Note : Les descriptions sont présentées dans la langue officielle dans laquelle elles ont été soumises.
7~
This is a divisional of Canadian Patent Application
Serial No. 285,325 filed August 23, 1977.
- The present invention relates to force coils for
use in apparatus for determining the amount and location of
imbalance in an article which is to be balanced about an axis.
Machines which determine the amount and location of
imbalance in parts which are to be balanced about preselected
axes by suspending such parts from their axes an~d allowing the
parts to tilt in response to the imbalance are commonly referred
to as static balancing machines. Typically such machines measure
and interpret the amount of tilt to obtain an indication of the
amount and location of imbalance. Static balancing machines are
generally accepted as being reliable, simple to operate, and
accurate. However, static balancing machines su~fer ce~tain
shortcomings. For example, most static balancing machines
include fixtures, such as spindles or chuc~s, upon which parts
to be processed are positioned. When a part is suspended from
its axis upon a static balancing machine and the machine spindle
`~ which supports the part begins to tilt, the mass of the tilting
spindle of the machine itself enters into any determination of
the amount and location of imbalance in the part.
:~ The above-noted parent application describes and
claims an apparatus for determining imbalance in an article
~ which is to be balanced about an axis includes a base and means
- for receiving and posltioning the article. The article receiving
means is supported upon the base for substantially free tilting
movement ~enerally about a point on a vertical axis defined by the
base. The article receiving means may include a spindle,
chuck of other locating apparatus
so that the axes of articles placed thereon will coincide with
a predetermined location on the article receiving means. The
apparatus includes means ~or sensing the orientation of the
article receiving means and means for applying a force at at
.. .. .
least one salected location to the article receiving means to
orient it at a substantially neutral position. The axis may
be an axis of rotation or some other axis of symmetry of the
article.
The terms ''neutral position", "neutral orientation"
and the like as used herein shall mean a balanced or rest posi-
tion or orientation. Typically, when the disclosed system is in
such position or orientation, the article receiving means will
be positioned so that an article being checked for balance will
assume a generally horizontal orientation.
The above-noted means for sensing the
orientation of the article receiving means
comprises at least first and second proximity indicator means
positioned approximately 90 about the article axis. The
proximity indicator means monitor distances related to the
distance between the article receiving means and the base.
Each proximity indicator means produces a usable output related
to the distance it detects.
The means for applying the force desirably comprises
first and second force coils positioned substantially 90 apart
about the axis of the article and responsive to signals related
to the proximity indicator means outputs to bring the article
receiving means into a neutral position. The forces exerted by
the f~rst and second coils can be resolved to a single force at
, '
-- 2
..s
, " , .
a single location about the axis. This force represents the
amount of imbalance.
The invention of the present divisional provides,
in one aspect a force coil comprising a housing, first and
second magnets for generating a magnetic field within the housing,
the housing including an inner wall defining the interior thereof
and means for providing return paths for the magnetic flux, at
least one turn of an electrical conductor for conducting current,
the first and second magnets being aligned generally axially of
one another, one pair of like poles of the first and second
magnets being directed oppositely in the.housing and.the other pair
of like poles facing one another to provide a substantially
uniform magnetic field which extends radially of the alignment
axis of the magnets and which extends to the return path
providing means of the housing, and means for conducting flux
disposed between and in contact with the facing poles of the mag-
nets, the return path providing means of the housing including a
flux conducting region of the inner wall which projects radially
inwardly into closely spaced apart relation with the flux
20 conducting means to provide an air gap therebetween through which
flux is directed radially, means for supporting the conductor in
the air gap for movement therein in response to force due to
current flow in the conductor, the housing inner wall comprising a
generally cylindrical portion disposed about an axis, the flux
conducting means comprising a generally cylindrical spacer which
is generally coaxial with the housing inner wall, the spacer
including an end wall in contact with each of the facing like
poles of the first and second magnets, the spacer end walls and
facing like poles being generally equal in size and shape, the
flux conducting region of the inner wall comprising an axially
and radially inwardly extending annular portion of the inner wall.
76
In an embodiment of the in~ention, the means for
indicating the amount and location of i~balance comprises a
visual display p2nel coupled to the proximity indicator means~
The proximity indicator means outputs are coupled to the panel
to produce a visual display of the location and amount of the
applied force necessary to bring the article receiving means to
the neutral position. A workman can use this displayed informa-
tion to correct the imbalance existing in the article.
The above-noted parent application also
,,
describes and claims a method for determIning the
amount and location of imbalance in an article which
is to be balanced about an axis. The method
: . .
- comprises the steps of placing the article on an article receiving
means with the axis in a predetermin~d orientation with respect
to the article receiving means, applying a force to the articIe
; receiving means to bring it into a neutral orientation and
; determining the amount and position of the forca applied.
In one embodiment of the method, at least first and
second-proximity indicator means measure the spacing between a
surface of the article receiving means and a surface of a base
upon which it is substantially freely movably supported.
