CA1330360C - Circuit testers - Google Patents

Abstract

ABSTRACT
The present invention relates to improvements in test equipment for electronic circuits already mounted and assembled, notably printed circuit boards. The invention provides observa-tion means based on the use of the electro-optical principles of certain materials. A functional tester in accordance with the invention has easy effective diagnostic processes.

Description

IMPROVEMENTS TO CIRCUIT TESTERS

The present invention relates to circuit tests and more especially to tests of assembled circuits, such as printed circuit boards.

Apparatus is known, arranged to effect such tests automatically, which makes contact with the circuit nodes, applies excitation signals to these nodes, and monitors the response signals produced by the tested circuit. Such apparatus is often called in the art "Automatic TPst Equipment" (abbreviated as ATE).

ATE may be classified into two types, namely functional testers and device testers. In a functional tester inputs are applied and outputs received only at the normal input and output nodes of the circuit~ such as for example an edge connector for a board mounted circuit. Input signals are applied and output signals monitored to assess the functionality of the circuit assembly as a whole.

Functional testers present the problem that, particularly where complex circuits are involved, a large number of input/output combinations (referred to in the art as test patterns) are required to fully exercise the circuit, and an exhaustive test may be very lengthy. A further problem is that it may be impossible to test some circuit components, notably digital memories and counters, since certain states of such devices cannot be controlled from externally accessible nodes alone. For example, there may be no external connection to the reset input of a counter~
~ ri~
'' .'' .'''.' '. ~'''' ''"".'~,",'"',, ~.'' "' ` - 2 - 1 3~ 0 3 6 0 Equally, some output conditions of some devices may not be externally distinguishable. For these reasons some circuit assemblies can be only partially tested with a functional tester.
On the other hand Device testers have the advantage of being able to test each component of a board mounted circuit individually, since they are designed to contact internal circuit nodes, that is nodes other than the external input and output nodes. However this advantage is obtained at the expense of a special piece of hardware, referred to in the art as a "bed of nails" fixture, comprising a plurality of spring loaded probes, positioned individually to contact nodes within a circuit. By this means component inputs and outputs may be driven andmeasured to establish the functionality of each component individually.

Disadvantages of in-circuit (device) testers are primarily the hardware required in providing a fixture adapted to each circuit to be tested, for example each board, and means for holding the board in place and in contact with the fixture. A less apparent disadvantage is possible damage to circuit devices when an output of a device is driven via a nail to a state other than that which would logically result from the signals at its inputs in order to test another device having an input connected to that output. Such backdriving, as such a circumstance is referred to in the art, can damage the backdriven device by for example causing excessive heating therein.

Further, the minaturisation of circuits leads to growing difficulty in access to the internal nodes of tested circuits. Finally, in-circuit testers are intrinsically poorly adapted to monitor the overall functionings of a circuit, notably those which interfere with the respective timing of signals produced by the various components. For these reasons, a certain number of reservations endure about the use of in-circuit testers.

` 1 330360 It is nevertheless desirable that ATE is able to provide contemporaneously analytic information allowing validation of the functionality of the components taken individually, and further derived information allowing validation of the global functionality of the circuit, which depends not only on the functionality of each component, but also on the interactions between components.

Whilst in circuit testers give priority to the first aspect to the detriment of the second, functional testers do essentially the opposite; also the so called diagnostic phase, in the course of which the latter serve to isolate the probable cause of a fault, is a delicate operation. In fact it is generally impossible to locate a fault by exercising external inputs and monitoring external outputs alone since there will not be a unique test pattern corresponding to every possible fault. rhis problem is to some extent overcome in some functional testers by providing a manual probe for use during the diagnostic phase on boards which have failed the functional test.

During such a phase, an operator plac~s the probe at ;~
various nodes of the circuit in such a way that an internal node may be monitored in a way somewhat similar to that in which it is monitored in an in-circuit tester. Such a procedure is however slow and necessitates a skilled operator. ~ ;
.' ~,.
In this context, an object of the present invention is to provide an apparatus which not only allows a functional test of a circuit comprising a plurality of components placed upon an isolating support and interconnected by a -~
network of conductors extending over at least one surface of the support, such as a printed circuit board, but which also provides a possibility of obtaining a diagnosis of the cause of an established functional fault without posing the ~ 4 ~ ~" 1 3 3 0 3 6 0 same mechanical connection problems nor the damage problems which are posed by an in-circuit tester.

In accordance with a first aspect, the apparatus of the present invention comprises a layer of an electro-optical medium of dimensions substantially the same as the dimension of the assembled circuits, and placed electrically proximate the conductor bearing surface.

From the prior art, for example from US Patent 4 618 819 the use of electro-optical materials for the remote monitoring of electrical signals is known, that is without mechanical contact with the conductors which carry the signals. The system described in that patent uses electro-optical crystals placed physically proximata the surface of a not yet encapsulated integrated circuit. A
polarised light beam is directed towards a region of the crystal in the neighbourhood of a conductor carrying~an electrical signal to be measured, and reflected. In accordance with well known laws, a characteristic of the reflected light is affected by the electric field around the conductor, such that by appropriate detection this electric field and the signal which produces it, may be detected. Thus an image of the electrical signal in the conductor may be obtained.

Although the invention adopts certain features of this prior apparatus, it is distinguished by a number of points, notably by reason that the known apparatus is intended exclusively for the measurement of signals circulating within an integrated circuit and not for the examination of signals flowing between components of an assembled circuit, such as a printed circuit board. Thus, not only does this apparatus not make apparent the technical possibility of practising an electro-optical method on a circuit which at the same time has large dimensions compared with those of ~ ~ 5 ~ 1 3 3 0 3 6 0 an integrated circuit and very significant height variations which are not present at all in an integrat2d circuit, but above all the known apparatus, which only leads to the gathering of analytical information not representative of global functionality, is not concerned with the problems posed by the testing of assembled circuits, and neither teaches nor leads to the application of an electro-optical method to satisfy the requirements o~
functional tests and incircuit tests.

In one possible embodiment of the invention, the electro-optical material may be constituted by a polymer film endowed with electro-optical properties. This film may be applied directly to the conductor bearing surface before the mounting and assembly of the components. This film may also be used to lie flat against the conductor bearing surface after component assembly. It then includes openings in which these components or solder pads are accommodated. It may ~e incorporated in an optical probe for testing, being placed electrically proximate to the surface of the conductors of a board to be tested to perform a test. Such a film may for example be made in the form of a transducer including: an element of such electro-optical ~ilm dimensioned and configured with regard to the dimensions and configuration of a circuit to be tested; on that face which is to be placed electrically proximate the conductors, at least one element re~lecting light which reaches it after crossing the thickness of the film; and on the other face a conductive transparent or semi-transparent layer serving as a reference electrode.

Apparatus in accordance with the invention intended to constitute ATE includes, in addition to the electro-optical layer:-- means for directing, at any region o~ the electro-optical medium and for receiving in return, light such that electro-optical effects arising in the medium may be detected, - means for applying an electrical test signal configuration at one or several external nodes such that a response signal is produced at at least one circuit node, the light being directable at a region of the electro-optical medium which is electrically proximate a conductor constituting the node such as to produce an analogue , - and means for comparing this analogue with the idealised response of a known to be good circuit and for providing an output signal representative of this comparison.

In accordance with an embodiment of the for~going apparatus, the light directing means includes acousto-optical means for deflecting the light beam striXing the electro-optisal medium layer.

