Note: Descriptions are shown in the official language in which they were submitted.
Docket ~o. CIIl-103
,~
SYSTEM FOR CONFIGURING, AUTOMATING AND CONTROLLI~G
THE TEST ~ND REPAIR OF PRINTED CIRCUIT BOARDS
Back~round of the Invention
This invention relates generally to testin~
and/or repairing of printed circuit boards and more
particularly to the configuring, automation and con~rol
of a multiple station test and/or repair facility.
(The term printed circuit board as used herein refers
generically to electrical circuit~ which are
constructed on or within supporting strata. The use of
this term is not intended to limit this invention to
electrical circuits having conductive members formed by
'~ "printing" technique~ nor to circuits formed on or
~ 15 within a supportin~ stratus o~ any specific
: construction).
: In oraer to test such printed circuit boards
(PCBs) it is necessary to establish electrical
connection between the test equipment and selected
- 20 nodes within the electrical circuits of the PCB.
Conventionally this electrical connection is
acccmplished with a ~ixture assembly upon which.the PCB
--2--
i~ positioned. The fixture assembly incorporates a
plurality of conductive probes (traditionally re~erred
to a~ a bed of nails) which are selectivel~ positioned
therein to correspond to predetermined electrical nodes
on the PCB. After the PCB i8 positioned upon the
fixture assembly the probes are biased into electrical
contact with such nodes~ The various probes of the
fixture as~embly are in turn electrically connected to
~timulus and measurement means within the te~t
equipment which stimulate the nodes and which mea~ure
the response~ which occur at the nodes of the PCB under
te~t as a result of such stimulus. A separate fixture
as~embly i~ generally required for each different PCB
to be tested due to the variation of the circuits of
the PCB and of the electrical nodes located therein.
Since the electrical connection between the electrical
nodes of the PCB and the test system is very critical,
since the trend i8 to design PCBs with electrical nodes
which are very densely packed, and since there is such
a variation in the size and configuration of the
di~ferent PCBs, the accurate loading or positioning of
a PCB with respect to the ~ixture assembly and the
accurate loading and positioning of the fixture
assembly as well as the electrical connection of the
fixture assembly to the stimulus and measurement means
of the test equipment have typically been manual
proces3es which make up an exce6~ively large proportion
of the total te~t time. Attempts to automate such
processes through the u~e of robotic arms or univeral
fixtures have been limited due to the required
accuracy, the inherent variablllty in the size, ~hape,
complexity, and fragility of ~CBs, and the associated
C08t and complexity of the robotic arm~ and the
universal fixtures. Once a robotic arm is programmed it
is al~o extremely difficult to alter the configuration
of the operating modules between which the robotic arm
i8 tasked to repetitively perform. Furthermore,
neither the robotic arms nor the universal fixtures
have achieved much ~uccess with reducing the amount of
time required to load and position the fixture
as~embly, load and po~ition the PCB with reRpect to the
fixture assembly, maXe the electrical connection with
the PCB, remove the PCB from the fixture a~sembly, and
prepare the test equlpment for testing the next PCB
(especially if the next PCB has a different
configuration). The~e concerns are compounded when
multiple t86t and/or repair stations are employed in a
test sy~tem.
Summary_of the Invention
The present invention provide~ an apparatus
and a method for quicXly configuring, automating and
controlling the routing, testing and/or repairing of
PCBs which substantially minimizes the aforementioned
: 20 limitations and which can be used to automatically
~elect, transport, and loa~ the fixture assembly or
comparable circuit means required to test or repair a
given PCB; transport and position the PCB to be tested
or repaired with respect to the circuit mean~;
electrically connect the same to the circuitry adapted
to selectively te~t the electrical circuit~ of the PCB;
initiate the testing of the PCB determine in view of
an interpretation of the test results whether the PCB
being tested is accepta~le, defective, or requires
` 30 additlonal testing or repair conduct the additional
tests or remove the PCB from the circuit means and
tran~port the PCB to another area with~n the test or
repair system corresponding to the determined status of
the PCB: determine whether there are other PCBs to be
tested or repaired which will utilize the ~ame circuit
--4--
means presen-tly loaded within the test system, and if
so, sequentially transport and position one of such
PCBs with respect to the circuit means or otherwise
unlo~d the circuit means; and de~ermine whether there
are other PCBs to be tested or repaired which require a
different circuit mean~, and ~f so, select and load the
appropriate circuit means, etc. It should be
emphasized that ~he sy~tem according to the present
invention i~ not ~trictly limited to testing Pcss. As
will become apparent, stages can be advantageously
included within the system to manufacture or repair
PCBs, e.g. to insert or remove part~, to maXe
adjustments to partsl to solder or unsolder parts, to
burn in parts, etc. To emphasize the avallability of
lS the3e alternative operations, this disclosure utilizes
the terminology "test and/or repair" to generically
refer to ~uch operations. For the saXe of brevity and
readability such alternative language has not, however,
been repeated in each and every instance. The failure,
there~ore, to ~pecifically refer to both term~ should
not be viewed as intentionally limiting the reference,
for example, to strictly test operations.
The testing and/or repair of a PCB may
requixe multiple stages and, therefore, the movement of
the PCB from one stage to another. According to the
present invention, reconfigurable modular conveyor
units and automation interfaces are provided to
efficiently transport and accurately load a PCB, for
example, onto a selected test stage prior to the test
and for removing the PCB from the test ~tage after the
te~t. In the preferred embodiment of the system
accordin~ to the present invention, the system further
includes control means to efficiently route the PCB
through the various ~tages required to test and/or
repair a given PCB. In its preferred embodiment the
--5--
present invention further include~ mounting the PCB
within a carrier which is adapted to uniquely position
and restrain the PCB therein. The carriers are
adjustable to accommodate various shapes and sizes of
PCBs and are adapted to facilitate the movement of the
Pcs~ into and out of the various test stages of the
test sy~tem by the automated conveyor means. The
carriers include structure which in cooperation with
other elements of the te~t system facilitate the
precise positioning of the PCB with respect to the
circuit means.
The invention further contemplates: the
adaption of test fixture assemblies and vacuum cover
assemblies (which a~emblies along with other items
useful in PCB test and/or repair, and the PCB carriers
are referred to herein as "components") for transport
by the same conveyor system; control mean~ for
coordinating, in accordance with a dynamic priority
rating system and a preemptive scheduling technique,
the transport of PCBs and associated components; an
automation interface for loadin~, unloading, precisely
positioning, and producing electrical enga~ement of a
PCB and related components at a test stage; and
buffering stage~ capable of holding and storing
multiple PCBs and components. Other significant
features of the invention are: its modularity which
facilitates flexible configuring and extension of the
test system; it~ ability to transport and automatically
load te~t fixtures as well as printed ~ircuit boards
wh~ch signifi~antly enhances the automation of the test
process: and its automated programming capability which
USe8 interactlve graphic~ to lay out a coniguration of
the ~ystem and automatically generates ~he ~ystem
interconnections andtcontrol programs needed to operate
the so configured system.
3L~ g ~
--6--
Description of the Accompanyin~ Drawings
The present invention will be further
descibed hereinafter with reference to the accompanying
drawings wherein: -
Figure 1 is a perspect.ive view of a testing
system according to the present inven~ionr
Figure la illustrates transport of a
component between stations in a test an~/or repair
system of the present invention;
Figure lb depicts alternate configurations of
the cystem of Fig. 1.
Figure 2 is a perspective view of a carrier
according to the present invention;
Figures 3 and 4 are partial sectional views
ta~en along lines 3-3 and 4-4 respectively, of Fi~ure
2:
Figure 5 i.5 a partial end view of a conveyor
unit according to the present invention:
Figure 5a depicts a pre.~erred component drive
mechanism of a conveyor unit
Figure 5b presents four views of a conveyor
unit of the present inventlon;
Figure 5c i8 a partial sectional view taken
along lines 5c-5c of Fig. 5;
- 25 Figure 5d pre6entQ five views of a conveyor
unit having component rotational capabilitie~;
Figure 5e i~ a partial perspective view
illustra~ing a component stop mechanism of a conveyor
unit;
Figure 6 i8 a per~pective view of an
automation interface according to the present
invention;
: Figure 6a ~chematically depicts component
movemen~ in a works~ation of the pre~ent invention:
--7--
Figures 6b, 6c and 6d are perspective views
depicting sequential movement of a component into an
automation interface;
Figure~ 7 and 8 are partial sectional views
taXen along lines 7-7 and 8-8 respectively, of Figure
6;
Figure 7a i8 an exploded view of the driven
roller mechanism of an automation interface;
Figure 7b is a top plan view of said drlven
roller mechanism depicting, in phantom, displacement of
the roller:
:~ Figures 7c and 7d are partial sectional and
perspective views, respectively, depicting drive
mechani~ms for the guide rails of an automation
lS interface;
Figure 9 is a perspective view of a fixture
assem~ly according to the present invention;
Figure 9a presents four views of the fixture
assembly of Fig. 9;
Figure 9b is an exploded view illustrating
the alignment of a stacked arrangement of components
with an automation interface at a test stage:
Figure 9c is an exploded view of the fixture
assembly of Fig. 9;
Figure 10 is a partial sectional view showing
the relative po~itioning and location of the cover
assembly, PCB, carrier, fixture a~sembl~, and circuitry
connecting the same to a te~t stage:
Figure 11 is a perspective Vi2W of a cover
aggembly according to the present invention;
Figure~ lla present a flow chart illustrating
the operatlon o the.automation interface of the
present invention~
Figure llb is an i~ometric view illustrating
a mechani~m for varying the elevation of a cover
--8--
a~sembly in an automation interface,
F'igure llc illustrates the sandwiching of a
PCB between two fixture as~emblie~;
Figure 12 i8 a perspective view of a
buffering stage according to the present invention;
Figure~ 12a-12c are simplified perspective
views illustrating the operation of a buffering stage
in a workstation;
Figure 13 is a partial section view taken
along line A-A of Figure 12 J
Figure 13a is an elevation showing the
relationship between a buffering stage and conveyor
units on either side thereof;
Figure 13b is an exploded view of drive wheel
lS mechanisms of a buffering stage;
Figur0 13c is a top plan view of a buffering
stage;
Figure 13d illustrates examples of a
multi-station test and/or repair system constructed
according to the present invention;
Figure 14 i8 a functional block ~iagram
illustrating the interconnection of the various modules
according to the present invention;
Figure 14a depicts an example of a decision
tree for a type of PCB;
Figure 14b 1~ a block diagram of a typical
control system of t~e pre~ent inventio~;
Figure lS i8 a flow chart illustrating the
various ~teps of a method according to the present
invention: and
Figure 15a is a flowchart depicting a dynamic
priority rating and preemp~ive schedulin~ ~ystem of the
present invention.
.
- 8a -
For convenience of presentation, the following of the
foregoing figures appears out of consecutive order and in the
location stated: Figures 5, 5a and 5c follow Figure 6; Figures
5b(i)-(iv~ follow the sheet comprising Figures 5a and 5c; Figures
5d(i)-(v) follow Figures 5b(i)-(iv); Figure 5e follows Figures
5d(i)-(v); Figures 7a and 7b follow Figure 8; Figure 7c follows
Figures 7a and 7b; and, Figure 7d follows Figure 7c.
.. . . ~
/
/
~, ,
Detailed Description of the Invention
A perspective view illustrating various
functional mod~les of the test system lO according to
the present invention is provided in Figure 1.
Depending upon the complexity and type of circuit or
PCB being tested ~t may be neces~ary to u~e multiple
test stage~. It may also be necessary to use various
combinations of test stages in different sequences to
reliably locate the faults which might be present
~ therein. As can be seen, the ~ystem 10 can include
such multiple test stages, for example, stages lla and
llb~ The tes~ stage~ 11 can be identical, or they can
- be of different types, e.g. functional, in-circuit,
combinational, special device or component testers,
bare board testers, etc. Such testors and the computer
controllers employed to operate them are commercially
available and therefor are not described in detail
herein. Although Figure 1 illustrates two te~t stages
11, the present invention is not intended to be limited
to this number nor to the descriptions provided above.