Desirably, the proximity indicator means comprises four proxlmity
- transducers positioned at 90 intervals about the axis to measure
the spacing. The proximity transducers produce outputs repre-
sentative of the spacing. The step of applying the force at a
: .
predetermined location to the article receiving means desirably
` is performed by a pair of force coils positioned 90 apart about
'`.'.'''
:.
, ':.
- 3a -
''
.
:
7 ~
the article axis between the article receiving means and the base.
The force coils are responsi~e to signals related to the p~oximity
transducer outputs to equalize those outputs, thereby bringing
the article receiving means, and the article thereon, into a
neutral orientation.
In the illustrated embodiment, the article receiving
means comprises a table and a locating fixture, i.e., a spindle
and chuc~. The proximity indicator means outputs are in analog
form, i.e., each of four proximity transducers produces an output
related by a predetermined known function to the table-base
spacing which it monitors. These analog outputs are processed
to generate first and second force coil input signals which are
provided to the first and second force coils, respectively. These
signals cause the force coils to bring the table-article system
; to neutral orientation. The proximity transducer output signals
are also processed and converted to digital information by an
analog-to-digital converter ultimately to provide a visual indi-
cation of the amount and location of imbalance in the table-
article system.
In an illustrative embodiment, the means for processing
the digital outputs of the analog-to-digital converter comprises
a digital computer. The means for producing the first and second
force coil input signals comprises electronic circuitry.
Generally, static balancing machines measure forces
generated along two perpendicular axes, generally referred to as
X- and Y- axes or 0 and 90 axes, respectively. Such forces are
generated by im~alance existing in the system comprising the
article to be balanced and the apparatus, usually a table,
'
-- 4 --
.
including a locating chuck and/ox spindle, upon which the axticle
is positioned. The ~orces arising due to the i~balance along
these two perpendicular axes are resolved to a single force at a
single location about the axis of the article. This force and
location are interpreted by a workman or by automatic machinery .
either to add or remove material at the location of imbalance.
Material is added or removed as necessary to correct the amount
of imbalance.
The term "amount and location of imbalance" as used
. .
hereinafter refers to an amount of material, usually expressed
as either a mass or a weight of material, which can be added to
an articIe at a particular location about an axis of the article
to balance it about that axis. It is to be understood, however,
that the instant invention is equally useful in situations in
which material is to be removed from an article to correct
imbalance.
- The invention may best be understood by referring to
; the following description and accompanying drawings which
~ illustrate the invention. In the drawings:
-A 20 Fig. 1 is a perspective view of an apparatus constructed
, . .
' in accordance with the present invention;
.
`- Fig. 2 is a side elevational view of a portion of that
; apparatus;
~,.;
~ig. 3 is a top plan view of the apparatus of Fig. 2;
Figs. 4a-b are partial sectional ~ertical elevational
- .
- and fragmentary perspective views, respectiYely, of an alterna-
~ . . .
tive detail of a portion of the apparatus of Figs. 2-3;
Fig. 4c is a partial sectional Yertical elevational
. .
~ 5 ~
.~
view of an alternative detail of a poxtion of the app ratus of
Figs. 2-3;
Fig. 5 is a sectional perspective view of a detail of
the apparatus of Figs. 1-3;
Figs. 6a-b are flow diagrams illustrating various steps
in the processing of outputs from the apparatus of the figures
to produce signals for use by that apparatus; and
Fig. 7 is a partly block and partly schematic circuit
diagram of a part of the apparatus of Fig. 1 useful for perform-
ing certain of the process steps listed in Fig. 6.
Fig. 1 illustrates a station 8 at which is determined
the amount and location of imbalance of a part to be balanced
~ about an axis. The station includes a table 10 movably mounted
-~ upon a base 12. me illustrated table lO is freely pivotally
mounted upon base 12 by a conventional pivot mechanism. Table
10 is generally flat and circular so that its center o mass
lies as close as possible to its geometric center. Two types
of pivots will be discussed hereinafter in connection with Figs.
4a-c. The base 12 rests upon a floor 14 on four corner leveling 2Q legs 16 which are adjustable ~o that the top surface 17 of table
lO is perpendicular to the direction of gravitational attraction
- at the center of the table.
The center of table lO is e~uipped with an orienting
spindle 18 and chuck 20 for positioning and holding an article
:;
to be balanced. In the illustrated embodiment, the articles to
be balanced are wheel-and-tire assemblies 22 which are to be
balanced about their axes of rotation. Assemblies 22 are con-
veyed along a conveyor 24 to station 8 where they are loaded
7~
onto table 10, one at a time. The amount and location of
imbalance in each assembly 22 is determined by the apparatus at
station 8, and each assembly 22 is marked with some indication
of its amount and location of imbalance, either manually or
automatically. Assemblies 22 are then unloaded from table 10
and proceed along conveyor 24 away from station 8 to a balance
correcting station (not shown). The axis of rotation of
assemblies 22 coincide with the axis 26 of spindle 18 and chuck
20 (see Figs. 2-3).