In an optical probe realised in accordance with the foregoing principles, a field lens may be provided having dimensions at least equal to the circuit to be tested.
Advantageously, this lens is plano-convex, its planar face being proximate the circuit to be tested. In accordance with one embodiment, the electro-optical layer is affixed to the plane Pace of the lens. ;`

For some applications, the electro-optical layer is chosen with a structure such that the component of the electric field which is perpendicular to the plane of the electro-optical layer is detectable by means of polarised light incident normal to the plane of the layer. For the case of an electro-optical polymer film the electro-optical properties of the film may be predetermined, by raising the film to a temperature sufficiently in excess of that at ~- l 33036a which its molecules acquire a measure of mobility and by exposing the film to an electric field of a predetermined orientation. Upon cooling the film, provided that the field is maintained, the molecules fix themselves, keeping their preferred orientation. In accordance with the invention, this orientation may be chosen to be for example perpendicular to the plane of the film or in the plane of the film. It is also possible to operate with polarised light of angled incidence with respect to the layer, for example 45.

The means for analysiny the electro-optical effects created by the local voltages of the circuits to be tested in the electro-optical material may be analysed by a polarimetric method as has just been described or, as a function of operating conditions, by an intefe~ometric method.

When the analysis means are of the polarimetric type it i5 advantageous to provide in the path of light reaching the electro-optical medium layer a reflector to enhance the component of the electric field in the direction of polarisatio~ of the initial incident light. In the case of a polarimetric analyser, in which output light from the electro-optical layer is analysed along two axes, it may moreover be beneficial to provide a means of imposing a phase shift in the output light form the electro-optical layer controllable as a function of the output signal of said analysis means.

In the case of an analysis in accordance with a interferomatic principle, for example that of Fabry-Perot, it is also advantageous to provide means permitting the adjustment of a functional parameter of the system suzh that the wavelength of the light used, in order to locate the analysis means at a substantially optimum region.

ATE in accordance with the present invention preferably operates as a functional tester, operating in a test phase in which only the external inputs and outputs are excited and monitored, and further, in the case of a failure in the test phase, in a diagnostic phase in which more nodes, particular internal nodes, are examined.

The nodes are advantageously examined in turn, and a multiplicity of test patterns applied during each test.
Preferably, all possible test patterns which affect the node being examined are consecutively applied~

Thus apparatus in accordance with the invention allows considerable modification and improvement in diagnostic techniques in a way not suggested by the prior art. In fact, with existing manual diagnostic probes, it is necessary to use complex test sequences at the circuit inputs to obtain a non-ambiguous fault signature at some diagnostic points only (seldom more that ten) to take account of the time and degree of intervention necessary on the part of the operator performing the diagnosis.
Specifying such multiple vector test sequences for each diagnostic point is complex and costly. By contrast, the apparatus in accordance with the invention offers the ability to probe, in diagnosis, a large number of points very quickly: as a result simple test sequences may be used at each point; the availability of measurements at numerous points allows ambiguities to easily be resolved.

In accordance with a preferred embodiment o~ the invention in the caæe where the electro-optical medium is not constituted by a polymer film directly applied to the conductor bearing surface, the medium is placed electrically proximate through the agency of an interface member, adapted to relay to the electro-optical member electrical potentials appearing on said conductor bearing ~ 1 3 3 0 3 6 0 72722-19 surface whilst preserving their relative spatial placement. This interface member may comprise a plurality of essentially parallel conductive columns isoIated one from another, the assembly taking the form of a flexible sheet of dimensions substantially equal to ~ ;
those of the member. This sheet may moreover have a profiled surface to marry with the shape of the assembled circuit, if this is not planar. `
In accordance with another aspect of the present invention, the light may be directed in a general way toward the `-electro-optical medium'or a part thereof. The electro-optical medium includes a plurality'of sites, which may be individually electrically'biassed, aIong the path of the light. In use, a single site may be made to function by such bias, such that the light detected at the output is representative of a single region `~ ~
of the medium. -~ ~`
The invention relates equal~ly and individually to the printed circuit board, the optical probe, and the transducer for putting into effect the principle in accordance with the invention disclosed above.
According to a'broad aspect of the invention there is ' ~ `
provided apparat'us for circuit testi~g, such as a printed circuit ~' board, the circuit cbmprising a plurality of electronic components '~
arranged on an isolating'support and interconnected by a network ~'~
of conductors formed on at least one surface of the support, the apparatus including a layer of an electro-optical medium of dimensions substantially equal to those of the circuit placed, in ' - 9a - 1 330360 72722~19 use, electrically proximate a conductor bearing surface such that an analysis of changes in incident light under the effect of voltages appearing in said conductors as a response to the application of test signals to the tested circuit may be made r the apparatus further including means to direct the light toward any :~
region of the electro-optical medium and for receiving light emanating rom this region, such that the electro-optical effects appearing in the medium may~be detected, and means for applying a pattern of electrical test signals to one or more external nodes such that a response signal is produced at at least one circuit :' ~ ' node, the light being directable toward a region o the electro-optical medium which'is electrically proximate a conductor ~.
constituting the node, so as to produce an analogue.
According to another broad aspect of the invention :
there is provided an optical probe for a circuit~tester, such as printed circuit boards, such a circuit comprising a~plurality of '~'' electronic components arranged on an isolated support and inter-connected by a network'of conductors present on at least one surface of the support, a layer of'an electro-optical medium having ~:
one face covered with a reflecting layer placed, at least during the test, electrically'`proximate test points on the conductors on ~ .
the circuit to be tested, and said circuIt being supplied with electrical test signalis so as to glve in response a voltage signal at at least one of the test points`of the circuit, this optical probe being chracterized in'that it includes: means to produce and direct a light beam in the direction of a predetermined circuit ~.~;' ' - 9b - 1 3 3 0 3 6 0 72722-19 test point of the circuit mounted in the tester; means to analyse light obtained in response from the electro-optical medium in ,~ , proximity of said test point; and means to control the displace-ment of said beam from one test point to another at the surface of said circuit to be tested.
According to another broad aspect of the invention ~ ~
there is provided a transducer for a testing circuit, such as a '' circuit on as a printed ciEcuit boa~d, the circuit being constitut- ' ed by an assembly of electronic components mounted on an isolating '~
support and interconnected by a network of conductors present on the surface of the support, and of-which certain regions at least are intended`to be subject to a voltage test characterized in that it includes: at least'an electro-optical polymer film ~' element having dimensions and configurati~n allowing one of its faces to be placed electrically proximate one of a plurality of test points on the network of con~uctors of said circuit at the support surface; a reflective covering on said face of the film ~ '~
placeable in electrical proximity to at least one test point and ' able to deflect~incident light'which it receives across said film toward the other face; and at least one conductive covering transparent to the inciden* test light connectable to a reference potential. , According'to another broad aspect of the invention there is provided a printed circuit'board comprising an isolating ~' - support on a surface of which'is ~resent a network of conductors allowing inter-connectlon of electronic components'mounted on the ''~.' ' ' ~ ~

- 9c - 72722-19 support characterized in that it includes above each of a plural-ity of test points at the surface o~ said conductors at least anelectro-optic polymer film element, said polymer film comprising on its surface proximate the test point a layer reflecting :
incident light toward the other face after passing through the ~:
thickness of the elec~ro-optical polymer film.
In order to give a better appreciation of character-istics and advantages of the present invention, particular embodiments will now be described, by way of non-limitative examp-es, with reference to the accompanying diagrammatic drawings, of which:-- Figure 1 represents an embodiment of ATE in accordance with the invention, - Figure 2 represents parts of the equipment of Figure 1, - Figure 3 represents an optical assembly, - Figure 4 represents an interface element, - Figure 5 represents an alternative interface element, ': .' ' '' -` .~` '' , ., :.: ..

