbH | Weighing device and method for operating the weighing device |
| US10098355B2 (en) * | 2014-11-19 | 2018-10-16 | Bakery Systems, Inc. | System and method for weighing items |
| NL2016315B1 (en) * | 2016-02-24 | 2017-01-17 | Xyztec B V | Digital creep and drift correction. |
| CN109063325B (en) * | 2018-07-30 | 2024-02-23 | 白杨 | Goods quantity calculating method based on weighing |
| CN109145398B (en) * | 2018-07-30 | 2024-01-05 | 白杨 | Goods quantity calculating method based on weighing |
| US11300440B2 (en) * | 2019-11-14 | 2022-04-12 | Blackberry Limited | Method and system for determining the weight of a load on a chassis |
| CN113694305B (en) * | 2021-09-01 | 2023-03-24 | 深圳市冠辰科技有限公司 | Weighing sensing dynamic compensation method and device, medium and electronic equipment |
Family Cites Families (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US3962570A (en) * | 1974-01-18 | 1976-06-08 | Reliance Electric Company | Scale with manual tare entry |
| US3986012A (en) * | 1974-05-02 | 1976-10-12 | Reliance Electric Company | Digital weight measuring and computing apparatus with automatic zero correction |
| US3916173A (en) * | 1974-06-27 | 1975-10-28 | Reliance Electric Co | Plural scale system with digital zero correction |
| US3951221A (en) * | 1974-09-30 | 1976-04-20 | National Controls, Inc. | Computing postal scale |
| DE2506985C2 (en) * | 1975-02-19 | 1986-08-21 | Reliance Electric Co., Toledo, Ohio | Procedure for correcting zero point errors on a digital scale |
| AR208424A1 (en) * | 1975-03-07 | 1976-12-27 | Timi Vaaka Hyholm | AN AUTOMATIC DEVICE FOR RESETTING SCALES WITH NUMERICAL DISPLAY |
| US4012953A (en) * | 1976-03-18 | 1977-03-22 | Ormond Alfred N | Short term creep compensation for load cells |
| DE2625476A1 (en) * | 1976-06-05 | 1977-12-22 | Berkel Patent Nv | ELECTRIC EVALUATION DEVICE OF SCALES WITH SELF-ADJUSTMENT OF THE ZERO POINT POSITION |
| JPS5327062A (en) * | 1976-08-25 | 1978-03-13 | Kubota Ltd | Zero tracking circuit |
| US4159521A (en) * | 1976-10-05 | 1979-06-26 | Reliance Electric Company | Digital scale with antifraud features |
| JPS548553A (en) * | 1977-06-21 | 1979-01-22 | Kubota Ltd | Electronic weighing machine |
| US4139069A (en) * | 1977-06-23 | 1979-02-13 | Acurex Corporation | Digital weighing method |
| US4139892A (en) * | 1977-07-18 | 1979-02-13 | Triner Scale And Manufacturing Company | Electronic postage scale |
| JPS6052365B2 (en) * | 1977-07-29 | 1985-11-19 | 東芝テック株式会社 | Weight measurement method and device |
| US4236222A (en) * | 1979-01-15 | 1980-11-25 | Reliance Electric Company | Digital scale |
-
1979
- 1979-09-20 US US06/077,468 patent/US4412298A/en not_active Expired - Lifetime
-
1980
- 1980-09-10 GB GB8029201A patent/GB2060172B/en not_active Expired
- 1980-09-17 DE DE19803035126 patent/DE3035126A1/en active Granted
- 1980-09-18 FR FR8020079A patent/FR2466005A1/en active Granted
- 1980-09-19 CA CA000360669A patent/CA1136167A/en not_active Expired
- 1980-09-20 JP JP13142480A patent/JPS5696225A/en active Granted
Also Published As
| Publication number | Publication date |
|---|---|
| JPH0346766B2 (en) | 1991-07-17 |
| DE3035126C2 (en) | 1992-04-23 |
| FR2466005B1 (en) | 1985-01-25 |
| US4412298A (en) | 1983-10-25 |
| DE3035126A1 (en) | 1981-04-09 |
| FR2466005A1 (en) | 1981-03-27 |
| GB2060172A (en) | 1981-04-29 |
| JPS5696225A (en) | 1981-08-04 |
| GB2060172B (en) | 1984-04-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CA1136167A (en) | Method and apparatus for tracking creep and drift in a digital scale under full load | |
| EP0171237B1 (en) | Creep-compensated weighing apparatus | |
| US4535854A (en) | Calibration method and apparatus for an electronic weight indicator | |
| US3665169A (en) | Electronic measuring unit | |
| US4156472A (en) | Apparatus for automatically changing the integration time and resolution of an electrical scale under predetermined operating conditions | |
| US4139892A (en) | Electronic postage scale | |
| US3869004A (en) | Method of fine adjusting balances and balance for performing this method | |
| US4890246A (en) | Electronic balance with calibrating system | |
| EP0124355B1 (en) | Zero-adjustment in weighing apparatus | |
| US4782904A (en) | Electronic balance | |
| US3439524A (en) | Automatic weighing scale calibration | |
| JPH0568645B2 (en) | ||
| JPS62280624A (en) | Automatic adjusting method of high-sensitivity electronic balance | |
| US4914611A (en) | Force measuring device | |
| CN113701859A (en) | Weighing counting method, device, system, computer equipment and readable storage medium | |
| US4344494A (en) | Measuring system having a balance with digital display having digit appearance modifying circuit | |
| CA1097703A (en) | Scale with digital display | |
| US4457386A (en) | Temperature-corrected weighing apparatus having electromagnetic load compensation means | |
| US4512429A (en) | Weighing apparatus for producing geographically-corrected measurements | |
| US4703815A (en) | Span adjusting device for weigher | |
| US4525794A (en) | Electronic balance | |
| KR910001147B1 (en) | Weighting meter system | |
| JPS62228118A (en) | Electronic balance | |
| US4792003A (en) | Span adjusting device for weigher | |
| SU1089423A1 (en) | Digital balance |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| MKEX | Expiry |