~- 10 Station 8 includes a data processor or digital computer
28 which is coupled to apparatus on table 10 and processes infor-
mation received regarding the amount and location of imbalance in
each assembly 22 processed thereby. Desirably, computer 28 in-
cludes a visual display panel 30 upon which is displayed infor-
mation regarding the amount and location of imbalance in each
- assembly 22 processed. Typically, the signals which drive the
.~ .
various display elements on panel 30 also control any automatic
marking apparatus 32 which is used to mark, e.g., the location
of imbalance on the processed assemblies 22. Other ~ignals
generated by the computer 28 can be used to control any automatic
equipment ~not shown) for correcting imbalance in the processed
assemblies 22, e.g., adding-or subtracting weight at the
location of imbalance.
; Referring now to Figs. 2-3, table 10 and base 12 are
illustrated in greater detail. Table 10 is supported from base
12 upon a pivot mechanism 34 which allows the table to move
freely with respect to the base. T-he pivot mechanism may be of
any suitable type. Two typical pivotal mechanisms 34 are
- 7 -
7~i
illustrated in Figs. 4a-c. The mechanism 34 of Figs. 4a-b is
typically referred to as a wire pivot. This pivot consists of
an upper carrier 40 which is attached to the underside of table
10 and ~hich has two down~ardly extending fingers 42, 44. At the
ends of fingers 42, 44 are bloc~s 46, 48, respectively. Blocks
46, 48 have adjacent surfaces, respectively, provided with
registering gocves 53. Surfaces 50, 52 can be drawn together
by tightening an adjusting cap screw 54, the head 56 of which
engages a shoulder 58 in bloc~ 46, and the threads 60 of which
engage threads 62 in bloc~ 48.
A lower carrier 70 includes upwardly projecting fingers
` 72, ~4 at the ends of which are formed blocks 76, 78, respec-
tively. Blocks 76, 78 include adjacent surfaces 80, 82 defining
registering grooves 83. Surfaces 80, 82 can be drawn togetAer
by an adjustment screw 84, the head 86 of which engages a shoulder
88 in block 76, and the threads 90 of which engage threads 92 in
block 78. A wire 94 is captured between suraces 50, 52 in
grooves 53 at its lower end and between surfaces 80, 82 in grooves
;; 83 at its upper end. Wire 94 is resilient to present minimum
resistance to movement of table 10 with respect to base 12. Wire
- 94 must be capable of supporting the weight of table 10 and the
assemblies 22 which are placed thereon.
A second pivot mechanism, illustrated in Fig. 4c is
referred to as a spherical hydraulic pivot. Tnis pivot comprises
an upper member 100 having a downwardly and outwardly facing male
spherical surface 102 and a lower member 104 having an upwardly
and inwardly facing female spherical surface 106. Surfaces 102,
106 have equal radii of curyature. A grooYe 108 extends about
., :
-- 8 --
7 ~
surfaces 106 to provide a channel through which a hydraulic fluid
is pumped. Fluid is introduced into groove 108 through bore 110
which extends into and vertically upwardly through lower member
104 from a hydraulic fitting 112. A reservoir 114 is formed by
lower member 104. Reservoir 114 collects hydraulic fluid which
is pumped from groove 108 under pressure by a pump tnot shown)
through hydraulic fitting 112 and bore 110. The hydraulic fluid
flows between surfaces 102, 106, allowing upper member 100 to
move freely with respect to lower member 104.
- 10 Returning to Figs. 2-3, table 10 includes a downwardly
facing surface 120. Base 12 includes upwardly facing surface 122.
Four proximity indicators 124, 126, 128, 130 are mounted on
surface 122 at 90 intervals about the axis 26. Indicators 124-
:
130 in the illustrated embodiment are of a type generally re-
- ferred to as proximity transducers. These transducers are series
- 05OPT proximity transducers manufactured by Schaevitz Engineering,
Pennsauken, New Jersey. Such devices have a substantially linear
.
voltage output for distances of from O to 0.05 inch between the
- transducer and a proximate ferromagnetic object. A cyLindrical
. ,
column 132 of a ferromagnetic material is attached to surface 120
of table 10 in axial alignment with each of transducers 124-130.
Each column 132 projects into close proximity with one of trans-
- ducers 124-130. The output signals of the various transducers
124-130 are-representative of the distance between the downwardly
-~ facing surfaces of coLumns 132 and the active faces of ~espective
transducers 124-130.
The imbalance determining system also includes first
and second force coils 140, 142, respectively. Force coil 140
- 2 -
7~
- is located adjacent proximity transducer 124. Force coil 142 is
located adjacent proximity transducer 126. A line extendtng
through axis 26 and the centers of transducer 124 and force coil
140 w;ll hereinafter be referred to as the 0 axis. A line
extending through axis 26 and the centers of transducer 126 and
force coil 142 will hereinafter be referred to as the 90 axis.
A line extending through axis 26 and the center of proximity
transducer 128 will hereinafter be referred to as the 180 axis.