Conveyor meanR are provided to load and unload such
PCBs and other transportable components from the test
and/or repair stages and to transport such components
there between. The conveyor means of the present
~5 ~nvention employs plug together, modular conveyor units
which are specifically adapted to afford flexibility in
the configuration of the test system lO. Conveyor unit
12 iq a linear conveyor which moves components in two
directions, io e., ~owards as well as away from a ~tage.
It should b~ noted that such forwards and backwards
motion i~ not commonly found in typical automated
production lin~s whl~h generally transport the items
- under process in one^common direction. A more complex
motion is however re~uired to enable conveyor unit 12
~5 to be utilized to transport components both in~o as
--10--
well as out of a test stage 11, as i~ included in the
present invention. To aford an even more flexible
test and rep~ir system, a conveyor unit 13 is provided
incorporating the capability for the rotational
movement of the components in addition to linear
movement. The need for the rotational movement
capability of conveyor units 13 is illustrated in
Figure la where an example transportable component 202
i8 trans~erred from the test station llh to a repair
3tation 15 (only the applicable portion of the system
10 being shown). The figure shows the progress of the
transportable component 202 at intervals of time, not
necessarily equal, as it i8 ~ransported through the
system. In Figure la~iv), a conveyor unit 13 is shown
lS a~ it is rotating 90 degrees counterclockwise in order
to redirect the path of the transportable component
202. In Figure la~vi), another conveyor unit 13 is
shown rotating 90 degrees clockwi~e to alter the path
of the transportable component 202 once again. That
the conveyor units 12 and 13 transport components in
two directions i8 ap~arent in that the transportable
component 202 in Figure la could al50 be transported
from the repair ~tation 15 to the test station llb.
Direct routing paths are in fact available between any
two ~tation~ in a system 10. Various combination~ of
conveyor units 12 and 13 can be utilized to facilitate
multiple arrangement~ of test stages 11 in order to
optimize the te~t ~ystem 10 to the particular user
requiremen~s, Utilizing different types of conveyor
units 12 and 13 not only provides increased efficiency,
but additionally allow~ a user to reduce the cost of
the te~t ~ystem by utilizing the less expensive linear
conveyor 12 for those application~ when rotational
movement is not required. For example, the two
conveyor units 12 labelled 200 and 201 in Figure la(i)
do not need to rotate, and the less expensive conveyor
unit 12 can be used. It al~o allows the user to ea~ily
alter the configuration of the test system 10 as the
requirements change.
Staqes other than test stages 11 can be
included within the test system 10. For example,
- Figure 1 illustrates a repair stage 15, such as a model
404 repair system which i~ commercially available from
the Factron/Schlumberger division of Fairchild Camera
and Instrument Corporation~ This repair stage 15
incorporates a repair and adjustment capability as well
as a testing capability. A PCB can be diverted to
stage 15 when repair (e.g. desoldering and replacement
of components) or adjustments (e.g. repositioning of
knobs or switches within the PCB) are required prior to
completing the testing of the PCB. In addition, manual
probing of the PCB may be accomplished such as may be
required, for example, to trace a fault indicated as
being on a common node having multiple connections
thereto.
Buffering or set-up stages 16 are also
included in system 10 to afford a smooth flow of
components in and out of the variou6 te~t or re~air
stages. The buffering stages 16 can also be utilized
to form a storage Rtation S or PCBs and other
components and in an input/output station 80 for the
test system 10. A~ shown, the input/output station 80
also includes a conveyor module 12 and bar code reader
81 (the operation of which will be described
herelnafter) and facilitates the loading and unloading
o component~ into and out of the tes~ system.
As illustrated in ~i~. 1, as~ociated with
each test stage 11, i8 an automation interface 38 which
facilitateE the load~ng and electrical engagement of
te~t fixture a6semblies and PCBs on the testor. The
~: .
-12~
construction and operation o~ automation interface 38
will be described in detail hereinafter. As shown in
the plan view of Fig. lb(i), each test stage ll,
together with it~ associated automation interface 38
and the immediately preceeding conveyor module 12 and
buffering stage 16 define a test station TS.
Similarly, repair stage 15 and immediately preceeding
conveyor module 12 and buffering stage 16 define a
repair station RS.
Physically interconnecting the repair (RS),
test (TsJ~ input/output (I/0) and storage (S) stations
of test system lO is a transport system, generally
denoted 18, consisting of a suitable arrangement of
plug together conveyor units 12 and 13. The transport
system allows direct routing of PCBs and other
component~ from any station to any other station, in
either direction, under the control of a ~aster control
unit and distributed controllers. To illustrate the
modularity of the system lO, alternate configurations
of the system lO are shown in Figure lb in which all
the same stations are used in three different
configurations. Figure lb~i) shows the configuration
of the system lO in a plan view as a reference. As an
example, it may be desireable to interchange the
locations of storage station S with the test station TSb
in a particular application. This configuration is
shown in Figure lb(ii). As a further example, it may
be desireable to interchange the locations of the
storage station S and the test station TSb, and al~o
the repair station RS and ~he test station TSa. This
configuration is shown in Figure lb(iii). In addition,
each of the pictured configurations can be expanded to
include additional conveyor units, buffering stages
and/or test/repair stages, e.g. at the locations
denoted "X" in Figures lb(i), (ii) and (iii). A
-13-
~ignificantly expanded sy3tem i3 shown in Fig. 13d. In
the preferred embodiment of the invention, the number,
type and physical layout of stations is determined
based on the number of PCBs to be tested and/or
S repaired, the number of test/repair stages required to
process the PCBs and the physical ~pace requirements of
the application. The details of the control system and
the overall operation of te~t system 10 will be more
fully explained after a detailed description of the
- 10 system hardware is presentedO
To facilitate the movement of various PCBs
throughout the te~t system 10 the present invention
utilizes a carrier 20, which is ~est illustrated in
Figure 2. The carrier 20 is designed to be adjustable
to accommodate a wide range of sizes and configurations
of PCBs 19. By utilizing adjustable carriers 20 the
test system 10 can accommodate various ~izes and
configurations of PC8s without requiring manual
intervention or downtime to adjust the remainder of the
test system 10 for a specific size or configuration
PCB. By combining thls feature with other aspects of
the present invention, the test system 10 of the
present invention can concurrentl~ accommodate or
intermix different PCBs 19, e.g. different shape~ and
sizes. This i5 particularly useful and cost effective
for users who do not consistently test a large volume
of the same PCBs 19. The carrier 20 includes a frame
having interconnected members 22a, 22b, 22c an~ 22d
circum~cribing and thereby defining an interior area
sufficiently large to accommodate the PCBs 19 which are
intended to be tested. At least two of the
interconnected members 22a and 22c are generally
parallel to each other ~nd form opposing supports for
tw~ suspension beams'23, The suspension beams 23 are
mounted in a manner which facilitates the lateral
-14~
adjustment of the suspension beams 23 along the members
22a and 22c tSee Fig. 3). The interior surface of each
of the members 22a and 22c has a longitudinal channel
24 therein which i~ adapted to loosely receive the
distal ends of the suspension beams 23. Such distal
end~ include a threaded bore having a corresponding set
~crew 25 engaged therein. Tightening such set screw 25
against one of the ~ide walls orming such channel 24,
whlch side wall i8 extended to overlap such bore,
cause~ the displacement of the beam 23 within the
channel 24 to a position in which it is frictionally
engaged with the opposing side wall of the channel 24.
Loosening such set screw 25 affords the movement of the
suspension beams 23 within the channels 24. The beams
23 can thus be displaced with respect to each other to
~ccommodate varying lengths (or widths) of the PCB.
One or more ~upport bracket~ 26 are adjustably mounted
on each of the beams 23 and adapted to engage the edges
of a PCB l9. The brackets 26 (See Fig. 4) include
~prings 30 which can be biased toward the beams 23 to
frictionally engage the beams 23 but which can also be
pivoted to a position spaced from the beams 23 to
release the brackets 26 from frictional engagement with
the beams 23 and afford the movement of the brackets 26
along the beams 23 in order to adjust for differing
sizes and shapes of PCBs. The edge of the brackets 26
which is proximate to the PCB l9 has a groove 28
therein which is adapted to receive an edge of the PCB
l9. Similarly, the proximate end 31 of the spring 30
is al~o adapted to receive this edge of the PCB l9.
The bracket 26 is designed such that the ab~ence of a
PCB l9 will allow tha spring 30 to pivot toward it~
spaced position thereby releasing the beam 23. The
insertion of a PCB lg between the proximate end 31 of
the spring 30 and the groove 28, however, biases the
-15-
sprin~ 30 to engage the beam 23. The re~ilience of the
3pring 30 also securely captures the PCB 19 bet~een the
grove 28 and the proximate end 31 of the sprin~ 30.
Since a typical PCB may have components mounted at
varying position~ adjacent its outer edge it i8
po~sible with the present invention to adjust the
brackets 26 along the beam 23 to a position which
minimizes interference with these component~. The
parallel ~ide members 22a and 22c also facilitate the
tran~port of the carrier 20 by the conveyor units 12
and 13. For this purpose at least one surface 21 of
the generally parallel side members 22a and 22c of the
carrier 20 is substantially planar and has a
coefficient of friction promoting the transport of the
carrier 20 by the conveyor units 12 or 13. Furthermore
the side members 22a and 22c include a projecting ridge
29, the operation of which, as well as the operation of
the planar surface 21 will become apparent as a result
of the description of the conveyor units ~supra). Each
of the carriers 20 also contain structure to accurately
locate the carrier~ 20 when they are positioned within
any of the various stages 11 or 15 making up the system
10, as will be explained. In the embodiment
illustrated the location holes 27 provide this locating
structure.
As has already been discus~ed the conveyor
units 12 and 13 provide for the transport of
components, e.g. the~carriers 20 and the PCBs therein
between the various ~tageq 11 and 15 of the test ~ystem
10. Although a variety of drive means can be utilized
to transport the components, the illustrated embodiment
(See Fig. 5~ employ~ driven rollers 32 which establish
frictional contact with the extsrior planar surface on
the members 22a or 22c of the carriers 20, or similar
~urface3 provided on other component~ to be
-16-
transported. The rollers 32 are driven in a
conventional manner by a dc motor 34. The driven
rollers 32 are supported on gu~derails 91 on chassis 31
in a conventional manner and positioned adjacent both
the oppo~ing ends of the conveyor units 12 or 13 to
maintain control over a tran~ported carrier 20 as it
pa~ses in either direction into or out of one conveyor
unlt 12 or 13 to an adjacent conveyor unit 12 or 13, or
into and out of a test stage 11 or a repai~ stage 15,
as is applicable. The direction of travel of the
rollers 32 is determined by the polarity of the dc
motor 34. Guide rails 91, also supported upon the
chassis 31 in a conventional manner and located
adjacent opposing sides of the path of the carrier 20,
contain channels 92 which are adapted to receive the
projecting ridges 29 of the carriers 20 or similar
pro~ectlng ridges on other transported components,
thereby guiding as well as supporting the components.
In the illustrated embodiment of the drive
means, (see Figure 5a) a belt 203 and pair of sprockets
204 or similar devices are use~ to transfer the
rotation of the dc motor 34 fro~ a fir~t roller 32,
mounted on the roto~ 33 of motor 34, to a second roller
32. Openings 205 in the guide rail 91 allow the
rollers 32 to contact the exterior planar surfaces of
members 22a or 22c of the carriers 20 when the
pro~ecting ridges 29 of the carrier 20 are captive
inside channels 92. As described hereinafter, the same
drive scheme may be advantageously employed in
buf~erinq stage 16 and automation interface 38.