O - `` 1 330360 - Figure 6 represents alternative ATE in accordance with the invention, - Figure 7 represents a characteristic o~ the electro-optical effect with respect to applied electrical voltage, - Figure 8 is a perspective view representing the elements of the ATE of Figure 6, - Figure 9 represents schematically an embodiment of a deflection system used to optically examine the surface of the tested circuit board, - Figure 10 represants an embodiment of an optical separator, - Figure 11 shows placement of contact elements during a test, - Figures 12a and 12b show the anisotropy index of an electro-optical film, - Figure 13 represents a sectional view of a printed circuit board on which an electro-opticai polymer has been incorporated for testing, - Figure 13b shows in some detail an alternative to Figure 13, - Figure 14 shows an embodiment in which a preformed electro-optical film is placed upon a board, - Figure 15 shows a plan view o~ the film used in Figure 14, - Figure 16 shows an alternative embodiment of the film used in Figure 14, - Figures 17 and 18 show schematically a method of preparing a polymer film use~ul for putting the invention into practice, - Figure 19 represents an arrangement permitting the detection of the Pockels effect by an interferometic Fabry-Perot technique, - Figure 20 shows an alternative to the technique of Figure 19, - Figure 21 represents an arrangement permitting a ~ ~ , ' .

detailed analysis of wave forms of the signals at circuit nodes.

In accordance with the invention automatic test equipment includes (Figure 1) a layer 10 of electro-optical medium of dimensions substantially equal to those of an assembled circuit, such as a printed circuit board 11 to be tested, the assembly being combined such that the electro-optical medium is electrically proximate the surface 12 of the board 11 which carries the electric circuit, the latter comprising components such as 15. The surface 12 carries conductors such as 16 and 17 which inter-connect the components, and constitute the nodes of the circuit. The medium 10 may be placed electrically proximate the surface 12 by the intermediary of an interface element 18 which serves to transfer to its upper surface 100 (in the arrangement of Figure 1) the electrical potentials that appear at its lower surface 19.

In use, the lower surface 1~ is in contact with the printed circuit board, while the upper surface 100 is in contact with the medium 10.
',:" '' "~' .

Light from a source 101 may be directed at and received ~.
from any region of the board surface 12 via positioning ;
means 102, received light being directed toward a detector 103 by virtue of a beam splitter 104 such as a semi-transparent plate. ; `

The detector 103 is sensitive to changes in the optical characteristics of the medium 10 which are induced by electric field variations in the neighbourhood of the region of incidence of light on the medium 10, in such a way as to produce an analogue of this field, adopting the ;~
form of an electrical signal at the output 105 of this detector, this signal therefore being also representative - 12 - - ~ 33~360 of any electrical signal which is present on the conductor the potential of which has been transmitted to the electro-optical medium, in the region of incidence of the light, by the interface member 18.

The board 11 has lengthwise connectors, such as 106 and 107 which are printed on the board and which constitute the external nodes of the circuit, with which it is possible to establish a physical connection via appropriate female connectors of standard type, the use of which is well established in prior art functional testers. A
support or fixture g3, shown schematically, allows manipulation and/or support of the board during a test.

A test pattern may be supplied to exercise the circuit via these external nodes, sufficient to cause a response at the node currently being examined electro-optically, producing an analogue of the response at output 105.

Based upon an analysis of the circuit to be tested, as is the practice in the A~E art, the output expected for the applied test pattern may be predicted in advance, and the prediction 108 used as a basis for comparison, in a ~ i comparator 109, with the obtained output 105. If the result of the comparison is that the produced and predicted responses differ, comparator 109 provides an indication of a test failure at its output 110.

It will be appreciated that Figure 1 is a sectional view taken through a circuit carrying boàrd. To assist the clarity of description, a plan view of the same board, with its conductors (hereina~ter called tracks) and an electrical component is shown (Figure 2), Figure 1 corresponding to a section being taken at II-II' of Figure 2. Typically, a circuit board might be 300mm by 200mm or - 13 ~ 1 3 3 0 3 6 0 more, the electro-optical medium 10 and the relaying member 18 having substantially the same dimensions. The electronic devices such as 15, which are mounted on the board are most commonly digital components in a dual in line package; one part of the track networX interconnecting these components is shown in Figure 2; to the extent that the components are inserted in the lower face of board 11 (in the arrangement of Figure 1), the component 15, not directly visible, is shown in ghosted marking in Figure 2, with the exception of its connection pins which, spanning the board to be soldered to the tracks carried by the surface 12 are shown directly in Figure 2.

The solder pads betw~en the ends of the pins of the components and the tracks constitute important discontinuities in the sur~ace of the board conductor carrying surface. In ~inished boards, this surface is completely covered with a protective lacquer which forms an intermediate layer of some micro-meters thickness which is between the board conductors and an external test probe.
Moreover, these conductors are often in practice covered with oxide which forms during manufacture and storage.
This oxide itself is a substance whose existence must be taken into consideration in realising industrial testers.

Lastly, frequently modern printed circuit boards also carry components on the face carrying the conductive track network. Nowadays boards carry tracks and components on both faces.
: :
In use, the interface member 18 is placed in contact with the surface 12 carrying the conductors, as partially shown in Figure 2. This interface member 18 includes a plurality of conductive columns, such as column 111, arranged in a matrix such that the potential of each conductor may be relayed to the electro-optical medium 10 (also shown in ~ . : . , ,: :: , ~ : 1: , - 14 ~ ~ 3 3 o 3 6 0 part), via at least one of the conductive columns. The nature and operation of the interface member 18 will be described later.

It will be noted that each of the columns corresponds to a separately observable region of the circuit and that light may be directed at each of them, and received by reflection from each of them. The light is constituted by a polarised light beam from a laser, and the analysis of the reflected light received consists of detection of the rotation of the plane of polarisation, for example by means of a polarimetric assembly including a Wollaston prism separating received light into two detectable beams. The detector in fact comprises two detectors, so as to permit measurement of the difference in intensity between the two beams, and compensation of intensity variations in the source.
'"
The electro-optical medium may be constituted by a crystal. In an alternative, the electro-optical material may be constituted by a polymer film endowed with electro-optical properties. This film may be used in place of an electro-optical crystal layer in the ATE of Figure 1. It may also be preformed to fit between components of the board to be tested in direct contact with its conductors without recourse to an interface member or only such a member of reduced thickness. As an alternative, the film may be placed directly on the conductor carrying surface before mounting the components and of which the face in contact with the conductors has, intrinsically or from a serigraphic deposition of metallic micro-particles, a degree of light reflectance, for example, the film may be integrated with the printed circuit board as it is made. Such a structure, which allows the interface member to be dispensed with, leads to the making of electronic boards able, by virtue of their - - 15 ~ 1 3 3 0 3 60 fabrication, to be subject to a test benefiting from all the advantages of the invention.

The mechanism by which signals may be detected by an electro-optical effect will now be described in more detail (Figure 3), more particularly with re~erence to an embodiment of the invention making use of an interface member. Tne essential characteristics described being however the samP in the case of use of a polymer film.
.. .. ..
Output light from a laser 30 (for example an HeNe laser of wavelength 633mm) is linearly polarised by a polariser 31 and concentrated, by a lens 32, onto an acousto-optical deflector 33, so as to be directed, via a lens 34, toward a point of inspection 35 of a reflective surface 36 of an electro-optical crystal 37. Generally a medium power continuous laser is used, typically 5 to lOOmw. -~

An acousto-optical deflector 33, known in itself, is controlled by a voltage signal continuously output by a -control device of the ATE (not shown), so as to deflect its received light to any point on the electro~optical medium, examination of which is desired. -In a preferred embodiment, two acousto-optical de.flectors may be placed in series. A first control signal voltage then controls beam deflection along a first direction, corresponding for example to one of the dimensions of the board to be tested, and the other voltage signal controls deflection along a second direction, preferably perpendicular to the first and corresponding for example to the other board dimension.

Such use of two deflectors allows reduction in the time required to move from one monitoring point 35 to another to ~ -a time of the order of some microseconds at most.

: :

.. ~ . . .. . . .. . ................ . . . . . . . . . . . . . . . . .