A line extending through axis 26 and the center of pro~imity
transducer 130 will hereinafter be referred to as the 270 axis.
Force coils 140, 142 are energized in a manner to be
explained subse~uently to produce either attractive or repulsive
forces between surfaces 120, 122 of the table 10 and base 12,
- respectively. The forces produced are independent of one another
:,
and can be resolved into a single force at a selected position
about axis 26 of table 10. Force coils 140, 142 are substantially
identical. Fig. 5 is a detailed drawing of an illustrative
construction of force coil 140.
Referring to Fig. 5, force coil 140 includes a generally
cylindrical housing 144 and an open bobbin frame 146 sized to be
loosely received in housing 144. Housing 144 is closed at its
~- upper end by an upper closure cap 148 and at its lower end by a
lower closure cap 150. Upper cap 148 includes two downwardly
projecting bosses 152, 154. Boss 152 forms a downwardly extend-
ing positioning ridge about the periphery o~ upper cap 148. Boss
154 forms a pedestal which is coaxial with upper cap 148 and
extends downwardly from the bott~m surface 156 thereof. Lower
cap 150 includes two upwardly projecting bosses 158, 160. Boss
.
158 forms a positioning ridge about the periphery of cap 150.
Boss 160 forms a pedestal which is coaxial with cap 150. Cap
150 also includes a pair of chordally extending slots 162.
Slots 162 loosely receive a pair of legs 166 of bobbin frame 146.
Legs 166 project upwardly from an attachment plate 168 at the
bottom of bobbin frame 146. A bobbin 1~0 is supported from the
tops of Iegs 166. Bobbin 170 includes a peripherally extending
outwardly opening groove 172 into which are wound several turns
174 of conducting ~ire. In the illustrated embodiment, there
are 100 turns of No. 31 copper wire. The leads 176, 178 of turns
174 are secured in grooves 180 on the external surfaces of legs
166, e.g , by gluing. A peripherally and radially inwardly
extending ring 182 is formed about the interior side wall of
~ . ,,
, housing 144. Ring 182 insures that the magnetic field will be
uniform and strong vertically above and below turns 174 for a
substantial distance. Thus, when the housing 144 moves relative
to the bobbin frame 146, the strength of the field within which
turns 174 are immersed remains substantially constant.
A pair of solid cylindricaI magnets l90, 192 are
situated on bosses 154, 160, respectively, of the upper and
lower caps 148 r 150. Magnets 190, 192 are positioned so that
their north poles face one another. A spacer 194 is placed
between the facing north poles of magnets l90, 192 during
assembly of the force coil. Spacer 194 is made of a ferromag-
netic material. ~agnets 190, 192 and spacer 194 thus establish
a quite strong radially out~ardly directed magnetic field through-
out the cylindrical spacs 196 between the sidewall 198 of spacer
194 and the cylindrical wall of ring 132. Turns 174 are immersed
:,
: - 11 -
47~
in this field. Thus, current flow through turns 174 causes a
force between turns 174 (and bobbin frame 146 upon which they
are wound) and housing 144. It will be appreciated that when
housing 144 is attached to surface 120 by one or more attach-
ment screws 200, and attachment plate 168 is attached to sur-
face 122 by one or more screws 202, forces of attraction and
repulsion can be generated between table 10 and base 12 by
energizing the turns of force coils 140, 142.
-~ In the illustrated embodiment, bobbin frame 146 is
constructed from aluminum. Housing 144 and caps 148, 150 are
made of a ferromagnetic material to provide flux return paths
`~ for magnets 190, 192. Magnets 190, 192 are produced from a
proprietary cobalt-samarium a}loy with the trade mark ~ICOREX,
available from Hitachi Magnetics Corp., Edmore, Michigan.
; Referring now to Fig. 7, an electronic circuit for
use with the apparatus of Figs. 1-3 will be explained. It is
to be understood that the circuit of Fig. 7 is intended for use
with one pair of diametrically opposed proximity transducers,
e.g., the pair 124, 128 or the pair 126, 130, and one of the
force coils, e.g., 140 or 142, respectively. However, identical
circuitry can be used to drive the remaining force coil in
; response to signals generated by the remaining pair of proximity
- transducers.
In the illustrated embodiment, proximity transducers
124, }28 are coupled to a s~gnal processing circuit 220 illus-
trated in block form. Circuit 220 in the illustrated embodiment
is a SMS/GPM/LVDT signal conditioning module manufactured by
Schaevitz Engineering.
.~
- 12 -
~ 147
:
- Each of proximity transducers 124, 128 includes an
exciting coil 124e, 128e, respecti-~ely. Coils 124e, 128e are
coupled in paralleI between terminals K, X of module 220 and
!~ are excited by module 22Q. Each of transducers 124, 128 also
:
` includes an output coil 124x, 128x, respectively. Coil 124x
is coupled across terminals P and N of module 22~. Coil 128x
~- is coupled across terminals R and S of module 220. Coils 124x
and 128x produce voltages across terminals PN and RS, respec-
~tively, which are representative of the proximity of transducers
124, 128 to their respective columns 132 ~see Figs. 2-3).