Figures 5b~i), (ii), (iiij, and (iv) show a
conveyor unit 12 in side, front, and top views and in
perspective, respectively. In Figure 5b(iv), a carrier
20 i Q shown outside of the conveyor unit. This figure
is included to illustrate how a carrier 20 is presented
~2~
to the conveyor unit. The carrier 20 i~ presented to
the conveyor unit 12 by an adjacent conveyor unit 12 or
13, buffering stage 16 or automation interface 38 (not
shown).
Fig~re 5c shows an enlarged section taken
along lines 5c-5c of Figure 5. The frictional contact
between the outer periphery of rollers 32 and the
planar ~urfaces of members 22a or 22c of carrier 20 is
clearly shown. As the rotor 33 of dc motor 34 rotates,
a force is applied to the carrier 20 by rollers 32
causing it to move. As shown in Figure 5c, the carrier
would move into or out of the page. The dc motor 34,
rollers 32, belt 203 and sprockets 204 are supported on
the guide rail 91 in conventional manner (not shown).
It should be noted that although a carrier is pictured,
a fixture as~embly (Figure 9) or a cover assembly
(Figure 11) could be pictured instead of the carrier
and would be moved in identical fashion.
In order to change the direction of transport
by other than 180 degrees, the conveyor unit 13 having
component rotational capabilities, includes a stepper
motor 95 and associated conventional bearing and
gearing mechanisms or their equivalent (not shown), to
rotate the chassis 31, supporting the roller~ 32, about
a vertical axis "V". Figures 5d(i), (ii), (iii), (iv)
and (v) show side, front, top, perspective and ro~ated
top views, respectively, of the conveyor unit 13. In
Figure 5d(v), the chassi~ 31 i~ shown rotated by some
amount. The stepper motor 95 and associated bearing
and gearing mechanisms or their equivalent (not shown)
- are used to effect the pictured rotation~ The stepper
motor is microproce 30r controlled to allow flexibility
in the OrientatiGn of the chassi~ 31 and con~equently
the path along wh~ ch component~ are tran~ported. The
necessity for flexibility in the path for components
-18-
was previougly de~cribed in conjunction with Fig. la.
The 3peed and polarity of the dc motor 34 and therefore
the driven roller~ 32, are microprocessor controlled to
afford controlled, bi-directional motion in the path
along which a given component is transported. Conveyor
unlt 12 and conveyor unit 13 are ~ub~tantially
identical except that conveyor unit 12 exclude~ the
stepper motor 95 and its a sociated mechanism. Each of
the conveyor units 12 and 13 also includes appropriate
sensors (see Fig. 14) to d~termine the presencs and
location of a carrier 20 or other transported component
therein. Similarly sensors also determine the relative
positioning of the chassis 31 of the rotatable conveyor
unit 13. The position of the chassis 31 of conveyor
unit 13 can be sensed, for example, when the chassis is
rotated to 0, 90, 180 and 270 degrees from a reference
orientation. This can be accomplished witn mechanical,
optical, magnetic, capacitive, or other type of
conventional sensors. If the test system 10 requires a
? conveyor unit to stop or hold the transport of a
carrier 20 or other transported component, stop
solenoids are also provided at the entrance and exit
ends of such conveyor units. When actuated the
moveable element of such ~olenoids is interposed within
the path oP the carrier~ 20 to prevent further motionO
Flgure 5e shows a portion of a conveyor unlt 12 or 13
with a stop solenod 206. The moveable element of the
fiolenoid 206 is show~ removed from the path of motion
of ~he fixture as~embly 47 so as to allow its motion,
and in phantom the moveable element i8 shown interposed
in the path of the fixture assembly 47 so a~ to prevent
lts mo~ion. A ixture a~sembly 47 is pictured to
illustrate that the fixture a~qsembly 47 is tran~ported
ln a conveyor unit 12 or 13 in the same manner as a
carrier ~See Fi~ure 5c~. The conveyor unit~ 12 and 13
--19--
are adapted to be utilized in modular for~. For
example, one of the conveyor units 13 can contain a
conveyor controller or microprocessor (See Figure 14)
which can provide control for some predetermined
quantity of slave conveyor units 12. These slave
conveyor units 12 can ~imply be plugged into the master
conveyor controller and such controller can be
programmed with the location and identity of the
additional conveyors 12 and 13 being utiliæed within
the te~t system 10 and plugged therein, thereby
- avoiding the need ~or redundant microprocessors within
each oP the slave conveyor units 12. Each slave
conveyor unit will, however, contain drivers, sensors,
and stop~ to the extent required, such as have been
described above.
The test system of the present invention also
include~ an automation interface 38 (See Fig. 6) which
interfaces the test stages 11 with the conveyor units
12 or 13. Similarly a ~uitably configured automa~ion
interface can be provided to interface repair or other
stage~ to the transport system 18 via a buffering stage
16 tmore fully described hereinafter). The automation
interface 38 of the pre~ent invention i8 adapted to be
utilized with existing testers such as are now
commercially available, The automation interface 38
incorporates mechanisms to transport components by
imparting a force to the side of the component via a
roller ~as is done in the conveyor units 12 and 13) as
well as mechanismq to change the elevation of
components after they are captive in the automation
- interface 38. These two functions are illustrated in
Figure 6a which shows a test worX station. The figure
shows a side view of a conveyor unit 12, buffering
st~ge 16, and an automation lntexface 38 on a test
~tage lla, with movement of a fi~ture assembly 47 shown
-20-
in phantom. The fixture a~sembly 47 is pictured in
four positions: A, B, C and D. The four positions
represent location~ the fixture a~sembly 47 would
occupy at distinct intervals in time aq the work
station is operated. The figure illustrates that an
eRsentially horizontal motion of the component
transports the component between the buffering stage 16
and conveyor unit 12, and between the conveyor unit 12
and the automation interface 38 and that an essentially
vertical motion i8 used to elevate the component
between the operating level of the conveyor unit 12 and
the operating level of the te~t stage lla. The
mechanisms to perform these functions will be discussed
in detail hereinafter. The automation interface 38
contains a driven roller 39 similar to those utilized
by the conveyors 12 and 13. The driven roller 39 is
positioned to engage the exterior planar surfaces of
parallel side member~ 22a and 22c of the carrier3 20 as
the carriers 20 exit from the conveyor units 12 or 13.
In t~is manner the roller 39 can maintain control of
the carriers 20 or other components when the roller~ 32
of the conveyor units 12 or 13 have lost effective
control. This i8 illustrated in Figures 6b, 6c and 6d
where a carrier 20 is shown at distinct intervals of
time during its transportation into an automation
interface 38. Shown are the carrier 20 with a PCs 19,
a conveyor unit 12, and relevant portions of the
automation interface 38 on top of a receiver 56 and
test stage 11. Much of the mechanism of the automation
interface 38 is not shown so that the function of the
roller 39 can be clearly seen. In Figure 6b, a carrier
20 with a PCB 19 is shown entering the conveyor unit
12. The buffering stage 16, in the preferred
embodiment of the invention, i8 adjacent to the
conveyor unit 12 at the end oppo~ite the automation
f9~
-21-
interface 38, a~ ~hown in Fig. 6a. At the interval of
time illustrated by Figure 6b, the motion of the
carrier 20 is being controlled by the conveyor unit 12,
wherein the roller~ 32 are engaging the carrier 20. In
Figure 6c, the carrier 20 is shown in such a position
that the rollers 32 of the conveyor unit 12 no longer
engage the carrier 20 and consequently have lost
effective control. However, roller 39 of the
automation interace i8 now engaging the side of
carrier 20 and exercising control over carrier
movement. At least one roller is always in contact
with carrier 20, 80 that it i8 alway~ un~er control.
In Figure 6d the carrier 20 is shown fully supported
in ide the automation i~terface 38. It is clear from
Figure~ 6a through 6d that although co~ponents must be
elevated between different operatin~ levels, the roller
39 can be fixed at the operating level of the conveyor
unit 12 or 13. It should be noted that although a
carrier 20 and PCB l9 are ~hown in Figures 6b through
6d, a ~ixture assembly 47 or cover assembly 62 could
have been shown. In the preferred embodiment of the
invention, the components are transported to the
automation interface in 5uch a manner as to create the
assembly shown in Figure lO. The driven roller 39 is
driven by dc motor 96, which i9 also under
microprocessor control. DC motor 96 i8 mounted upon a
~upporting column 97 by a bracket 98. See Fig. 7a.
The supporting column 97 is in turn rotatably mounted
upon a bracXet assembly 99 which is a~fixed to the
frame lO0 of the automation interface 38. The frame
100 is an essentially planar, rectangular surface
c~rcum~cribing an es6entially rectangular opening, upon
which the structure and m~chanisms of the auto~ation
in~erace are mountedO Also mounted upon the bracket
as~embly 99 is a solenoid lOl or similar device which
-22-
can be actuated to extend an arm 102 to the position
indicated by 102a (as shown by dashed lines in Fig. 7).
The arm 102 is pivotably connected to the supporting
column 97. Thexefores the extension of the arm 102
will rotate the column 97 through the slotted extension
arm 207. Such rotation of the column 97 will pivot t'ne
dc motor 96 to a position 96a where the driven roller
39 passes through an opening 208 within the rails 44
and frictionally engages the carriers 2G. The rotation
of column 97 and subsequent displacement of the roller
39 to a position labelled 39a is illustrated in the top
view of Figure 7b. The purpose of this displacement
will be discussed later. The opening 208 in the rails
44 is illustrated in Figures 6b through 6d. Similarly
the retraction of the arm 102 will disengage the driven
roller 39 from the carriers 20. The automation
interface 38 further contains elevator means 40 (See
Fig. 7c and 8) which i8 adapted to raise and/or lower
the components with respect to the operating level of
the conveyor unit~ 12 or 13. In the preferred
embodiment the elevator means 40 comprise two spaced
rails 44 which are substantially parallel to each other
which elevate at the same rate such ~hat ~he ~wo ~paced
rail3 44 remain essentially parallel, and that the
plane defined by the two spaced rails 44 remains
essentially horizontal throughout the travel of the
rails 44. Each of the rails 44 contains a longitudinal
channel 45 therein which iQ adapted to receive and
~: support the projecting ridge 29 extending along the
side members 22a and 22c of the carriers 20. This is
illu~trated in Fig~. 6b-6d. The rails 44 are
themselves supported on air cylinders 43 or the liXe
which afford the relatlve movement of the rails 44 from
a ~o~ition supporttng the carrier~ 20 to a po~ition
where the raiis 44 are spaced from the carriers 20.
~'
--23--
This motion is illustrated in Figure 7c. The rails 44
are shown spaced from a fixture assembly 47, and in
phantom in a position where they would be supporting
the fixture assembly ~7. ~he necessity for this motion
is apparent in that in order to deposit a component
and/or elevate the rails 44 to some other operating
level, they must be spaced from the side of the fixture
assembly. The necessity to elevate the rails 44 to
some other operating level arises from the need to
stack components as shown in Figure 10. The air
cylinders 43 are in turn supported by the shuttle
member~ 42 of lead screw mechanisms 41, and the lead
screw mechanisms 41 are driven by a stepper motor 36
which is microprocessor controlled. A drive belt 35 in
combination with various pulley~ are driven by motor 36
,~ and in turn drive lead screw mechanisms 41 located at
each of the opposing corners of the rails 44. Rotation
of the multiple lead screw mechanisms 41 by the motor
36 will raise or lower the rails 44 in a controllable
manner. This is shown ~chematically in Fig. 7d. The
rails 44 can therefore be positioned at the operating
level of the conveyors 12 and 13, and the air cylinders
~3 can be actuated to move the rails 44 to the position
- at which they can support the carriers 20. In addition
solenoid 101 can be actuated to bring the driven roller
39 into frictional en~agement with the carrier 20
through the opening 208 in the rail 44, ~ince the
roller 39 and rails 44 are at the operating level of
the adjacent conveyor unit 12 or 13. The purpose of
30 the solenoid lOl and pivotably mounted brarket S7 shown
in Figures 7 and 7a is now clearly visible. Since the
rails 44 are elevated between the operating level of
the adjacent conveyor unit 12 or 13 and a level which
is below that level ~as seen ln Figure 7c) the roller
39 must be removed from contacting the side of the
-24-
component which is supported inside the rails 44 in
order to raise or lower that component. A carrier 20
can then be transported by the driven rollers 32 of the
conveyor units 12 or 13 and/or the driven rollers 39 of
the automation interface 38 into the channels 45 of the
rails 44 as seen in Figures 6b~6d. Conventional
~echanical stops or solenoids (See Fig. 14) can be
utilized to ensure the carrier 20 is fully transported
and accurately positioned within the rails 44 aa seen
in Fig. 6do The solenoid 101 can then retract the
driven roller 39 and the lead screw mechanisms 41 can
be driven to raise or lower the carrier 20 from the
operatlng level of the conveyor unit 12 or 13 to the
operating level of the test stage 11. The automation
interface also includes appropriate sensors (See Fig.