-- - 16 - 1 3303hO

The reflective surface 36 is isolating, at least in the sense that an electrical potential applied to one point on the surface is not propagated to others. Thus, if the reflective qualities of this surface are obtained by a metallic deposition, intrinsically a conductor, this does not consist of a uniform layer, but a matrix of reflective particles deposited separately and not interconnected, each of which is however connected to the region of a point to be tested, such as a test point 38 of the board 39, via interface number 300 (directly in the case of a polymer film). In accordance with one possible embodiment, the reflective surface 36 may be foxmed of an intrinsically isolating material, for example a thin layer of a dielectric material such as an alternating structure of layers of titanium oxide Tio2 and silicon oxide SiO2.

A transparent electrode 301, for example constituted by a deposition of ~old or aluminium is deposited as a layer on the first face, that struck by the incident light, of the electro-optical crystal 37, and this electrode i5 kept at ground potential as a reference; thus, if the electrical potential of the test point under inspection differs form this reference, the polarisation of the reflected light is, by reason of the electric field applied across the thickness of the electro-optical crystal 37, different to that of the incide.nt light, and this difference is detected.

For a cubic crystal structure, such as Bismuth Germanium Oxide Bi4 Ge3 012, the crystal axis (100~ of which is optically orientated normal to a quarter wave plate 302, the phase shift of the light due the sustained electro-optical effect is proportional to the potenkial difference between the faces of the crystal 37 at the monitoring point and independent of the electric ~ield ~. -, . .. . .

- 17 -' 1$303~0 distribution in the crystal.

The reflected light is reflected toward the quarter wave plate 302 via a splitter 303; downstream of the plate 302, the light is directed, by a lens 304, towards a Wollaston analyser 305, from which stem two beams reaching respective photo-electric detectors 308 and 309 which produce respective electrical signals I1 and I2 at their outputs 306 and 307.

This type of polarimetric detection gives the following relationships:-Il = I.(l+m) andI2 = I.(1-m), Wherein I is an intensity proportional to the luminous intensity of the laser 30 and in which m is the phase lag detected, expressed in radians and implicitly small.

A differential amplifier 310 receiving the signals of outputs 306 and 307 provides at its output 311 a signal I1-I2, thereby equal to 2Im. In the case where, during a test, the overall luminous intensity variations of the laser 30 correspond to frequencies very much lower than the frequency of variation of m, the value m ma~ be directly obtained by adequate filtering of the electrical signal available at the output 311 of the amplifier 310. As an alternative, a signal I1 + I2 may be used to regulate the luminous intensity of the source.

For a crystal of non-cubic structure, suah as that of Lithium Nobilium Oxide Li N6 03, the electro-optical effect is also proportional to the potential difference between the faces of the electro-optical crystal. The use of such a crystal however requires some care in the case where the coefficient of proportionality between the potential difference and the effect obtained depends upon the ~' ~ "

- 18 ~ 1 3 3 0 3 6 U

orientation of the crystal cut.

The desirable properties of the electro-optical crystal are a low absorption, a low diffus~on, and a low circular birefringence and good linearityO Where a highly birefringent material is used, such as Lithium Nobilium Oxide for example, a small variation in the angle of incidence of the light leads to a variation sensitive to static phase. Thermal changes causing crystal thlckness variations have similar effects, and both must be avoided.

To improve the performance of such a birefringent material, control of the polarisation of the incident light may, in accordance with the present invention, be provided, for example in response to angle of incidence or temperature.
Additionally or alternatively, the apparatus of the invention may include, relative to the electro-optical medium, a crystal structure presenting perpendicular orentations, but thicknesses substantially identical, in such a way as to attenuate or cancel tha birefringence.

The electro-optical medium advantageously presents a resistivity at least of the order of 10 ohm.cm, especially for testing low frequency circuits, and a di-elecric constant at most of the order of 100 so as to introduce only a low capacitance (lpf or less).

Internal to the interface member 18 (Figure 4), a conductive column 111 is sunk into an isolating ~lexible substrate 112. Each column, such as column 111, is for example cylindrical. The substrate 112 contains an assembly of other mutually parallel columns each spaced lengthwise and extending widthwise along the interface member 18.

In use, this member is placed in contact with the conductor - 19 - 1 33 ~ ~ 5 bearing surface 12 of a print~d circuit board 11 such that a lower surface 113 of the column 111 is proximate a conductor 114 on the surface 12; surfaces of other columns similarly being proximate other conductors. The electrical potential in the conductor of column 111, may be monitored at its upper surface 115.

A conductive film 116 is applied to upper surface 117, excepting the regions of the upper surfaces of the conductive columns, which may be aarthed to provide a sharp step in potential at the upper surfaces of the columns for examination of the electro-optical effect as herein before described.

It will be appreciated that since no current is required to pass along the conductive column 111, a high resistance can be tolerated in the region of proximity to the conductor 114. Thus, the apparatus may be used to examine a circuit mounted on a board, such as board 11, to which a film 118 of an insulating protective lacguer has besn applied. This is a particularly important advantage of ATE
in accordance with the present invention. It allows boards to be tested after all production steps, including lacquer application, have been completed. With prior art device testers lacquer cannot be applied as the nails have to make contact with the conductors directly: even with functional testing, lacquer must be penetrated if the manual probe is to be used. With tha present invention, boards in finished condition may be tested.
. .
When the conductor bearing surface of the assembled circuit to be tested is uneven, as is the case for the ~oard 5 represented in Figure 5, the corresponding surface in the interface member is profiled, moulded or fretted to a suitable shape. The board 50 has for example a circuit using surface mounted components, (such as 52) the - 20 ~ 3 3 ~ 3 6 0 connection pins of which are directly affixed to the conductive tracXs 53 without penetrating the board 50. In such an arrangement, a plane conductive surface may not be present, thus the interface member 51 is adapted to receive the surface mount components, to maintain a plane surface proximate the elsctro-optical medium 54.

In an embodiment of a tester in accordance with the ~i invention represented in Figure 9, a mechanical deflection as well as an acousto-optical deflection is used to extend the scanned circuit surface. ~s in the example of Figura lO a laser 430 emits the test beam at an acousto-optical deflection apparatus 433 via a polariser 431 and a lens 432. The apparatus 433 comprises two acousto-optical deflectors for example of the types made by AUTOMATES ET AUTOMATISMES, l9, rue de Paris 78460 CHEVREUSE
France.

The first deflector is mounted inside a housing 433 to deflect the laser beam in a first direction a and the second deflector deflects the beam so deviated in a perpendicular direction b, such that the two deflector combination allows the sweeping of a square surace of 50x50 millimetres at a distance of 1000 millimetres from the output of the lens 432 (corresponding to its focal length). -The beam emerging from the apparatus 433 falls on a pivoting mirror 441 of a first mechanical deflector 443 which reflects it onto a pivoting mirror 442 of a second mechanical deflector 444. The combination of the displacements of deflectors 443 and 444 in two directions A
and B parallel to 3 and b allows the emergent beam 445 to sweep a rectangular surface of 500 x 500 millimetres in the focal plane of the lens.

- 21 ~~ 1 330360 The beam 445 falls on the convex face 449 of a plano-convex lens of which thQ dimensions are 500 x 500 millimetres, made for example in BK7 glass.

A rectangular layer of a mosaic of BGO ~Bismuth Germanium Oxide, Bi~ G33 012) crystals is affixed to the plane face of the lens. The dimensions of the lens 450 and the electro-optical crystal layer 451 substantially correspond to the total area to be swept by the beam 445 exiting the system of deflectors 443, 444.

The BGO layer 451 has a thickness in the region of one (1) millimetre. In this example, the mating of the BGO
layer 451 to the plane face of the lens 450 allows meshanical vibrations which occur naturally in the crystal as a result of piezo-electric resonance phenomena to be annulled. Such vibrations would be evident by virtue of the parasitic optical signals that they would cause by photo-elastic effects in the crystal.