.
Module 220 operates in the following manner: If the
potential across terminals PN exceeds that across terminals RS,
- the potential on the output terminal, T, is negative. If the
potential across terminals RS exceeds the potential across
terminals PN, the potential on terminal T is positive. If the
potentials across terminals PN and RS are equal, the potential
. .
on terminal T is zero.
Terminal T of module 220 is coupled to a rate feedback
networ~ 224 comprising a resistor 226, a capacitor 228 and a- 20 potentiometer 230. This circuit is coupled to an input terminal,
pin 2, and the output terminal, pin 6, of an operational ampli-
fier 232. The feedback network 224 varies the amount of feedback
at pin 2 according to the rate of change in the potential at
terminal T. If the potential at terminal T is unc~anging, or
the rate of change in that potential is very small, little or no
feedback appears at pin 2 of amplifier 232. As the rate of change
of potential on terminal T increases, the amount of feedback due
to network 224 increases. This prevents excessively high rates
~"
- 13 -
'
of change in the current through the tuxns 174 of force coil 140.
Limiting of the rate of change of current through tuxns 174
reduces the possibility of large excursion, high-speed movement
by table 10. Such movement is to be avoided to prevent un-
~ desirable oscillation by the table as it seeks to bring a parti- .
-j cular ass~hly 22 to a neutral arientation.
Operational amplifier 232 in the illustrated embodiment
is an LM 301 operational amplifier. In the description of Fig.
7, unless otherwise specified, all operational amplifiers are of
; 10 this type.
- A zero control potentiometer 234 is also coupled to
- pin 2 of amplifier 232. Potentiometer 234 is coupled between
+V and -V voltage supplies to supply a selected offset potential
to pin 2. Potentiometer 234 can be used to bring table 10 to a
neutral orientation before any assemblies 22 to be balanced are
placed thereon.
A gain control circuit 236 comprising a resistor 238
and a potentiometer 240 is coupled between the output terminal,
.
pin 6, of amplifier 232 and pin 2 thereof. Circuit 236 provides
for adjustment of the gain of amplifier 232. Desirably, this
gain will be adjusted as high as possible, ~ut low enough to
prevent undesirable oscillation of table 10.
A rate control circuit 242 is coupied ~etween pins 2
and 6 of amplifier 232. Rate control cixcuit 242 comprises a
potentiometer 244 and a capacitor 246. Circuit 242 provides for
adjustment of the upper corner frequency of amplifiex 232. Such
upper corner frequency can be adjusted to prevent high frequency
oscillation of table 10.
",.'
-
.
!
, .
47~
"'
A gain control circuit 250 is coupled between pin 6
of amplifier 232 and the input terminal, pin 2, of an opera-
tional amplifier 252. Circuit 250 includes a potentiometer 254,
a capacitor 256, and a resistor 258.
,:
A "slow zeroing'~ operational amplifier 260 is also
-:: coupled between the output terminal of amplifier 232 and the
input terminal of amplifier 252. Signals are supplied to input
pin 3 of amplifier 260 through an ~C network including a resistor
262 and a capacitor 264. Feedback is provided for amplifier 260
by a potentiometer 266 and a resistor 268 which are coupled
. between the output terminal, pin 6, and an input terminal, pin
.
- 2, of amplifier 260. Amplifier 260 helps to prevent excessively
rapid oscillation of table 10 by generating at pin 2 of amplifier
.,
252 a very slowly changing correcting signal for large-magnitude
deviations of the orientation of table 10 from its neutral
orientation.
Pin 6 of amplifier 252 is coupled through a res.istor
270 to a biasing network 272. ~etwork 272 includes two oppositely
poled diodes 274 r 276, which couple network 272 to a driver cir-
cuit 278 for turns 174 of force coil 140. Circuit 278 includes
complementary symmetry predriver transistors 280, 282. The
circuits of transistors 280, 282 include clamping diodes 284,
286, respectively, coupled between the emitters of transistors
280, 282 and ground. The output circuit of transistors 280
includes a frequency limiting capacitor 288. The output circuit
. .
of transistor 282 includes a frequency limiting capacitor 290.
Drive circuit 278 further includes complementary
symmetry driver transistors 292, 294. The base of transistor
.... 30
- 15 -
7~
292 is coupled to the collector of transistor 280. The base
of transistor 294 is coupled to the collector of transistor 282.
The emitter of transistors 292, 294 are coupled to the ~V and -V
voltage supplies, respectively. The collectors of transistors
292, 294 are coupled to the bases of two complementary symmetry
output transistors 296, 298, respectively. The emitters of
transistors 296, 298 are direct coupled to lead 176 of force
coil 140. Transistor 296 thus provides current flow in a first
.
direction through turns 174, while transistor 298 provides
current flow in the opposite direction therethrough. Lead 178
is coupled through a feedback resistor 300 to ground.