14) similar to those described with regard to the
- conveyor units to determine the presence of a carrier
20 or other transported component therein. Similarly,
sensors are also available to determine the positioning
of the elevator means 40 and the rails 44.
As has already been stated, PCB testers
traditionally use fixturing assemblies to achieve
electrical contact with the PCBs under test.
Traditionally, these fixture assemblie~ are manually
loaded onto the testerq and the probes therein are
electrically connected to the various stimulus and
measurement means included within the testers. The
PCBs which are to be tested are then individually and
manually placed llpon such a fixture assembly and
accurate~y located with respect to the probes. A
detailed discussion of a fixturing system such as is
de~cribed above i8 ~et forth in U.S. Patent 4,352,061
~hereinafter "061 Patent"). Such a fix-turing assembly
as is taught by the 061 Patent can be adapted
!
r^--
~ . .
-25-
to permit its transport by the conveyor units 12 and
13, buffering stages 16 and automation interface 38.
The fixture assembly is best illustrated in Figures 9,
9a and 10. In Figure 9 the fixture assembly 47 i5
shown in perspective with a portion cut away to expose
the interior structure of the assembly. A complete
picture of the fixtur~ assembly 47 i9 shown in the
bottom, side, front and top views, respectively, of
Figures 9a(i), (ii), (iii) and (iv). In Figure 10, a
partial section is shown of a fixture assembly 47, a
carrier 20 and a cover assembly 62 aq they would be
configured just prior to invoking a test. The stacked
arrangement of component3 is shown restiny on a
receiver 56, which would be supported by the test
station 11. It should be noted that each of the three
components (fixture assembly 47, carrier 20 an~ cover
as3embly 62) would be transported to the automation
interface 38 by the oonveyor units 12 and 13 and
buffering ~tages 16, as required, and that each would
be lowered to an operating level by the elevator means
of the automation interface so as to produce a desired
stac~ed arrangement of components, e.g. a~ shown in
Fi~ure 10. The fixture assemblies 47, adapted
according to the present invention, include a frame 48
having interconnected members 48a-48d circumscribing
and thereby defining an interior are~ sufficiently
large to accommodate the larger PCBs which are intended
to be tested. The frame 48 has at least 2 parallel
side members 48a and 48c adapted to afford the
transport of the fixture assembly 47 by the conveyor
units 12 and 13, buffering stage 16 and automation
interface 38. For this purpose at least one surface on
each of the parallel side members 48a and 48c of the
~ixture assembly ~7 i~ substantially planar and ha~ a
coefficient of friction which will promote the
-26-
transport of the fixture assembly 47 by the conveyor
units 12 or 13, buffering stage 16 and automation
interface 3B. slde member3 48a and 48c also include a
projecting ridge 58 which will be received by the
aforementioned channels 45 of the automation interface
and 92 of the conveyor units. In addition, location
holes 52 similar to those de~cribed within the carrier
20 are also added to facilitate locating the fixture
assembly 47 with respect to the automation interface
38. The means by which the fixture a~sembly 47 is
located with respect to the automation interface 38 is
illustrated in Figure 9b. In the figure, distances a
and b show that the location of the automation
- interface 38 is referenced to receiver 56 via the two
column~ 60. The two col~mns 60 also reference the
fixture assembly 47, carrier assembly 20 and cover
assembly 62 with respect to the receiver 56. Hence,
the fixture assembly 47 is referenced to the automation
interface 38. Supported within the frame 48 is a
non-conductive plate 49. Typically this plate 49 is
manufactured from materials having characteristics
resulting in a plate 49 which is both lightweight and
yet capable of withstanding substantial forces without
signifi~ant deformation. One such material is
manufactured and sold by General Electric Company under
the de~ignation G-10 epoxy glass cloth, although
comparable glass, paper vr cloth laminates are also
useable~ A predetermined array of electrical probes 51
are supported within and pass through thi~ plate 49.
These probes 51 are po~itioned in a pre-arranged
pa~tern to correspond with selected nodes on the PCB 19
which i~ to be tested. ~s has already been described,
this pattern will typically be uniqu-e for each type or
model of PCB 1~ whlch i~ to be tested. The array o~
; 35 probe~ 51 is electrically connected to an array o
-27-
po~ts 53 which are supported within a contact panel
plate 5~. Either individual wiring or a personality or
universal matrix platen concept such as is described in
the 061 Patent could be utilized to provide this
electrical connection. The contact panel plate 54 is
also constructed from materials similar to those
de~cribed with respect to plate 49, and is similarly
affixed to the frame 48. Spacers 55 between plate 54
and plate 49 ensure that plates 54 and 49 remain
substantially parallel and rigid. A view of a fixture
assembly 47 is shown in Figure 9c. All of the
componen~s of the assembly are shown except the means
of electrical connection between the array of probes 51
and the array of posts 53. The test stages ll contain
a receiver 56 also having an array of probes 57 which
correspond to the array of posts 53. The probes 57 of
the receiver 56 are electrically connected to the
various stimulus and measurement mean3 included within
the test stages ll. Actuating means (not shown) such
as the vacuum described in the 061 Patent are utilized
to draw the posts 53 of the contact panel 54 toward the
probes 57 of the receiver 56 thereby establishing
electrical connection between the probes 57 of the
'receiver 56 and the posts 53 o~ the fixture assembly
: 25 47. Alternative actuating means (not shown) such as
electronic motors, hydraulic cylinders, lever
mechanisms, camming mechanism~, etc. can also be
utilized to establi~h this electrical connection.
In a manner already described with regard to
the carriers 20, the fixture assemblies 47 can be
transp,orted by the conveyor units 12 and 13 to a
desired automation interface 38 of the required stage
ll.' Slmilarly the elevator means 40 o~ the automation
in~erface 38 can be u~ed ~o ral~e or lower tha ~ixture
as~embly 47 from the operating level of the conveyor
-28-
units 12 and 13 to a level where the contact panel
plate 54 can be brought into preliminary engagement
with the receiver 56. Simultaneously the air cylinders
43 supporting the rails 44 can be actuated to cause the
rails 44 to release the projecting side members 58 of
the fixture a~semblie3 47. The exact po~itioning of
the fixture assemblies 47 with respect to the stages 11
is achieved by including guide columns 60 within the
test stages lla and llb~ These guide columns 60
cooperate with the location of the location holes 52
within the fixture assembly 47 to accurately position
the fixture assemblies 47. See Fig. 9b. Pref~rably
either the column~ 60 or the holes 52 are tapered to
ensure engagement.
- 15 A PCB 19, positioned within a carrier 20, can
then be transported to the test stage in a manner
already described and po~itioned upon the fixture
a~sembly 47. See Figures 9b and 10. Once the PCB is
loaded upon a fixture assembly 47, the test systems'
actuating means, e.g. a vacuum, is utilized to draw the
PCB 19 and the fixture assembly 47 toward the receiver
56 and thereby compress the spring loaded probes of the
fixture assembly 47 to ensure electrical connection
between the fixture assembly 47 and the conductive
paths of the PCB 19. To ensure that a vacuum can be
drawn between the PCB 19 and the fixture assembly 47
the preferred embodiment of the present invention
utilizes a cover assembly 62 (See Fig. 11) which
includes a frame 64 having interconnected member~
64a-64d circumscribing and thereby defining an interior
area sufficiently large to accommodate the PCBs 19
which are intended to be tested. At 1 ast two of the
interconnected members 64~ and 64c have at lea~t one
surface whlch i8 ~ub~antially planar and which ha~ a
coefficient of friction promoting the tran~port of the
.
9LR~O~
--29--
cover member 62 by the conveyor units 12 and 13,
buffering stage 16 and automation interface 38. Side
member~ 64a and 64c also include a projecting ridge 63.
Fr~me members 64b and 64d contain location holes 61
analogou~ to those o the carrier and fixture assembly.
Suspended across within the frame 63 is a sealing
diaphragm 65 which is constructed of a resilient
mater~al Ruch as that utilized in the construction of
~urgical gloves. This diaphragm 65 is supported within
the frame 64 in a manner restricting passage of the
atmosphere between the frame 64 and the diaphragm 65.
The diaphragm 65 is adequately elastomeric such that it
is capable of envelopln~ the Pcs 19 in a manner similar
to ~hrink wrap parkaging, without damaging the
components on the PCB or without itself becoming
damaged. The use of such ~over assembly 62 affords the
elimination of custom edge seals and ~asket~ which are
required by conventional fixtures. In a manner already
discus~ed, the cover assembly 62 can be transported by
the conveyor units 12 and 13, buffering 3tage 16 and
automation interface 38 to an appropriate stage lla or
llb.
Since the cover assemblies 62 are not PC~
specific and will function with practically all PCBs 19
and ~ixture assemblie~ 47, the automation interface 38
of the present invention is adap~ed to include separate
lifting means (See Figure~ 6 and llb) for the cover
a~emblies 62~ These lifting means include a lifting
carriage 85 having interconnected members forming a
substantially U-~haped configuration. Two legs of the
carriage 85 are sub3tantially parallel and have
solenoid3 87 (or comparable deviceq) mounted thereon
which are adapted to engage cover suppor~ holes-88
within the frame 64 of the cover a~embly 62. The
lifting carriage 85 i~ ~upported on an air cylinder 89
~'~7~
-30-
and slide mechanism 90. Once a cover assembly 62 has
been transported to a given automation interface 3B,
the carriage 85 can be lowered over the cover assembly
62, and the solenoids 87 can be actuated to engage the
support holes 88, thereby locking the cover assembly 62
to the carriaye 85. The air cylinder 89 can then be
actuated to lift the cover assembly 62 within the slide
90 above the operating level of the conveyor unit,
thereby clearing the path for the tran~port of
additional carriers 20 and fixture assemblies 47
without requiring the unloading and transport of a
cover assembly 62 for every chanye in carriers 20 or
fixture assemblies 47O Obviating the need to
repetitively transport a cover assembly 62 between an
automation interface 38 and a buffering stage 16
further reduces the time taken by the system 10 of the
present invention to test and/or repair PCBs. To
facilitate the vacuum seal, either the frame 64 of the
cover asseJnbly 62 or the frame 22 of the carrier 20
- 20 contains appropriate gasket means to create a seal
between the carrier 20 and the cover assembly 62 when
the two are placed adjacent to one another. In the
preferred embodiment the diaphragm 65 i8 extended over
the frame 64 to provide such gasket means. Similarly
either the frame 22 of the carrier 20 or the frame 48
of the fixture assembly 47 will also contain
appropriate gasket means 67 to ~eal the space there
between when the carrier 20 (and PCB 19) i8 placed upon
the ixture assembly 47. Likewise gasket means 68 are
interposed between the fixture assembly 47 and the
contacting surface of the te~t stage 11. It is
apparent to one s~illed in the art that such gasket
means in cooperation with the cover as~embly 62 will
facilitate a ~ealed chamber which can be evacuated ln
order to draw the diaphrag~ 65 downward in re~ponse to
-31-
the greater atmospheric pressure above the diaphragm 65
resulting from the vacuum, and thereby force the PCB 19
and the fixture assembly 47 toward the ~eceiver 56.