The laser 430, polariser 431, acousto-optic and mechanical deflectors 433, 443, 444 and also lens 450 with the BGO
layers 451, assembly is integrated with the optical recovery and detection system for the reflected light by the layer 451. This assembly, not illustrated in Figure 9, is shown as 319 in Figure 3 and comprises the splitter 303, the quarter wave plate 302 on lens 304, the WOLLASTON Prism 305, both photoelectric detectors 308 and 309 and the differential amplifier 310 forming an optical probe incorporated in the ATE head.

In usa, an interface member 462 analogous to the interface member 18 (of Figure 1) having anisotropic conductive properties is firstly placed against a face to be tested of a printed circuit board 460 (Fi~ure 9) to form an analogue of the card 460 conductor voltages in contact with one of - 22 ~ ` 1 3 3 0 3 6 0 its faces at its other face 463. This face 463 and the free face of the electro-optical crystal 451 at the heart of the optical probe are in contact for the test and ~irmly urged together by pressure means not represented allowing the elimination or minimising of all parasitic spacing between the contacting faces of the BGO layer 451, the interface member 462 and the board to be testedO In this example an elastomer sheet made by JSC TECHNIC in the Federal Republic of Germany and sold under the name 'ZEBRA
bidimensionnel' has been used for the interface member; its thickness being between 0.1 and 5 millimetres for example in accordance with the board type and its surface discontinuities.

In Figure 9, the checker region 470 corresponds to an elementary surface of acousto-optical scanning of the crystal of 50 x 50 mm. By virtue of a displacement by the mechanical deflection system, acousto-optical scanning of one hundred (100) such elementary surfaces next to each other over the surface of the electro-optical crystal 451 is possible.

Typically, the acousto-optical scanning rate which may be obtained is 100 KHz (the frequency of getting form one test point to another). The mechanical scan allows a change form one elementary region 470 to another in 50 milliseconds or so.
:
Considering Figure 9 a rectilinearly polarised light beam 445 at the input of the crystal layer 451, its polarisation remains unchanged while crossing the layer if the conductor of the board to be tested in the region of the point of incidence of the beam has no voltage load. The application of a voltage creates a phase shift p(u) between the two components of the electric field of the light. The state of beam polarisation which passes back through the lens 450 ''.` ~ ' '' ' "' ' ' - . "" ": ' ' ' : ' ' . ~ ' ' : ` 1 ' ' ' - . . ' ' ' . .: ' ' ' ' ` ' . ' ' ' . ' ' - 23 ~ ' 1~3 03 6a at the output of the BGO layer 451 is then elliptical.

The emergent beam is redirected by the splitter 303 (Figure 3) toward a detection lens 304 then is divided into two components by the Wollaston Prism 305 and the intensities of these components are represented by Il = 1/2 Io (1 + cos (p + Po)) I2 = 1/2 Io (1 - cos (p + Po)) in which Io is the incidence intensity and Po is the static phase lag introduced by the ~uarter wave plate 302.

The differential amplifier 310 gives a signal S

S - Il - I2 = Io (P + Po) If the phase shift p is zero, the signal s is negligible in many cases due to the small amount of polarisation variation produced in the layer 451. ~ `~

By way of example, for a BGO crystal of index 2 and of electro-optical coefficient equal to 1 pm/v, for a laser wavelength of 647 mm (Krypton laser) a value of P =
8.10 4 rad/v. is obtained.

In response, with a quarter wave plate, Po = pi/2 and S = Io x P `
the signal S varies linearily with V and values of P of the order of 10 4 radians may be straightforwardly detected.
By virtue of the differential method adopted, this is true even if laser source intensity is subject to relatively slow fluctuations in time (up to about 10 per cent).

The deformations caused by fixing the BGO crystals may give rise to parasitic birefingences which are different ~rom one point to another. As a result Po varies from one point to another. Pl is biased to return to P1/2 after each ;~
;;'~ ~' ' . '''';"`

,' '~.~''". ;`'' - - 24 - ~ 30360 positioning of the beam and before electrical measurements by inserting a Xerr cell 480 between the quarter wave plate 302 and the Wollaston Prism. This apparatus is constituted by a plate of electro-optical material placed betwe2n two transparent electrodes in a variable electric field supplied by a feedback loop 482 comprising a switch 484 and a variable amplifier 486 from the output 311 of the differential amplifier 310. The phase shift introduced by the plate ~80 between the beam field components which pass through it then makes adjustment to the level reguired to cancel the background component of the signal S
corresponding to the point tested. With test signals of a suitably high frequency, the high frequency component of the signal S contains the desired test information.

The modulation depth of the arrangement may be further improved by introducing an imperfect polariser 340 between the crystal layer output and the quartar wave plate 302.

In Figure 10 this polariser is a cube 340 intercepting tha beam 335 leaving the lens 34 of Figure 10 and used in place of splitter 303.

A system of axes 350 represents the polarisation of light emerging from the lens, P being the angle of polarisation with respect to the initial direction of linear polarisation along the Y axis. The x and y components of sector 352, representing the polarised wave of the beam 33 5, represent the maior axes of the elipse of polarisation, the Wollaston Prism 305 (Figure 3) being orientated so as to split the light components along these two axes.

The splitter cube 340 separates the beam 335 into two beams, one transmitted 334, the other reflected 336. The interface 355 of the cube is arranged such that it behaves ~ 25 l 330360 72722-19 as a bad polariser for the reflected light and increases its ellipticity of polarisation. The component along the Y axis is greatly reduced as against the ~ component as is shown by the diayram 360 of Figure 10.
sy contrast for the transmitted portion of the beam 334 this is linearly polarised along Y ~diagram 370).
The components Il and I2 received after separation of the beam 336 by the Wollaston Prism are expressed as follows:
I1 = (I~Ap) Io~2A2 I2 = (I+Ap) Io/2A2 ` ~:-A is a coefficient greater than 1 given by the characteristics of the polariser cube ~40. These relationships show that a more powerful laser may be used to increase the modulated depth without saturating the detectors at the output of the Wollaston Prlsm.
The interfaces between the printed circuit board 460r . : :
the interface member 462 and the crystal layer 451 have been shown schematically in Figure 11. Two conductors 467 and 468 forming two test points are indicated one under a voltage V1~ the o~her at ground potential. The conductors are often covered in oxide.
When the board is completed, they are covered with a lacquer. The layers of oxide or lacquer form a spaclng 469 between the upper ` :
face of the board and the opposing face of the interface member 462. .~
An electrical equivalent circuit diagram of the ~ .
. .
arrangement is shown in Figure 11. By virtue of the capacitances of the crystal~ of the conductor and of the spacing 469~ the voltage V2 which reaches the crystal in response to V1 is all the weaker as the capacitance of the spacing 469 is :~ ~
.

26 - ~ 1 330360 lower and that of the crystal is greater. Moreover, for a given interface member 462, the thickness of which is determined by the components mounted on the surface of the card 460, the more the interval 469 is made larger and the more the crosstalk, represented by the ratio V3/V2 is increased.

In one example, the parameters were as follows:-- thickness of the interface member 462 e = 2mm - thickness of the BGO crystal 451 = lmm - dielectric constant of the crystal e'= 16 - distance between the two test points d = 400 micros.

The maximum allowable crosstalk being about 10%, the height h of the spacing 469 must not exceed 1.5 microns.

It is therefore desirable to deposit an electro-optical material having a dielectric constant as low as possible which at the same time presents an electro-optical co-efficient as high as possible.

The relationship between the electro-optical sensitivity index of the crystal n3r (n baing the refractive index of the medium and r its electro-optical coefficient) and the dielectric constant e' constitutes a yardstick (figure of merit) for the selection of material for use as the electro-optical layer 451.