Voltage feedback from force coil 140 is provided by a
parallel RC circuit 302 coupled between lead 176 and input pin 2
of amplifier 252. Current flowing through turns 174 is directly
related to the force exerted by force coil 140. Substantially
all of this current also flows through resistor 300. A voltage
signal directly related to force exerted by coil 140 is thereby
produced. To process this signal, a current feedbac~ circuit
304 is coupled between lead 178 and pin 2 of amplifier 252.
Current feedbac~ circuit 304 includes a potentiometer
306 and an input resistor 308 coupled between lead 178 and an
input terminal 2 of an amplifier 310. A feedback resistor 312
is coupled between the output terminal, pin 6, of amplifier-310
and input pin 2 thereof~ A resistor 3-14 is coupled between pin
6 of amplifier 310 and pin 2 of ~mplifier 252. Circuit 304
provides at pin 2 of amplifier 252 a Yoltage signal which is
proportional to the force exerted by force coil 140 between
table 10 and base 12.
. .
- 16 -
, . .
:-
7~
An additional force coil driving circuit 320 is
provided in parallel with biasing network 272 and driving
circuit 278. Circuit 320 includes an operational amplifier
322 and a series resistor 324. Amplifier 322 is connected
in noninverting configuration, with an input terminal, pin 2,
and output terminal, pin 6, thereof connected together. The
noninverting input terminal, pin 3, of amplifier 322 is coupled
to pin 6 of amplifier 252. Pin 6 of amplifier 322 is coupled
through resistor 324 to lead 176. Driving circuit 320 is
provided to prevent the "backlash" or dead space which is fre-
quently encountered in systems such as the instant one. Such
bac~lash results when an error voltage, such as the voltage at
pin 6 of amplifier 252, is too small to produce an error-
correcting current flow in turns 174. In the particular system
illustrated in Fig. 7, error voltages at pin 6 of amplifier 252
which are not in excess of the bias offset voltages.produced by
biasing network 272 and driving circuit 278 will result in no
output current flow from one of transistors 296, 2~8. In the
absence of driving circuit 320, no current would flow in turns
174. Thus, such small error voltages at pin 6 of amplifier 252
would mean that no force would be generated by coil 140 between
table 10 and base 12. It would be virtually impossible for table
10 to be driven to its neutral orientation with no current flow
through force coil 14Ø Circuit 320 alleviates this problem,
amplifying the very small error voltages at pin 6 of amplifier
252 to provide a small correcting current flow through resistor
324 and turns 174, allowing force coil 140 to drive table 10 to
its neutral orientation from small error orientations.
'''
- 17 -
In order to produce signals related to the amount of
force exerted by force coils 140, 142 between ta~le 10 and the
base 12 and to use these force-related signals to calculate a
single balance-correcting force at a single location about each
article 22 to be processed, interface circuit 330 of Fig. 7 is
provided. The orientation error correcting voltage signal at
pin 6 o amplifier 232 is coupled through an input resistor 332
to a gain adjust operational amplifier 334. A gain adjust
potentiometer 336 is coupled between pin 2 of amplifier 334 and
the output terminal, pin 6, thereof. Interface circuit 330 also
includes a two-pole filter circuit 340 for rem~ving from the
error correcting signal appearing at pin 6 of amplifier 232 all
variations having frequencies above a predetermined cutoff. In
the illustrated embodiment, the cutoff frequency is 1 Hertz (Hz).
Circuit 340 includes three serially coupled operational
- amplifiers 342, 344, 346. Pin 6 of amplifier 334 is coupled to
the input terminal, pin 2, of amplifier 342 through a resistor
348. Feedback is provided from the output terminal, pin 6, of
amplifier 342 through a parallel RC circuit 350 to pin 2 thereof.
Pin 6 of amplifier 342 is coupled througn a series resistor 352
to the input terminal, pin 2, of amplifier 344. Feedback is
provided between the output termina}, pin 6, of amplifier 344
and pin 2 thereof through a resistor 354. Pin 6 of amplifier 344
is coupled to an input terminal, pin 2, of amplifier 346 tnrough
a resistor 356. A feedback capacitor 358 is coupled to the out-
put terminal,- pin 6, of amplifier 346 and pin 2 there~f. A
feedback resistor 360 is coupled between pin 6 of amplifier 346
i and pin 2 of amplifier 342. Very low fre~uency tl Hz m æ imNm~
..
- 18 -
7~
~ signals related to the force generated by force coil 140 thus
appear at a terminal 362 which is coupled to pin 6 of amplifier
;- 346.
~ Circuitry identical to that hereinbefore described is
.. coupled between proximity indicators 126, 130 and force coil 142.
~: Such circuitry (not shown) produces on a terminal 364 very lowfrequency signals related to the force exerted by force coil 142.
: Terminals 362, 364 are terminals of two high speed read
relays 366, 368, respectively. Terminal 362 is one terminal of
a switch 370 of relay 366... Terminal 364 is one terminal of a
switch 372 o relay 368. The remaining terminals of switches 370, -
372 are coupled together and to an input line 373 of an analog-
. to-digital (A/D) converter 374~ The actuating coils 376, 378
: of.relays 366, 368, respectively, are coupled between +V voltage
.