Suitable sensors (not shown) are included with the
present invention to ensure the presence of an adequate
vacuu~, which i~ also an indication that the fixture
assembly 47 and the PCB have been drawn into contact.
An example is now provided showing how the
automation interface 38 operates with fixture
assemblies 47, carriers 20, cover assemblies 62,
conveyor units 12 and buffering stages 16 in Figure
lla. This figure illustrates the sequence of events
for te~ting two diPferent types of PCBs. These two
types are called type A and type B. In the example, 2
PCBs of type A and 1 PC8 of type B are tested. The
fixture assemblie~ associated with each of these PCB
types are called fixture A and fixture B respectively.
It is assumed that all of the necessary components for
the two tests are present in the buffering stage 16
adjacent to the conveyor unit 12 which is itself
adjacent to the test 3tage lla and llb. (See Figures 1
and 6a) Discussion is provided hereinafter as to how
and when the components are delivered t~ the
appropriate buffering stage 16 and how the buffering
stage 16 facilitates the transport of components. Also
discussed later is the scheme used to decide which PCB
type (A or B) is to be given the highest priority in
term~ of utilization of the test stage 11. The example
illustrates only the phyRical operations and motions
that take place in order to test the threa ~pecific
PCBs in this discussion.
Gen~rally, as shown in Fig. lla, components
required to perform the desired test are transported
to, loaded into and aligned in the automation interface
38, the test is invo~ed, and then components no longer
3~
-32-
needed are unloaded and transported from the automation
interface. In the illustrated example, the cover
assembly 62 is employed to test all three boards and
fixture A used in the test of both Type A PCBs. More
particularly, as depicted in Fig. lla, test fixture A
is fixst transported to automation interface 38,
aligned and engaged with receiver 56. Cover assembly
6~ i8 ~ext transported to and loaded into the
automation interface. The irst Type A PCB i~ then
transported to the automation interface and aligned
with fixture A. The test i~ then invoked followed by
unloading and transport of the first PCB Type A from
the automation interface. The second PCB Type A can
~hen be transpor~ed to the automation interface,
aligned with fixture ~, tested, unloaded, and
transported from the automation interface. To test the
Type B PCB, fixture A is unloaded and transported out
of the automation interface and fixture B then
transported in and aligned and enga~ed with receiver
56, The Type B PCB can then be transported to the
automation interface, aligned with fixture B, tested,
and finally unloaded and transported out of the
automation interface.
In Figure lla, the procedure labelled (i) is
expanded in Figure lla(i) to show the low-level
operationR needed to tran~port a component from the
buffering stage 16 to the automation interface 38.
Since this sequence of operations is repeated for
different components, it i8 generalized for any
component. Similarly, the low-level operations needed
to load a fixture assembly onto the test ~tage, load a
cover assembly into the automation interface, load a
PCB onto a fixture assembly, invo~e a test of a PCB,
unload a PCB from a fixture assembly, transport
component~ from the automation interface to the
o~
-33-
buffering stage 16, and unload a fixture from the test
stage are labelled (ii) through (viii), and are
detailed in Figure~ lla(ii) through lla(viii),
respectively.
In Figure lla(i), the following operation~
taXe place to transfer a component from the buffering
stage to the automation interface. The number
de~ignations corre~pond to the blocXs in the Figure.
(1) The rails 44 of the automation interface
38 are elevated to the operating level of the conveyor
unit 120 (See Figure 7c) The number of motor steps
required to elevate the rail~ to this level is known by
the workstation controller.
(2) The rails are extended so that the slots
45 in the rail~ can support the projecting ridges 58 of
the fixture assembly 47, the projecting ridges 29 of
the carrier 20, or the projecting ridges 63 of the
cover assembly 62.
(3) The arm 98 of the as~embly in figures 7
and 7a iq extended by actuating the solenoid 101 so
that the roller 39 will contact the side of a component
when lt is captive in and supported by the rails 44.
The side~ of components include: sides 48a or 48c of
the fixture assembly 47, sides 22a or 22c of the
carrler 20, and side~ 64a or 64c of the cover a~sembly
62.
~4) The dc motor 96 attached to arm 98 is
energized in a controlled fashion so as to cause the
component to move into the automation interface 38.
(5) The dc motor 34 in the conveyor unit 12
iQ energized in a controlled fa~hion so as to cause the
~omponent to move ~oward the automation in~erface 38.
Whan the side o~ the component contact~ the roller 32
in the conveyor unit 12, it i8 transported through the
conveyor unit 12. ~See Figure 5e~ (The buffering stage
~ -34-
16 causes the component to contact the rollers 32, but
th~ exact method is discussed later.~ When the
component exits the conveyor unit 12, one of its sides
contacts the roller 39, and it is driven into the
automation interface. (See Figures 6b, 6c and 6d).
(6) The dc motor 34 is de-energized.
(7) The dc motor 96 is de-energized.
(8) The arm 98 is retracted by actuating the
solenoid lOl, so that the roller 39 is not in contact
with the side of the component.
In Figure lla(ii), the following opera~ions
take place to load a fixture onto a test stage:
(9) The rails 44, supporting the fixture
assembly, are lowered by the stepper motor 36, via a
system of belts, (See Figure 7d) until the contact
panel plate 54 is brought into preliminary engagement
with the receiver, and the fixture assembly is
supported by the receiver. The number of motor steps
required to vary the elevation of the fixture assembly
in this manner is known by the worXstation controller.
(10) The rails 44 are retracted so that the
fixture is no longer supported by the rails (See Figure
7c).
(11) The means to establish electrical
connection between the probe~ 57 of the receiver 56 and
the posts 53 of the fixture assembly 47 i5 now
activated. The fixture i5 now ready to be used for
testing.
In Figure lla(iii) the following operations
take place to load a cov~r assembly into the automation
interface:
(12) The rails 44, supporting the vacuum
cover assembly 62~ are now lowered until the vacuum
cover assembly 62 i3 supported by the fixture as~embly
47. The number of stepper motor steps required ~o
-35-
effect this motion i8 known by the workstation
controller.
(13) The rails 44 are retracted. (See
Figure 7c)
(14) The lifting carriage 85 is lowered
until it contacts the cover assembly 62.
(15) The solenoids 87 are extended, engaging
the support holes 88.
(16~ The lifting carriage 85 is raised,
carrying the cover assembly 62, locked to it by the
solenoids 87. Figure llb shows the cover assembly 62
loc~ed to and raised by the lifting carriage 85.
In Figure lla(iv), the following operations
taXe place to load a PCB onto a fixtur~ a~sembly:
(17) The rails 44, supporting a carrier 20
and PCB 19, are lowered until the carrier 20 is resting
on the fixture assembly 47. The number o stepper
motor steps to do this i~ known by the workstation
controller.
(18) The rails 44 are retracted. (See
Figure 7c)
In Figure lla(v), the following operations
take place to invoke the testing of a PCB:
(19) The lifting carriage 85, to which the
cover assembly 62 is attached, is lowered until the
cover assembly rests on the carrier 20.
(20) The chamber beneath the vacuum cover
diaphragm 65 is evacuated establishin~ electrical
contact between the probes 57 of the fixture assembly
and the nodes of the PCB.
- (21) The test is executed. This f~nction is
carried out by commercially available equipment.
(22~ The chamber beneath the vacuum cover
diaphragm 65 is allowed to return to atmospheric
pressure at the completion of the test.
~-36-
(23) The lifting carriage 85, supporting the
cover a~sembly 62 i~ raised.
In Figure lla(vi), the following operations
take place to unload a PCB from a fixture:
(24) The rails 44 are extended. (See Figure
7c)
(25) The rails 44, supporting a component,
are elevated to the operating level of the conveyor
unit 12. The number of motor ~teps required to effect
thi~ motion is known by the worXstation controller.
In Figure lla(vii~, the following operations
take place to transport components from the automation
interface 38 to the buffering stage 16. This sequence
i~ similar to that illustrated in Figure lla(i) except
the order is reversed.
(26) The arm 98 is extended by actuating the
solenoid 101, making contact between the side of the
component and the roller 39.
~ 27) The dc motor 34 in the conveyor unit 12
is energized in a controlled manner so as to cause
components to move toward the buffering stage 16.
(28) The dc motor 96 in the automation
interface 38 is energized in a controlled manner so as
to cause components to move toward the buffering stage
16.
Steps (27) and (28) cause the component to
move toward the buffering stage 16. The buffering
stage 16 cooperates with the conveyor unit 12 so as to
move the component completely off the conveyor unit 12.
This cooperation is discussed later.
(29~ rrhe dc motor 95 is de-energized.
(30) The dc motor 34 is de~energized.
The second PCB type A is then loaded onto the
type A ixture in the automation interface, tested, and
unloaded and transported from the automation interface.
-37-
In the example, PCB type~ A are no longer going to be
te~ted. Con~equently, fixture A mu~t be exchanged for
fixture B on the test ~tage. It ~hould be noted that
in this example, the same cover as~embly 62 i5 used
with both PCB types A and B. The carriage lift
a~sembly 85 can be u~ed to raise the cover assembly 62
above the operating level of the adjacent conveyor unit
12, making it possible to exchange ixtures without
removing the cover a~sembly 62 from the automation
interface 38. This decrea~e3 the time required to set
up the test stage for a different PCB type, and makes
the operation of the system more efficient.
In Fig. lla(viii) the following operations
ta~e place to unload a ixture a~sembly from a test
~tage ~o that it can be tran~ported out of the
automation interface:
(31) The rails 44 are retracted. (See
Figure 7c)
- (32) The rails 44 are lowered to a height
which will facilitate the engagement of the rails with
the pro~e~ting ridge 58 of the fixture assembly 47.
(See Figure 7c) This height, and the number of stepper
motor ~teps which correspond to it, are known by the
worXstation controller.
(33) The means to establish electrical
contact between the probe~ 57 of the receiver 56 and
the po3t8 53 of the contact panel 54 are deactivated.
(34) The rail~ 44 are extended.
(35) The rails 44, ~upporting fixture A, are
elevated to the operating level of the conveyor unit
12. The number of motor ~teps required to effect this
motion i~ known by the worksta~ion controller.
The step ~hown in Figures lla(i), (ii) and
(~v) - (viii) are then repeated for the type B PCB.
-38-
It should be noted that the system lO of the
present invention can also be utilized with
double-sided ~CBY, i.e. PCBs which have nodes on both
opposing surface~ which must be acces~ed in order to
S te~t the PCB. In this event a second fixture assembly
47' is transported and positioned adjacent the PC~ to
be tested, i.e. the PCB is ~andwiched between the two
fixture assemblies. See Fig. llc. In the preferred
embodiment the array of probes 51' of the upper fixture
assembly 47' are electrically connected to an array of
posts 53'. Such posts 53' of this upper fixture
assembly are, however, supported in plate 54' exterior
to that portion of its surface area which is directly
opposi~e the PCB when the PCB is positioned adjacent
thereto. In addition the lower fixture assembly 47
contains an array of probes 51 which are adapted to
contact the post~ 53' of the upper fixture assembly
when the vacuum or other actuating means draws the
~arious components together.
It should be noted that circuit means other
than the fixture assemblies 47 thus far described can
be utilized as part of the present invention to achieve
electrical contact with the selected nodes of the PCB
to be tested.
A~ has previously been discussed, the present
invention include~ one or more buffering stages 16.