The table below gives an indication of this relationship for different crystals.

Crystal rn3/e' Av a il a b le - 27 - 1 3 3 0 3 ~

Dimensions Origin Cm2 .

MNA 50 1 France ZnTe 10.8 1 USA
AsGa 3 100 France CuCl 1.9 1 France LiNbO3 1.6 10 USA
Bil2Sio20 1.2 12 Japan Bi4Ge3012 0.5 100 France KNbO3 0.5 2 France With the apparatus with the BGO layer as envisagPd above, reliable detection of test bits of 1 volt amplitude at a frequency of 5 MHz applied to test points 2mm apart and separated by a ground conductor has been obtained. The thickness of the elastomer interface member was 2mm, the contact surfaces without oxide and covered by a lac~uer layer of 5 microns thickness.

Consideration of the preceeding table shows that an organic composition such as MNA (2 msthyl -4 - nitroaniline) shows an increased figure of merit making this type of substance most interesting for the envisaged applications.
Such a substance may be used not only in crystalline form, but also in a form combined with a support material in which the molecules of such an active electro-optical substance are incorporated. As a support material perspex (PMMA - poly methyl methacrylate) may thus be used with a density of MNA about 15%. The MNA is therefore then used as a dopant having molecules which are retained in the PMMA
matrix, to make the composition electro-optical. Other possible dopants possessing electro-optical properties are for example the following~
~. '~' '' ' ~:

DAN [4- (N,N-dimethylamino)-3- acetamidomitrobenzene]
COANP [ 2-cyclo-octylamino -5- nitropyridine]
PAN [4-N-pyrrolydino -3- acetaminomitrobenzene]
MBANP [ 2 - ( alpha-methylbenzylanino) -5-nitropyridine ]

This type OI polymer substance is the topic of many development programmes currently being undertaken by several enterprises, universities and research centres in the f ield such as the Lockheed Missiles and Space Company , Inc ., Hoechst Celaneses Corporation and other large houses in the chemical field. tSee for example the conference proceedings of the Symposium Entitled NATO
Advanced Workshop: "Polymers ~or non linear optics", Sophia Antipolis June 19-24 1988. ) The polymers obtained may be used to form films, fibres or thin layers or thicknesses on a known substrate.
'' A particularly interesting characteristic of these polymers is the potential to use them in the form of films capable of being produced in large quantities and at reasonable cost. These films may be used in thicknesses of some microns (for example 10 microns) on transparent ~;upports such as glass . They may also be del ivered directly in the form of resilient films made up o~ several adjacent layers of elementary ~ilms (for example 500 microns thick).
After drying, the molecules of the electro-optically active ingredient are captured in a support material in the amorphous state without particular orientation. To make the material electro-optic it is necessary to heat to a temperature suf f icient to allow the active molecules to regain a certain mobility inside the matrix. The value o~
this transition temperature may vary with the polymer but may typically be around 100 to 120C. In this state they ~ -may be subj ected to orientation with respect to the support - 29 - ~ 3 3 0 3 6 0 under the effect of an electric field. The molecules tend to orientate thamselves in the direction of the exciting field. The stronger the exciting field, the higher the proportion of active molecules orientated in the direction of the field.

When the temperature is again lowered whilst the molecules are oriented under the effect of the field, they keep this orientation. The material thus keeps an orientated structure which may be shown by an anisotropic optical behaviour (Pockels effect) in the presence of an electric field. Thus when the material is struck by linearly polarised incident light in the absence of an electric field the material does not give any change in polarisation.

-By contrast, in the presence of an electric field, the transmitted light sustains an elliptical polarisation linearly related to the intensity of the applied field.

In Figure 12a the index distribution in an electro-optical film 500 realised in a polymer of the type previously described, the molecules of which have been orientated by the application o~ an electric field, called an aligning field, in a direction normal to its surface whilst the temperature was lowered below the transition region beyond which the electro-optical dopant molecules loose their mobility is shown. An ellipsoid represented as 502 describes the refractive index distribution of the material whilst an electric field (called a detecting ~`ield) perpendicular to its plane is applied. This ellipsoid shows the spatial refractive index variations of the material. It has a symmetry of revolution with respect to the normal 504 in the plane of the film ~00 which re~lects the fact that the material is optically isotropic parallel to the plane of the film.

_ 30 _~ l 33~3~a I~ the film 500 is illuminated with a polarised incident beam 506 normal to the film, the light recovered from the film (by transmission or following re~lection from the opposite face 508 of the film) will not show a changed state of polarisation. This explains that when such a film is used for example in place of the BGO layer 451 of Figure 9 voltages at the scanned circuit node cannot be detected. On the other hand an incident beam 510 angled with respect to the plane of the film will have its state of polarisation altered.

It is necessary therefore in this case to use a polarised incident light beam angled with respect to the electo-optical layer in order to reveal the Pockels effect created by the voltages in the tested circuit.

In Figure 12b the ellipsoid of indices obtained with the same material under the action of a detecting electric field, normal to the plane of the film, is shown when the molecul~s of the film 500 have been previously orientated in the plane of the film by an aligning electric field. In this case the ellipsoid of the indices 512 has a symmetry of revolution with respect to an axis 513 in the plane of the film 500. A normal incident beam 506 then has its polarisation changed as a function of the difference in indices r2 and r3 along the principal axes of the ellipse in the plane of the film. Under these conditions the assembly of Figure 9 (that is with normal incidence) with a polarimetric analysis apparatus such as represented in Figure 3 allows the exploitation of the Pockels effect created by the circuit to be tested in the polymer ~ilm incorporated in the optical probe or the board.
An apparatus for orientating the molecules of the active composition in the plane of the polymer film is illustrated in Figures 17 and 18 schematically. A strip of polymer 800 is placed in a tunnel oven 801 heated to a temperature T of about lOOAC ~slightly higher than the transition temperature described above.) The strip is guided in the interface between two opposing metal plates 802 and 804 which are maintained at a continuous identical electric potential + V. At the output 805 of the tunnel oven, the plate enters a second tunnel 810 defined by two opposing metal plates 812 and 814 which are maintained for example at zero potential. The result is that the spacing 815 between the heating tunnel exit 801 and the second tunnel 810 is subject to an electric field substantially uniform and parallal in the plane of the strip 800 which lays in this space 815. The electro-optically active molecules of the film tend to orientate themselves parallel to this direction under the local effect of the field at the exit 805 o~ the tunnel 801. Two nozzles 818 and 819 to either side of the film 800 open in the space 815 where they deliver a flow 820 of inert gas (for example argon or a sulphur hexafloride SF6) cooled to -40C to both faces of the film 800. The gas is exhausted from between the plates of the second tunnel by means not shown. The temperature of the film 800 traverses the transition region within the space 815. The active molecules maintain a preferred orientation in the plane of the film which cools in the second tunnel.

The preceding data concerning the orientation of the film structure and the incidence of the test light permits use of the electro-optical polymer film 500 for the testing of the signals in a circuit by the polarimetric method.

Alternatively, and in accordance with another aspect of the invention, a Fabry-Perot interferometric method may be adopted in place of polarimetry to detect the Pockels effect under the influence of the voltages to be tested.

- - 32 ~` 1 3 3 0 3 6 0 In Figure 19 part of a circuit board 840, typical of an assembled circuit, is shown comprising a support 841 of isolating material on a face of which several el~ctronic components are mounted: these are interconnected by conductive links such a 848.