- supply and two input lines 380, 382, respectively, of a flip-1Op
384. A flip-flop control signal is coupled to an input line 386
.~. of flip-flop 384 from a clock circuit in the digital computer 28
of Fig. 1. The cloc~ circuit generates on line 386 a signal
. which causes current to flow alternately in coils 376, 378. Such
.: 20 current closes.switches 370, 3~2 alternately, effecting time-
division multiplexing of AfD converter 374. The output lines
........... 388 of A/D converter 374 alternately carry digital signals repre-
. senting the forces generated in force coils 140 and 142. The
. computer, being the source of the multiplexing control s~gnals
~ on }ine 386, detexmines whether the signals it is receiving from
-. A/D output lines 388 are representative of the force generated
by coil 140 or, alternativeIy, the force generated by coil 142.
Tuxning n~w to F~g. 6, the ~anner in whIch the c~mpute~
~ 7 ~
utilizes the digital signals present on lines 388 to calculate
the amount and location of imbalance will be explained.
In the computer program illustrated in flow chart form
in Fig. 6, the orce generated by force coil 140 is referred to
as the "A" force and the force generated by force coil 142 as
the "B" force. The computer program starts at block 400 of Fig.
6. The program proceeds to block 402 to cause the computer 28
of Fig. 1 to check the position of a "servo calibrate" switch on
the computer 28 front panel. This switch is a multi-position
switch. In a first position, this switch disables flip-flop 384
of Fig. 7 to cause digital signals representative of the ~orce
;; generated by coil 140 only to appear on output line 388 of A/D
- converter 374. In a second position, this switch causes the
output lines 388 of A/D converter 374 to carry digital signals
representative of the force generated by coil 142 only. In a
; third position, this switch places the system in its normal
operating mode in which the analog signals representative of
the forces generated by coils 140, 142 are multiplexed onto the
input line 373 of A/D converter 374.
- 20 When the "servo calibrate" switch is in its third
position, the program next proceeds to block 404, causing the
computer to check the position of a press-to-actuate "check zero"
button. If the "check zero" button is depressed, the program
causes the multiplexed forces generated by coils 140, 142 to be
,. .
stored in the computer. See block 406. The "check zero" button
will not be actuated by an operator unless the table 10 (Figs.
1-3) is empty. The purpose of this portion of the program is to
store any inherent imbalance existing in table 10 and its
.
- 2~ -
.
47~
..
associated components prior to processing of axticles 22.
Typically, the "check zero" button will be depressed in-
frequently, e.g., after every tenth assembly 22 is processed.
For the several assemblies wAich are processed between operations
of the "check zero" button, the error in the balance table of
table 10 which will be used by the computer program for further
calculations will be the last error stored in the operations in-
dicated in blocks 404, 406.
When the "servo calibrate" switch (block 402) is in
either the first or third position, the digital signal repre-
sentative of the force being generated by force coil 140 is
- input for further use by the computer as indicated by block 408.
The computer program is controlled as indicated in blocks 410,
412 by an operator-controlled "auto zero" switch which allows
the operator automatically to zero tbring to neutral orientation)
table 10. With this switch in the "auto zero" mode, the computer
utilizes the error stored by the operation discussed in connection
with blocks 404, 406. This error is subtracted fr~m the input
.j .
from force coil 140 after an article 22 has been placed on
table 10 for processing. The computer produces a signal for
display related to the force necessary to correct the article 22
imbalance only. With the ~auto zero" switch in a normal oper-
` ating position, the "raw" digital input related to the force
coil 140 force is processed by the computer without adjustment
for any pre-existing imbalance in table 10. Thus, a digital
output display provided by the computer shows imbalance exist-
ing in table 10, if any, plus any imbalance existing in an
article 22 which is being processed.
- 21 -
L147~i
Next, in block 414, the computer considers the sign
of the digital signal representing the force being exerted
between table 10 and base 12 by force coil 140. Such force
can be either positive (tending to pull table 10 and base 12
toward one another) or negative ttending to push table 10 and
base 12 away from one another~. The computer stores the sign
of this force and uses the magnitude, ¦A¦, of this force for
further processing. The magnitude and the sign of such force
are then displayed on the display panel 30 ~Fig. 1), as indi-
cated by block 416.
If the "servo calibrate" switch tblock 402) is in the
second position, the program causes the computer to accept for
-further calculation digital input information representative of
the force being exerted by coil 142. The steps of this process
will be explained hereinafter.
.,. ~ . . .
The computer next chec~s the position of the "servo
calibrate" switch again. Block 418 illustrates this step. If
the switch is in the first position, the program returns to the
step illustrated in block 408. If the servo calibrate" switch
is in either the second or third position, the computer accepts
from A/~ converter 374 ~Fig. 7) digital information representing
the force being exerted by coil 142. This step is illustrated
by block 420.