The buffering stages serve as an interface which
connect the various te~t, repair, input/output, 3torage
and other stations that make up a system to the common
transport system 18. From the example illustrated in
Figure lla, one can see that the buffering stages can
be used to store (or "buffer"~ components that are used
in the test station. Similarly, in the preferred
embodiment, component~ can be buffered at each station.
For this reason e~ery test, repair etc. station
~7~
-39-
includea and is joined to the transport sy~tem 18
through a buffering stage 16. From Figure la, it i8
apparent that components are transported into and out
o buffering stage~ 16. It is also apparent that only
one component can be transported in a conveyor unit 12
or 13 at a time, and consequently the buffering stages
16 provide a location where components can be buffered
until the conveyor units are free. (The methodology
or allocating the conveyor units i~ discussed later.)
Since the transport system 18 transports components at
an elevation which can be diferent from the elevation
in the workstations, the buffering ~tages provide a
means to vary the elevation of components between the
two levels. Such buffering stages 16 (See Figs. 12 &
13) include a large receptacle 70 having internal wall
portion~ 71 forming a plurality of slots or cells
adapted to receive and support the carriers 20, the
fixture assemblies 47 and the cover assemblies 62, in a
manner permitting them to move interchangeably in or
out of the receptacle 70. The receptacle 70 i~
supported within a frame 72 so as to be moveable in a
generally upward or downward direction with respect to
the operating level of the conveyor ~nits 12 or 13.
Typically this movement is achieved by driven pinion
gears 73 rotatably mounted on opposing sides (only one
iQ shown) of the receptacle 70. The pinion ~ears 73
are driYen by a conventional stepper motor (not shown).
Corresponding ~lotted racks 75 which are adapted to
engage such pinion gears 73 are fixed to the two
adjacent sides (only one i~ qhown) of the frame 72. In
this manner the rotation of the pinion gears 73 hy the
~tepper motor will move the receptacle 70 in either an
upward or downward direction, to afford the positioning
of the receptacl~ 70 with any preselected one of its
cells at the operating level of ~he conveyor units 12
-40-
or 13. The operation of a bufferinq stage 16 as it
relates to adjacent conveyor units 12 is illustrated in
figures 12a through 12c. In these figures, the frame
structure of the buffering stage is not 6hown so that
the functionality can be better illustrated. The dc
motors, stepper motor, rack and pinion gear of the
buffering stage 16 and the drive mean~ of the conveyor
units (dc motor 34, belt 203, sprocke~s 204 and rollers
32) are also not shown.
In Figure 12a, a carrier 20 is shown driven partly
into a conveyor unit 12 labelled "left". Fi~ure 12b
shows the carrier partially driven into the receptacle
70, where the projecting ridges 29 of the carrier 20
are supported by the Rlots in the internal wall
portions 71. The means by which the carrier is driven
into the receptacle 70 is discussed later. In Figure
12CI the receptacle 70 is shown elevated so that the
carrier is at the operating level of the conveyor unit
labeled "right", and the carrier 20 is shown as it is
being transported from the buffer stage 16 to this
conveyor unit. To facilitate the accurate and aligned
movement of the receptacle 70 it is mounted with
respect to the frame 72 by a plurality of bushings 77
slideably supported on columns 78 which are in turn
affixed to frame 72. The buffering stage~ 16 also
- contain two drive wheels 76 driven by a dc motor in the
~ame manner as has previou~ly been described with
respect to the conveyor units. These drive wheels 76
are pivotally mounted on frame 72 at the operating
level of the conveyor uni~s 12 and 13. To provide
clearance for the receptable 70, ~o that it can be
driven up and down loaded with component~ withln the
frame 72, by the stepper motor via the rack and pinion
mechanism, a solenoid 79 or comparable device is used
to pivot the drive wheels 76 in a manner similar to
-41-
that described with regard to the automation interface
38, from a position at which they are in frictional
contact with the parallel side members of either the
carrier 20, the cover as~embly 62, or the fixture
assembly 47 which i~ within the cell of the receptacle
70 po3itioned at the operating level of the conveyor
units 12 and 13, to a second position which is spaced
from the receptacle 70. It ~hould be noted that the
receptacle 70 contains cut out portions within the wall
portions 71 to permit the drive wheels 76 to intrude
within the interior of the receptacle 70 so a~ to
contact the variou~ carriers 20, fixture as~emblies 47
and cover assemblie~ 62 located therein. The ~tepper
motor of the buffering stage is controlled by the
wor~station controller to facilitate the exact vertical
positioning of the receptacle. The timely pivotal
action effected by solenoids 79 and controlled rotation
of the drive wheels 76 are al~o controlled by the
workstation controller. Appropriate sensors of the
types ~lready deqcribed are included (See Fig. 14) to
verify the presence of a carrier 20 or other
tran~ported article within the cell~ of the receptacles
70. Similarly sen~ors also verify the relative
positioning of the receptacles 70 and the drive wheels
76.
It should be noted that buffering stage~q 16
are also designed to be modular, and a variety of
configurations are therefore po~sible. For example,
multiple slave buffering stages 16 can be stacked
adjacent one another for mass storage of cover
a~semblies 62, fixture assemblies 47, and carriers 20.
In this configuration it i~ useful to have separate
drive wheel~ 76 and a corresponding dc motor on the
side of the buffering ~tage 16 proximate to a conveyor
un~t, and separate drive wheels 76 and a corresponding
-42-
dc motor on the side of the buffering stage 16
proximate to adjacent buffering stage 16. This afford3
more accurate control over the carriers 20, fixture
assemblies 47, and cover assemblies 62 as they are
being transported between adjacent buffering stages 16.
The drive wheels 76 in buffering stages of such a
storage station 5 are preferably all at the same
vertical height. It is also possible with this
configuration to utilize a master buffering stage
controller or microproces~or connected to multiple
~lave buffering stages (5ee Fig. 14). This concept has
already been discussed with respect to the conveyor
units 12 and 13. The ~lave buffering stages while
containing similar motors, sensors, and stops, do not
lS require separate microproce~sors. Alternatively, or in
addition, buffering stages 16 can be utilized as
interface devices to individual repair or test stages.
Thi~ configuration can shorten overall testing and
repair time since, as an example, a batch of carriers
20 containing PC8s 19 requiring the same fixture
assembly 47 can be ~tored proximate a given test stage
and immediately ~equenced through that test stage as it
becomes available. In this configuration it i~ useful
to have separate drive wheels 76 and a corresponding dc
motor operating at different vertical level3 within the
buffering ~tage 16. The conveyor unit3 12 and
buffering stage 16 shown in Figure 13a illustrate how
- the drive wheels 76 in the buffering stage 16 are
positioned at the operating level of the adjacent
conveyor unit~ 12, and that a conveyor unit which is
part of a workstation (labelled right~ can be at a
different level from one which is part of the transport
system 18 (labelled left). In this figure, the two dc
motors 209 are shown at the operating levels of the two
conveyor units. These motor~ are each mounted to a
~-q~
-43-
respective bracket 98' which is mounted upon a
~upporting column 97' (See Figure 13b). The 3upporting
column 97' is in turn mounted to the frame 72. The
other drive wheels 76, similarly mounted to supporting
column~ 97'', are mounted to the opposite side of the
buffering stage 16. Figure 13b further illus~rates the
pivotal movement imparting mechanisms which are
identical to that of the automation interace. Figure
13c shows view B~B from Figure 13. In this figure, the
four drive wheels 76 are shown. It should be
remem~ered that typically the drive wheels near the
bottom of the figure are positioned at a different
level fro~ those near the top of the figure. In Figure
13c, one ~et of drive wheels 76 is shown in dashed
line~ at the extended positions 76a wherein the wheels
are contacting the parallel side members of the carrier
20. From this figure, it is apparent that by
energizing the dc motor 209 when the associated drive
wheel~ 76 are contacting the side of the carrier, the
carrier will be moved. In Figure 13c, this motion will
be toward the top or bottom of the page. For example
components could enter and exit the buffering stage 16
from or to an adjacent conveyor unit at the conveyor
operating level, while at the same time entering from
or exiting to an adjacent te~t stage at the operating
level of such test ~tage.
When the test system 10 according to the
present invention is initially installed the User will
be required to identify certain information about the
system configuration and the PCBs tha~ are to be tested
and/or repaired. This information includes the type of
test stages llt repair stages 15, conveyor units 12,
conveyor units 13, and buffering stage3 16 ~hereinafter
referred to generically a~ "modules") which will maXe
up the system 10, as well as the relative location of
IZ ~4~
each of these modules within the ~ystem 10. It is
possible with this system 10 to create a large variety
of dlfferent physlcal configurations using the standard
modules as described. Several example configurations
S are ~hown in Fig. 13d. It also seems very likely that
an installed system may be modified from time to time
for purposeR such as the addition of new modules or the
creation of additional or alternate paths to high
volume modules, etc. In a traditional automation
~ystem these changes would require the user to
experience a relatively long period of production
shut-down in order to make several mechanical and/or
- software revisions to be able to use the newly
configured system. An important benefit of.the present
inventi.on is that such changes are possible with a
minimum of down-time and virtually no software changes.
Since this ~ystem i~ comprised of a relatively small
set of modules whose functions are known, any
configuration of the system can be reduced to a list of
modules, their types, and the necessary relationship
between them with regard to control and orientation.
Rather than force the user to determine the desired
layout by paper and pencil methods and later input a
fairly large amount of data, the.present invention
provides a convenient method for arranging the modules
of a system and at the same time automatically capture
the required data to operate the system 10. During the
initial start-up the master control unit 17 presents
the user with a grid pattern on a graphical display
device, and a menu upon which the various potential
modules are repreqented schematically, e.g. by icon~.
These icon~ may be selected by the user and placed on
the grid. In .this manner a co~figuration may be built
up ~n an interactiYe way which represents the deqired
configuration of the system. The user may remove and
,~2r~
-45-
replace modules, select new module type3, and
reconfigure the module orientations until the
configuration appear~ as desired. The configuration
may be an entirely new layout, or an exi~ting
configuration may be recalled and "edited" to achieve
the desired result. A configuration may be "saved" at
the request of the user ~o be used either a~ the
default or controlling configuration file for the new
system or as reference when con~idering more than one
po~ible configuration for the 3ystem. Once the
desired physical layout has been determined by the user
a verification routine may be invoked. The
verification routine applies a set of design rules
regarding module placements and orientation to be
certain that the proposed configuration is operable.
An example would be that a corner of the transport
system must contain a conveyor unit 13 with rotation
capability. Should an error be located the user i3
informed of the conflict and the input phase is
reinstated. After passing the rule check the user i~
then prompted for the additional data required to
define the configuration which i3 not ascertainable
from the graphic repre~entation, such a~ controller ID
numbers and the desired interconnections of slave~
units to master controllers. Once this in~ormation is
entered a user may re~uest the 3ystem 10 to print out a
configuration listing. ~his listing describes the
configuration, for example, a list of the modules u~ed
by type, a connection list for cables (including power,
logic, air, etc.) by module, e~c. Thi3 listing aid~
the user in the actual physical connection of a new
configuration. By u ing this interactive method new
configuration3 may be efficiently created by the u3er,
and the resulting data file will automatically contain
the required information to allow the system 10 to be
` ~279~8
-4~-
controlled without requiring any additional input or
changes to the control system software.
Once the configuration is established a
software routine within the operating system recognizes
the icons and provide~ each module with a unique
identity code (typically based upon the controller ID).
The routine then identifies the interconnections
between the modules and const~ucts all the possibLe
paths between the various modules indica~ed and the
re~uired operation of the various modules to facilitate
these paths. The operating system utilizes this
information to con~truct a table which correlates all
the ways to get from the various starting modules (e.g.
buffering stage 16) at which a component may be located
to the various destination modules (e.g. test stage 11)
to which a component will be transported. Such
technique~ as a~e required to formulate this
information are commonly available as evidenced by a
text entitled, "Compute~ Algorithms", written by Sara
Baase, and published by Addison Wesley, 1978. A file, here-
inafter referred to as the system configuration file, is
created containing this configuration and path information.