A layer or plate of electo-optical material 842 is placed proximate the circuit supporting board 841 in contact with the upper surface of the conductor 848. It receives, given normal incidence, a laser beam emitted by a laser source 844. The surface of the film 842 facing the laser source is covered with a layer 845 of conductive and semi-transparant material which forms a semi-transparant mirror of average value coefficient of reflection. The other face of the film 842 (opposite the circuit board to be tested) is covered with a layer 846 of an electrically conductive and high coefficient of reflection material, for example a layer of aluminium thicker than the layer 845 so as to reflect a substantial part of the light which reaches it, the remainder being absorbed. A layer 847 is provided to absorb that part of the light transmitted by the reflecting face 846. The layer 846 may be formed of particles or fragments isolated one from another to avoid creating short circuits between neighbouring conductive tracks on the board to be tested 840.

Interference is established between the light reflected by the parallel mirrors 845 and 8~7; the light beam emanating from the film (shown in the Figure with a double arrow) is defected by a semi-transparent mirror :851 and superimposed on the common path of the emitted beam 843 and the output beam 849 toward optical analysis means 850 for measuring the intensity of the light signal 849. As is known, the interference phenomenon is a function of the distance separating the mirrors, that is the thickness (e) of the ~ ' '. ,' .

~ 33 ~ ~ 3 3 o ~ ~ O

film, the wavelength (lambda) and the refractive index (a) of the material of the film. The index "n" varies with the electric field to which the film is subject, itself a function of the potential V which it is desired to measure.
.
Therefore, measurement of the luminous intensity of the signal at the receiver 850 allows a measurement of the potential V at a given point on the circuit 800 to be obtained. It will be noted in particular that if the layer of electro-optical material 842 is a polymer film of the type described above (and which could be integrated with the board 840) the measurement of the Pockels effect is possible with perpendicular incidence with this interferometric technique, even when the electro-optical molecules of the film have been originally orientated perpendicular to its plane.
, ~:
In order to alleviate the consequences of thickness differences in the electro-optical film or layer, the light source is adapted to emit a variable wavelength light beam. This means that the operating point of the apparatus may be shifted on its intensity characteristic curve (Figure 2 0) , of intensity as a function of PHl = [ 2pi/lamba]n.e from a point A situated on a first part of the curve where I = Imax whatever PHI (therefore of zero sensitivity) to a point B situated on a second portion of the curve where I (being between Imax and 0) varies greatly as a function of PHI; the sensitivity is a function of the variation in I on the second portion o~ the curve about the point B; preferably point B corresponds to about Imax/2. When point B is fixed, the measurements are then affected by the wavelength corresponding to the point B.
.' In Figure 13, a printed circuit board 600 includes a substrate 602 of a material traditionally composed of epoxy and glass fibre which comprises on each of its faces 603 ~'''.

. . - . . : . . . ~ . , ~

_ 34 -- ~ 330360 and 604 metallisations or conductive tracks such as 605 and 606, forming a network to which the components mounted on the board are connected.

The face 604 of the board is covered with a layer of polymer electro-optical film 608 which extends over substantially all the sur~ace above the metallic tracks 606 which are formed there. The film 608 is incorporated on the card after the formation of the metallisations, but before the placement of components, for example by bonding. One side of the face 604 is covered with a checker-board pattern of elementary mirrors by means of a layer of reflecting aluminium 609. The size of the particles or pieces of the pattern is such that each piece cannot give rise to a short circuit between two nei~hbouring metallic tracks. On the opposing face, the film 608 is covered with another layer of aluminium 610 of a thickness sufficiently low to be transparent to test light projected at the board from direction 612. The layer 610 constitutes a reference electrode for the electric fields generated in the thickness of the film 608 by the voltages applied to the conductors 606.

The board 600 carries components such as discrete element 614 or integrated circuit package 616 on its surface 603. The components have connection plns 618 which penetrate the board substrate in holes 620 placed at right angles to the metallic tracks 606 on the other surface o the board, to which they are connected via solder pads 622. Also shown is a component 624 mounted on the surface 604 of the board, its two outer faces 625 and 626 being connected to two metallic tracks 606 respectively by two beads of solder 628. Before assembly of the components 614, 616 and 624 openings such as 630 have been formed in the polymer film 608 around the expected solder beads 622 and 628 to avoid any electrical contact between the ;

:~"'':.'.'' ~

- 35 -` 1 3 3 0 3 6 0 metallic solder and the referance electrode 610. These openings may be formed by grinding before placement of the components on the card or be made in the film 608 before its joining to the surface 604 of the card. Alternatively, a film or film elements, praformed as a function of the board regions carrying the conductors to be tested may be placed upon the board before mounting. contact regions such as 632 between the upper surface of the metallisations 606 and the film 608 provide the test points. These may be monitored by projecting an incident laser beam and analysing the beam reflected by the corresponding metallic mirror 609. This analysis is made with a test probe analogous to the apparatus of Figure 9 in which the BGO
layer 451 and the inter~ace layer 462 have been xemoved.
In fact, the light leaving the plano-convex lens 450 strikes the card mounted adjacent its plane face directly.
The beam returned by each point of the tested board 600, changed by the effect of the voltage driving this point is detected and analysed by an arrangement such as that 319 of Figure 3. The closeness of contact between film 608 and the tested conductor 606 ensures excellent optical conservation of the signals to be monitored and good spatial resolution. In certain board arrangements, (Figure 13b) wherein the conductive regions 606A to be examined are by construction placed proximate a ground connected region 606B, an electric field parallel to the plane of the electro-optical film is established (field lines 640, 641 figure 13b). The existence of this field may be directly tested without any opposite polarity electrode that might otharwise be necessary. The Pockels e~ect then appears by virtue of a substantially parallel field in the plane OI
the film, and no longer perpendicular.
- :~
In accordance with another embodiment of the invention, use of a electro-optical polymer film directly in a tester optical probe is envisaged. In fact, the good .... ...... ~ .. .. ..

- 36 - l 3 3 0 3 ~
-characteristics of these materials, what with the high level of their electro-optical coefficient and their low dielectric constant, make them well adapted to this application. They may be made in the form of a layer for example bonded on the plane face of the plano-convex lens 450 (Figure 9) in place of the BG0 crystal layer 451.
. .
In accordance with another advantageous technique, the use for each type of board of an electro-optical transducer specific to that type of board is envisaged. In this connection the fact that the costs associated with polymers of the type indicated are not very great is put to good effect. It is therefore possible to make for each new type of board to be tested a transducer designed to be associated with the optical probe during the test but which is adapted to that form of board.

The transducer 700 (Figures 14 and 15~ is in the form of a plate 702 of glass or a transparent plastics material of low photo-elasticity and straightforwardly workable so as to be able to create interior openings or cavities in which the componsnts, the solder pads and other discontinuities in the upper surface of a printed circuit card 705 are located when the transducer 700 is coupled for the test.

An electro-optical polymer film of the type already described with reference to Figure 12 is deposited onto the face 710 of the transducer 700. Film 712 is covered by an aluminium transparent electrode 714 where it mates with the support plate 702. The lower face of the film 712 is ~ -covered with a reflective layer 716, also in aluminium. As previously, this layer is not continuous, but formed of particles mutually spaced apart such that a particle in ~-contact with a conductor on the surface of the board cannot -~
also contact or have its potential influenced, by a neighbouring conductor. ~ --' ~

1 3303~0 The openings 720, 721, for example are arranged in the transducer 700 to allow components such as 724 to enter or beads of solder such as 725, by which the pins 726 of components 727 passing through the board 705 are fixed to conductors 728 to the surface 730 of the board against which the transducer 700 is placed, to be accommodated.