The program next causes the computer to check the
position of the "auto zero" switch. See block 422. The computer
checks this switch position to determine whether the operator
wants the computer automatically to zero the digital information
representative of the coil 142 force to null any force component
, .
- 22 -
7~;
attributable to imbalance in table 10, or to present the "raw"
digital data representatiYe of the coil 142 force. Again, the
- "auto zero" mode will cause the program to go to block 424 and
subtract from the raw imbalance data received from force coil
142 any information stored as a result of the operations illus-
trated in blocks 404, 406.
- The computer next checks the polarity of the digital
signal representative of the coil 142 force. See block 426.
As with coil 140, the force exerted by coil 142 can either be
- 10 positive or negative. If the sign of the force is negative,
this sign information is stored in the computer and the force
is inverted to obtain its magnitude, ¦B¦. The magnitude and
~` sign of the digital signal representative of force generated
by coil 142 are then supplied to the display panel 30 (Fig. 1)
of the computer, as indicated in block 428.
The computer next checks the position of the "servo
':'
calibrate" switch again. As block 430 illustrates, if the switch
is in the first position, the computer returns to block 408 to
accept digital signals related to the force exerted by coil 140.
If the switch is in the second position, the computer continues
to accept digital information representative o~ the force exerted
by coil 142 beginning at block 420. If the "ser~o calibrate"
switch is in the rhird, or normal operating position, the
computer proceeds to the next step illustrated in block 432 of
the flow chart.
Now the computer begins the computations necessary to
combine the two forces applied to the table along the 0 and 90
axes by coils 140, 142, respecti~eIy, into a single force at a
- 23 -
,:
7~
single location about the axis. 26 of table 10. The computer
squares the digital signal representing the force being exerted
by coil 140. See block 432. The computer squares the digital
. signal representing the force being exerted by coil 142. See
block 434. The computer then adds t~ese squares together, block
436, and obtains the square root of the sum, block 438. Next
the computer multiplies the square root by a scaling factor G.
- See block 440. Since computer 28 (Fig. 1) in the illustrative
.: - .
.. embodiment does not have floatng point capability, scaling
factor G causes the computer to adjust the range of the square
root calculated in block 440 to provide maximum significance to
.:
. the digits of.the square root for subsequent computation. The
computer next divides the G JA~ ~ B~ product by a scaling factor
.. ~, as illustrated in block 442. In the illustrated embodiment,
; . H equals 32 since the amount of imbalance of any article 22 to
...... .
be processed is to be resolved to 1/32 ounce increments.
. . In blocks 444, 446, the computer converts the square
. ~ root, adjusted by the factors G and ~, to binary coded decimal
. (BCD) information and displays this force on display panel 30.: 20 It must be understood that this displayed forca is representative
of the amount of imbalance existing in the article 22 being
:~ processed.
In block 450, the computer compares the magnitudes of
the forces generated by coils 140 and 142. If the magnitude of
the. force generated by force coil 142 is greater, the computer
divides the square root of the sum of the squares of the forces
;~ generated by both coils into the force generated by coil 140, as
: indicated in block 452. If the force generated by coil 140 is
'
- 24 -
. .
7~i
greater, the computer divides the square root of the sum of the
squares into the force generated by coil 142, as indicated by
block 454. If the magnitudes of the forces generated by coils
140, 142 are e~ual, the computer recognizes that t~e imbalance
, . .
in the assembly 22 being processea lies at one of the following
locations with respect to the zero axis; 45; 135; 225; or,
315.
If either of the forces is greater than the other, the
computer proceeds to the step illustrated in block 456. Stored
in the computer memory are sines of all angles between 0 and 45
in 1 increments, and the angles corresponding to these sines.
The computer locates the angle corresponding to the sine calcu-
lated in either block 452 or bloc~ 454.
- Regardless of the outcome of the decision indicated in
block 450, the program causes the computer to consider the signs
of both forces. The signs are determined from the information
stored in the computer as described in connection with bloc~s 414,
426. Assuming that negative forces exerted by coils 140, 142 are
forces tending to urge the bobbins 146 and housings 144 of the
2Q coils tsee Pig. 5) in opposite directions, and positive forces
generated by coils 140, 142 are those tending to draw their
respective bobbins 146 and housings 144 toward one another, the
signs of the generated forces can be compared to determine in
what quadrant of the circle whose center lies on axis 26 the
imbalance exists. This step is performed as indicated in bloc~
458.
; Next, in bloc~ 460, the information relating to which
of the forces is greater is again considered. This consideration
- 25 -
l4~
allows the computer to determine which half of the quadrant
determined in block 458 contains the location of imhalance.
Once this octant has been determined, the angle determined
in block 456 is used by the computer to calculate the
location of imbalance to within approximately 1~2. As
indicated in block 462~ this location of imbalance, in
angular form, i9 displayed on display panel 30 of computer
28 ~see Fig. 1). The apparatus of Fig. 1 is then ready to
begin its next operating cycle.
..
, 10
:
- 2~ -