This file is stored either in the memory associated with the
master control unit 17 or a host computer interfaced with
the master control unit 17, or within a disk file which can
be provided as an input to the master control uni-t 17 when
the system 10 is to be used. This modular approach of the
present invention and the ease in which configuration
changes can be programmed or specified, affords the user
flexibility to easily adapt the configuration of the system
10, to tailor the system 10 to the particular user require-
ments.
-47-
Prior to testing and/or repairing PCBs with
the system 10, the user i8 also required to input
information pertaining to the various types of PcsS
which are to be tested, and the testing sequenCe which
will be required to process each PCB type. Each step
in the ~equence i~ performed by some test or repair
stage which provide~, a~ a re~ult, a coded indicator
(interpreted by the master control unit 17) to
determine which of po ~ible alternate steps of the
sequence i~ to follow. For example, a specific type of
PCB might need to be tested first with a functional
tester. If it fails ~hese tests, (as indicated by the
result supplied by the tester) the PC8 may secondarily
be te~ted on an in-circuit tes~er. Alternatively, the
results may indicate that the PCB need~ to be
transported to a repair station instead of the
in-circuit tester. After repairs and/or adjustments
are made to the PCB it may be routed to the functional
tester to repeat the tests which it previously failed.
Simila~ly, certain test results may require the
performance of certain tasks such as manually probing
various nodes of the PCB. The PCB would therefore have
to be routed to a module affording such tasXs. Thus a
unique proceq~ 3equence is determined for each type or
batch of PCB which is to be tested. This information
i8 stored as a deci ion tree or similar device, where
the process step~ form the node of the tree and the
coded indicators determine the branches between nodes.
The sequence information and a base priori~y code
(associated with the PCB type and used by the system in
processing PCBs of this type~ are stored ln a file,
hereinafter referred to as the process ~equence file,
which can be provided as input to the master control
uni~ 17. The deci~ion tree includes all the various
sequential event3 which can take place in order to test
~Z~4~8~8
that PCB type. The path throu~h the tree for a
particular PCB i dictated by the actual results of
each event (i.e. process step) as they occur. As i8
apparent, the tree will typically have multiple
branches due to the alternative results which might
occur at a given node, e.q. pass, fail, etc. Methods
for creating, storing and performing operations on such
"tree" structures are readily available as evidenced by
a text entitled "Data Type~ and Structures" by C.C.
Gotlieb and Leo R. Gotlieb publi3hed by Prentice Hall
ih 1978 (reference chapters 6 and 7).
An example process sequence file is shown in
Figure 14a. In this figure, each node of the tree is
shown as an overlapping group of labelled boxes. With
each node in the tree, there are identified the process
name (e.g. functional te~t, repair), the physical
location in the sy~tem where the process can take place
(e.g. test ~tage lla or test stage llb, repair stage
15) and a reference to any associated components (e.g.
a particular type of fixture assembly 47) required to
"fiet-up" and complete the proce~ of that node. The
coded indicators associated with the results of the
proces~ are indicated as triangles with internal
label~. The results of each operation are used to
- 25 determine which step will occur next for a particular
board being proce~ed. The results are supplied to the
master control unit by the workstation controller~ via
the host and are not the detailed output of the test
stage. A base priority is as~ociated with each process
sequence file. The priority i~ u~ed as described
hereinafter to regulate resource allocation between
board typeq. In addition, a file i~ created which
contains a look-up table or similar device correlating
a PCB type to a fixture as~embly 47 which corresponds
to such PC~ type. Thi~ fil~, hereinafter referred to
~ 79~8
-49-
as a fixture correlation file, i8 also provided as
input to the master control unit 17. Also included
within the fixture correlation file i8 a liQting of any
other components such as cover assemblies 62 or upper
fixture assemblies 47' which will be required to test
and/or repair that PCB type. Typically the process
sequence file and the ~ixture correlation file will
correlate the PCB type to the corresponding components
by utilizing a serial number located on the PCB, of
which certain digits refer to the PCB model or type,
and the remaining digits uniquely identify each
individual PCB of that type. In the preferred
embodiment, thiQ serial number is associated with the
particular carrier in which the PCB is mounted. The
carrier is uniquely identified by a bar code sticker
permanently affixed to the carrier which can be
automatically scanned by the test system 10 according
to the present invention. Thus once the bar code for a
given PCB is scanned, the operating system can search
the process sequence file through the master control
unit 17 and ascertain the various station~ within the
test syQtem which will be required to test that
particular PCB 19. Similarly the fixture correlation
file can alRo be ~earched to identify the other
components which will be required to teQt that PCB. A
mesQage can be displayed to the operator to make
certain that the correct fixture assemblies 47 and,
depending upon the size of the batch o~ PCB~ to be
tested, an adequate number of cover assemblies 62 and
flxture assemblies 47, etc., are loaded within test
system 10. In the preferred embodiment, fixture
as~emblie~ 47 and cover as~emblies 62 are stored within
a storage staton S when not in use. The carriers 20
(containing the PCB~ 19 which are to be te~ted) can
then be individually inserted into the te~t system 10
, - - .
' .
-50-
or these components can be stored within a buffering
~tage 16 and automatically transported by the conveyor
means. In the preferred embodiment a file, hereinafter
referred to as the component location file, is created
containing component location information and i8
maintained by the ma~ter control unit 17. In this
manner the loading and unloading of the various t~t
stages 11 and 15 and the transport of the various
fixturing assemblies 47, cover assemblies 62, and PCB~
1~ 19 can be facilitated by the master control unit 17 in
conjunction with diQtributed controllers of the test
system 10. It ~hould be noted that the information
thus far described a~ being required as input can be
provided through standard input (e.g. Xeyboards,
lS magnetic storage devices, LAN connection~ with another
computer, etc.) interfaces (See Fig. 14) to the master
control unit 17.
In the preferred embodiment of the present
invention, a component input/output station 80 is
included within the system 10. The basic input/output
station 80 comprises a conveyor unit 12 combined with a
bar code reader 81 and buffering stage 16. Since the
carrier 20, the fixture assembly 47 and the cover
assembly 62 all have the same outer configuration, the
input/output station 80 can be utilized as a system
input/output device for all components. Accordin~ly, a
bar code can be placed on all component~ to provide an
identity to the master control unit 17. Upon insertion
of a component into the input/output station 80, the
bar ~ode reader 81 would scan the component and input
the erial number associated with the component to the
master control unit 17. The master control unit 17
could then ascertain a destination for th-e component by
examining the proces~ sequence file, e.g. an empty cell
within a bufferin~ stage 16, and direct the conveyor
08
units to transport the component to that cell. The
identifying number for the component and a code
identifying the destination, e.g. the buffering stage
16 and the particular slot within that stage are stored
in the component location file. As the components are
processed by the system 10 this component location file
is dynamically updated to provide an indication of the
current location of each of the components.
Reference is now made to Figures 14b, 15 and
15a for a description of the overall operation of test
system 10. During start-up of the test system 10 the
master control unit 17 reads the system configuration
file to determine what modules exist within the test
system 10 as well as their functions and relative
locations. The master control unit 17 then initializes
the system configuration by interrogating each of these
various modules over the networX 14 and determining
` whether the modules are on-line and functioning. If a
particular module is not functioning the user is so
informed. This module can then be mapped out of the
system or the malfunction can be corrected as
determined by the user. The control programs for each
- of the ~icroprocessors located within the various
modules are then downloaded from the master control
unit 17.
Figure 14b depicts a control structure for
the system 10 of Figure 1, in a block diagram fashion.
The master control unit 17 i8 shown at ~he top of the
- figure with local area network 14 connecting it to the
various controllers. Each station of the system 10 is
equipped with a microprocessor controller.
Additionally, each conveyor unit 13 contains a
microprocessor controller for it and some number of
slave conveyor units 12. This distributed controller
3s epproach gives the maeter ccntrol un~t 17 ccntrol over
,
.,
-52-
all aspects of the system lO without requiring that all
- software be executed on the master control unit.
In the preferred embodiment, the controller
for storage station S i5 housed in one of the buffering
stages 16 making up the station. This controller
operates the buffering stages 16 on direction from the
master control unit 17 to effect the storage and
retrieval of componentsO
The controller for each conveyor unit 13 is
hou~ed in the base of the unit and controls the
conveyor unit 13 and slave conveyor unit~ 12 on
direction from the master control unit 17. In the
preferred embodiment each conveyor unit 13 in a system
is equipped with a microprocessor controller. The
master control unit 17 coordinates activity between the
controllers of the conveyor units 13 to effect the
transfer of components between stations of the system
10 .
Each test station containing a test stage 11
contains a microprocessor controller. In the preferred
embodiment, the controller operates the buffering stage
16, the conveyor unit 12, the automation inter~ace 38,
and the test stage lla or llb of the station on
direction from the master control unit 17. The
controller additionally ascertains, through an
interface with the test ~tage, the result of any test
which is performed. ~his result (for example
PAS5/FAIL) is later used by the master control unit 17
i~ routing the PCB 19 to any subsequent station.
The microprocessor controller in the repair
station controls the buffering stage 16, the conveyor
unit 12, the I/O device, and the repair stage 15 in the
repair station, on direction from the master control
unit. In a manner ~imilar to the test station, the
controller in the repair station also appraises the
-53-
master control unit 17, via the host, of the completion
of any repair activity to facilitate the routing of the
repaired PCs 19.
The input/output station 80 contains a
microp~ocessor controller which controls the buffering
~tage 16, the conveyor unit 12, and the bar code reader
81 of the input/output station on direction from the
master control unit 17. In the case of the
input/output ~tation 80 the microprocessor controller
also appraises the master control unit 17 of the
arrival of a carrier containing a PCB 19 to be
processed. In the message, the bar coded identity of
the carrier 20 and the PCB 19 are provided to the
master control unit 17. These items of information are
maintained by the master control unit 17. The master
control unit 17 also maintains the process sequence
file which contains routing information and a base
priority for the new PCB 19. The identity information
i8 used in accordance with the proce3s sequence file to
control the path taXen by the PCB 19 in respon~e to the
result~ from the stations to which it is sent.
The preferred embodiment of the present
invention utilizes substantially identical
microproceqsors in the various modules and stores the
specific control program~ for the variou~ types of
modules in memory or in a file which can be down loaded
to the variou~ microproces~ors as required. Each of
the microprocessors i~ equipped with a ROM based
monitor program which execu~e~ on power-up and readies
the microprocessor to receive command3 over the network
14. The specific control programs which are downloaded
are determined by the configura~ion of the ~ystem and
the 3pecific module~ employed. This concept along with
- the master/slave concept already discussed with regard
to the conveyor units and the buffering stages affords
-54-
the user lower costs and added flexibility for easily
reconfiguring the system to tailor it to the particular
requlrements present. Furthermore it eliminates the
need for fixed memory or disk drives at each module.
Once the various control programs are loaded, the
moveable members of the various modules are "homed" to
Xnown positions, and the appropriate sensors are
interrogated to ensure that such homing has occurred.
~fter every module has been checked and initialized in
this manner the user is informed that the system is
ready for use.
Testing and/or repair of PCBs within the test
system 10 is done on a priority basis, with the intent
of keeping the test and repair stages occupied to the
fullest feasible extent. Through one of the various
input devices, the user associates a base priority code
for the PCB type dependent upon production
requirements. When a PCB is entered into the system 10
as previously described it is assigned the priority
code which was provided as input to the master control
unit 17 as part of the process sequence file for Pcss
of this type. As has been previously stated, the
master control unit 17 in combination with the bar code
reader has already ascertained the identity and type of
the PCB being supplied to the system 10 for test and/or
repair. In the preferred embodiment of the present
invention thi~ priority code i8 dynamicO That is, a
PCB awaiting an operation to be performed at a
particular stage or module will have it~ priority code
or rating increased as a function of the time it is
waiting. This prevents a PCB from being held at a
given stage or module in the system for an unlimited
period of time such as might be the case when it is
preempted by a large batch of hi~her priority PCBs.
The PCB returns to its base priority, i.e. that taken
~'~79~
-55-
from the process sequence file, after the required
operation has been performed at that particular module
or stage. Since the PCBs must share the various
modules, a pre-emptive scheduling technique i8
employed, in which operations on PCBs having the
highest priority will generally be processed first.
Scheduling algorithms to implement a pre-emptive type
system are readily available as evidenced by a text
entitled "Operating System Principles" by Brinch
10~ Hansen, published by Prentice Hall in 1973 treference
chapter 6).
A flowchart illustrating the dynamic priority
scheme for any given PCB and the preemptive scheduling
technique for using the dynamic priority to schedule
the use of a test/repair station is provided in Figure
15a.
The essential purpose of the control system
of the present invention is to track and control the
movement of PCBs between the various workstations. A
mix o~ different board types must be tracked and
controlled simultaneously. In a fully automatic manner
the PCBs are routed from station to station with any
necessary setup being performed as required. The
method of establishing the rules, i.e. process sequence
files, used to guide the different types of PCBs
through the test and repair system, was described
` earlier. An example of how the rules are applied by
the master control unit will now be provided. It
should be remembered that although PCB routing is the
primary func~ion o the master control unit, it is by
no means its sole activity. Monitoring the ~tatus of
work~tations and conveyor units is continuous as well
as providing status reports on production and equipment
to a factory HOST computer on demand.
-~6-
Basically the routing control portion of the
master control unit software i8 an infinite loop,
repeatedly examining the status of each printed circuit
board within the control of the system and acting on
S any changes provided by workstation activity. This
loop can be regarded as having two sections as shown in
Figure 15a. The top section determines all PCBs which
have completed a step in their processiny and
ascertains the next required step. The bottom section
applies load balancing and priority guidelines in order
to determine which PCBs will be granted workstation
time to proceed and which must wait.
As earlier discussed in connection with the
creation of process sequence files, PCBs which are
handled by the system are regarded as progressing in a
step by -qtep fashion. At the completion of a step, the
product "asks" for the next step based on the result of
the first, thereby taking a "route" determined by
actual events. The steps are regarded as "proce~se~"
which may be any required action at some point in the
manufacturing or testing o~ the PCB. These processes
can be performed by workstations, with perhaps several
stations being able to perform a process by virtue of a
"setup" change. For example, several workstations may
contain test equipment which can be used to perform
tests on different types of PCBs with appropriate
fixtures and test programs. Workstations are regarded
aR the system resources with processes competing for
themO The determination of which workstations are
performing some process for a given PCB type and for
how long, i.e. a preemptive scheduling technique, is
- what will now be described.
A workstation which has performed some
proce~s step on a PCB notifie~ the Master Control Unit,
via the host, of the completion of that step and any
~'~7~4~
associated results, for example, pass or fail in the
case of a testing step. Those PCBs for which such a
completion message has been received since the last
pass through the loop are considered ready to advance.
For each such PCB the appropriate process sequence file
i~ referenced, with different PCB types perhaps having
different process sequence files. From the process
sequence file for each PCB a determination is made of
the next step or procsss required for the PCB, a base
priority associated with the process sequence file i5
attached to the PCB, and a "bid" for that process is
registered on behalf of the PCB. This sequence is
repeated for all PCBs which have been deemed ready to
advance with the distinction that each PCB left waiting
by the last pass through the loop has an increment
applied to its priority before a new bid i3 registered.
This is done to prevent a PCB with a low base priority
from waiting indefinitely for higher priority PCBs to
be processed.
Once all the bids for processes have been
regi~tered, the next section (labeled as "for every
bidding process" in Figure 15a) applies the priorities
of the bids in order to make a selection of which
processes will be performed at workstations. The
selection may be done in essentially the following
manner. Each process step for which a bid has been
registered has associated with its process sequence
file, a list of the worXstations which may be used to
per*orm that process, and also the identity of any
fixture which might be required by the station.
Contention for workstation time may be resolved by
granting time to the proceRs for which the total bid is
highe~t, That is the sum of all ~he bids. If the
worXstation in contention is currently serving the
highest bidder, PCB~ requiring that step are advanced
~Z7~38
-58-
and will be physically routed to that station. In the
case of more than one station serving a function, PCBs
are routed to the station with the least product in its
station buffer to keep stations balanced.
If the workstation in contention is serving
another process step, it may be preempted. This means
that any fixture change required will be invoked as
well as any program changes needed by the station
equipment.
These stepR are repeated for all processes
requesting worXstation time. PCBs which cannot be
serviced at the end of thi~ section are delayed and
reconsidered in the first section of the next pass
through the loop as noted earlier.
In its preferred embodiment the system lO
will process multiple PCBs in parallel. For ~implicity
and clarity purposes, the discussion which follows will
view the system 10 from the perspective of a single
highest priority PCB under test. It should be noted,
however, that generally other high priority PCBs are
being processed concurrently, i.e. in parallel. The
configuration file and appropriate process sequence
files for ~uch highest priority PCB are read by the
master control unit 17. These files have been provided
as input to the master control unit 17 by one or more
of the alternatives previously described. The master
control unit 17 will determine from these files the
station which i5 required to perform the first testing
or repair operations to such highest priority PCB.
Similarly, the required other components, e.g., the
fixture assembly 47 and cover assembly 62 for repairing
and/or testing such PCB will also be established by
reading the fixture correlation file. The master
control unit 17 will then read the component location
file to determine if such components are within the
~L~d7~' ~
-59-
system and the present location of the required
components and the PCB. If such components are not
within the system 10 the user will be instructed to
input such components. These other components required
to test such highest priority PCB are then
automa~ically tran~ported to the required module. The
ma~ter control unit 17 i~ designed to interrogate on a
periodic basis the various sensors included within the
system 10 to update the location information (as
contained within the component location file) for all
components within the system 10 and ensure that the
required components are present and accounted for and
that on a real time basi~ two or more components (e.g.
carriers 20, fixture assemblies 47 and cover assemblies
62) never try to occupy the same space at the same
time. Similarly the appropriate sensors will also be
interrogated for the status of each of the modules
within the sy~tem 10 which form the various paths
available to tran~port the components, to ensure that
the modules are functioning, available, and properly
positioned for such transport. Methods for controlling
the use of resources such as conveyor units and fixture
assemblies by competing tasks (PCBs in this
application) are readily adaptable from methods
outlined for computer peripheral control in such text
as "Operating System Design, The XINU Approach" by
/ Douglas Comer, published by Prentice Hall in 1984
(reference chapters 12 and 16). In this manner the
optimum aYailable path to such test and/or repair
stages can be established for each component and the
proces~ of transporting such highe~t priority
components from their present location to the required
te t and/or repair stages will begin. From the-
information contained in the configuration file, the
master control unit 17 can ascertain the rotations of
' :
-60-
the rotating conveyor unit 13 and the direction of
rotation of the dc motors 34 in the conveyor units 12
and 13 required to transport these components to the
specified stages 11 or 15 and the instructions which
will be required to be issued on the network 14 to the
appropriate drivers within the various modules to
perform the transport. Accordingly the master control
unit 17 will timely activate the appropriate drivers to
perform ~uch transport. Similarly the appropriate
sensors can be activated or interrogated to ensure that
the required actions have ~aken place. This drive and
sensor information is conveyed to and from the master
control unit 17 via the networ~ 14. Concurrently the
component location file is dynamically upaated as a
result of such sensor information to, as previously
described, continually track the location of the
variou~ components within the system. The master
control unit 17 is therefore kept current as to whether
a required module has performed the required operation.
After such operation has been performed, and depending
upon its priority rating the PCB will eventually be
transported to the required test or repair station
according to the order specified by the process
sequence file. When all the requ~red components are
present at the specified test or repair station 11 or
15, the master contrcl unit 17 or the host computer to
which it is connected will activate the station to
begin performing the required operationsO In the
preferred embodiment the management of test data is
performed by a host computer, not part of the present
invention, interfaced with the master control unit 17
via the network 1~. The particular test programs which
are required to test the PCB are stored within the host
computer and downloaded as required. The identity of
the appropriate test program is part of the process
-61-
sequence file. Furthermore the host can track the test
history o the PCBs being processed in order to create
management reports for the user.
The completion and results of the tests of a
given PCB at a given workstation (e.g. PASS/FAI~) are
conveyed to the master control unit by the controller
of that workstation. The master control unit 17 via
the process sequence file determines the next event or
operation which is to occur. Assuming an adequately
high priority, the required conveyor ~nit activation
will be analyzed and carried out according to *he
process as has been described to facilitate this next
event or operation. The master control unit 17 will
thus read the process sequence file and ascertain the
next test or repair stage required. If no additional
- test or repair operations are required, e.g. the PCB is
finally accepted or rejected, the PCB will, according
to its priority rating, be transported to the buffering
stage 16 of the component I/0 station 80 for removal
from the test system 10 as indicated by the process
sequence file for PCBs of that type. If additional
test or repair operations are required, the appropriate
stages and modules will be interrogated and selected.
In addition, the master control unit 17 will determine
the associated components which will be required and
which have been specified within the fixture
correlation file. These may or may not be the same
components used for the first test. If the appropriate
fixture assemblies 47, cover assemblies 62, etc. are
already within the system, their current location will
be known by an entry within the component location
file, otherwi~e such components and the relevant
lo~ation information must be provided via one of the
above mentioned input devices. A priority rating is
effectively give~ to the fixture assembly 47, cover
~Z~ 8
-62-
assembly 62, etc., based upon the current priority
rating of the PCs being tested. Finally, the
activation of the conveyor units which will be required
to transport the components to such specified test or
repair stage will be determined and initiated.
This process is ~epeated until such time as
all test and repair operations to be performed by the
system 10 have been completed. In the same manner as
has already been described, the completed PCBs 19 and
19 associated components are transported to the buffering
stage 16 of the component I/0 station 80 for removal.
- lf required, the system 10 will also allow the manual
removal of a PCB and associated components at a test or
repair stage 11 or 15. If this is done the master
control unit 17 must be informed of such removal via
the miscellaneous I/0 devices discussed, e.g. a bar
code reader. It should be noted that, typically,
fixture assemblie~ 47 and cover assemblies 62 will
remain at a given test station until the tests
requiring such components at that station are completed
or until the master control unit 17 determines that the
particular station will be required to test a higher
priority PC8. This may, or example, occur when the
~um of the priority codes for all the PCBs 19 of the
same type (i.e., a new batch) which are being entered
into the system 10 exceeds the sum of the priority
codes for all of the PCBs 19 of the type currently
under test (i.e., the batch being tested). In this
event the fixture and/or cover presently in that
station may be temporarily bumped or replaced by those
required for the higher priority batch, as depicted in
- ~ig. lla.
Thus the system 10 according to the present -
invention is able to concurrently and automatically
process PCBs of different types, sizes, shapes, and
,.~. , ~.
~2~
-63-
configurations with little or no manual intervention
required. It i9 able to do so in an accurate, timely,
and cost effective manner, with maximized throughput,
thereby substantially eliminating the aforementioned
limitations associated with the prior art. It must be
understood that the description of the present
invention as et forth above does not attempt to recite
all the advantages associated therewith nor does it
attempt to recite in detail all the structure employed.
Furthermore, for the sake of clarity and understanding,
certain operations and the detailed structure of
various components, which are considered to be within
the scope of one skilled in the art in light of the
teachings set forth herein, have not been discussed.
Having thus described several embodiments of the
present invention, it must also be ~nderstood that
changes may be made in the configuration, size, makeup,
shape, or order of some of the parts, circuit~, or
methods described herein without departing from the
present invention as recited in the appended cl~ims.