The plate 702 gives rigidity to the transducer sufficient to allow the electro-optical polymar film 712 to be placed against the conductors such as 728 and 732 at the surface 730 of the board 705 when the board 705 and the transducer 700 are brought together, by manipulation apparatus, not shown. The parasitic capacitances between the sensitive film 712 and the contacts are at a minimum level which allows good spatial resolution to be obtained, for example O.lmm, with a lacquer of 10 microns thickness. The result is substantially better that the resolution possible with a relatively thick interface elastomer such as described with reference to Figures 1 to 3. The reduced thickness of this lacquer allows crosstalk between very close test points to be kept to acceptable values.
~ , The support plate may be made by commonplace mechanical means. The sensitive film (Figure 15) may itself be made by laser. The transducer assembly 700 may be fixed in contact directly with (or immediately adjacent to) the planar face of the field lens (cf. 45, Figure 9) of the optical probe in place of the arrangement formed by the BG0 lacquer 451 and the elastomer interface 462. Alternatively it may be manipulated separately form this lens at the time of inserting the board to be tested. As in the case of an electro-optical board of Figure 13, the film portions or electro-optically sensitive films may be separate and reduced to a predetèrmined number of test points or regions distributed over the board surface. Regions without i`'````; ' ` .,'` " '```` . ' ` "`~ ' ` '`' .`;' '`1 ` " 1~`'. `

mirrors or openings 727 are provided in the transducer 700 to allow the positioning of reference marks 728 from which the beam deflection system is calibrated to precisely direct the incident beam toward the conductors for nodes to be tested in the circuit. The deflection assembly 433, 443, 444 ~Figure 9) control system may then be programmed to selectively interrogate the nodes of the board corresponding to the regions of the transducer which are provided with electro-optical film or material.

Figure 16 shows schematically a sectional view of an electro-optical film 740 provided on a transparent reference electrode 742 on one of its faces. Its other face is covered by a checker pattern of small mirrors 744 formed by a thick deposition of aluminium. To further increase the contrast of measurement, the polymer is etched (chemically for example) between the mirrors to create dips 706 of depth of the order of the required resolution (0.lmm for example). Thus the capacitance between two neighbouring mirrors 744 is reduced. This film may be used in the transducer 700 of Figures 14 and 15.

Over and above the fact, already outlined, that the invention allows completed boards to be tested, that is boards covered with lacquer, another advantage of the invention is that the activity of internal nodes may be observed with a disruption all but negligible with respect to prior art testers. In fact, all contact with a probe, that is an element in which current flows, disrupts the normal operation of the monitored circuit, such that it may, with prior art testers, not be operating in the same way when it is tested and when it is not.

In conventional testers, the maximum speed at which a monitored circuit may function may, in some cases, be limited by the requirements of the test. Thus, tests are - 39 - 1 3 3 0 3 ~

not conducted in the same operational environment as that in which the circuit ought normally to function. This constitutes a major problem with prior art testers, the manual diagnostic probe of which has a high capacitance ~of the order of 100pf). On the other hand, the interface member 18 may only have a capacitance of lpf, thus allowing circuit operation at full speed during test.

The electro-optical material mentioned above producas a substantially linear electro-optical effect at the wavelengths envisaged ~Pockels effect), that is variation of angle of polarisation of a beam crossing these materials under the effect of an electric field in the region of crossing in proportion to this field. Other materials present a quadratic effect (Kerr effect) in accordance with which the variation in the angle of polarisation is proportional to the square of the electric field. This property is put to use in an alternative embodiment of the invention (Figure 6), wherein the light is directed generally and laterally at a quadratic electro-optic medium 60 placed electrically proximate conductors, such as conductor 61, printed on a board 62 of an electric circuit to be tested.

Light is received at an opposite edge face of medium 50 by a detector 63 which produces an output 64 in accordance with the principles already outlined. Output 64 may be used by ATE as herein before described.

Electrodes, such as electrode 65, are deposited upon khe upper surface of the electro-optical medium ~0 along the path of the light. Electrical connections (not shown) allow each electrode to be biased to a chosen potential, for example to zero or to a potential of twenty volts with respect to a reference potential.

~ 40 ~ 1 33 03 60 Under these conditions, the nature of the transmitted light received at detector 63 will be dependent upon electrical potentials appearing across the electro-optical medium. As the circuit below the medium is exercised these electrical potentials will be relayed to the underside of the medium 60 via a relaying member 66. Taking as an example a digital circuit, having logic 0 at 0 volts and logic 1 at 5 volts and assuming that all biasing electrodes 65 are biassed to the 0 volts, either a zero or a 5 volt potential will be applied across the electro-optical medium 60.

The characteristic (Figure 7~ of ths electro-optical medium chosen for this embodiment of the invention, for example a PLZT ceramic composition of the type used for the control of optical gates on laser beams, is a curve 50, defining effect against voltage. `~

Thus for 0 to 5 volts applied potential, the electro-optical effect will be detected between a and b along the ordinate. Assume now that one ~lectrode, say electrode 65, is biassed to a potential of 20v. The effective potential appearing across the medium 60 in the region of the conductor 61 is thus either 20 volts (logic 0) or 15 volts (logic 1). These potentials will produce an effect between c and d.

It will be appreciated that dynamic signals appearing in the region of a biassed electrode may therefore be discriminated from cumulative effects produced in the unbiassed region, the swing c-d being much greater than the swing a-b. Hence by selectively biassing the electrodes, activity in any region along the tract of the light may be examined.

For examining any region of activity, an array of detectors 80 (Figure 8) extending edqewise is required.

, . .. : . :,: . . ~ : : , , - 41 ~ 33036~

Where parts of the embodiment of Figure 6 correspond to those of Figure 8, common reference numerals have been used.

Biassiny electrodes, such as electrode 65 extend laterally stripwise across the medium 60 in a direction substantially perpendicular to the mean direction of light propagation in the medium, such that electical activity in any region may be examined by energising the electrode which biasses that region and by selecting the detector output signal, such as 81, which receives the light which crosses that region.

The way in which ATE in accordance with the present invention may be arranged to operate will now be considered.

In an assembled circuit, such as a printed circuit board, the state of each of the outputs clearly depends upon the previous states of the inputs; for example, the state SK of the Kth output, for a known to be good circuit, is a function FK of the array [E] of the input states, this may be written: SK = FK([E30).
':~' ~ :
The purpose o~ prior art functional testers is to verify that for an array [E] of intput states assigned to the circuit, the state of each output, such as state SK of the Krh, is well founded on the array [E] of input states through application of a function, such as FK, which characterizes correct operation o~ the circuit.

If that is not the case, that is if the state of at least one output, such as the kth, is not correct and, for example, S'K takes the place of SK, the monitored circuit is clearly faulty.

This information, however, provides no help for the repair - 42 ~ 1 3 3 0 3 ~ O

of the circuit, it will also be appreciated that the dependance of SK and [E] is very complex.

Upon further investigation, it will be seen that the state of each of the outputs depends upon the state of the inputs via the intermediary of the states adopted by the internal circuit nodes. For a good circuit, it may be shown that the state SK of the kth output depends, by virtue o~ a function denoted as GK, on the array [E] of the input states, not only directly, but through the intermediary of -the state I1 of a first internal node, the state I2 of a second internal node, etc....,this may be expressed, for the case of a circuit with n internal nodes~

SK = GK (Il, I2,...., In) If, instead of adopting the state SK, the Kth output adopts the state S'K, which maybe because instead of adopting the -state Il, the first node adopts a state I'l, and/or because instead of adopting the state I3 the third node adopts a state I'3, etc...., this gives rise to a series of possibilities, such as:- ~
' ' ;:~ ' S'K = GK (I'l, I2,...., In) or S'K = GK (Il, I2, I'3,.... In) or S'K = GK (I'l, I2, I'3,.... In), etc............................... ;
. .
In general, the a~pearance of an abnormal state SIK may be a priori due to a large number of possible causes, amongst ~
which a prior art functional tester cannot detect the `
actual cause without recourse to long and difficult additional functions. ~;~

By contrast, ATE in accordance with the present invention ~ -may, during diagnosis, that is after a functional problam with the tested circuit has been established, shown by at _ ~ 43 ~ 1 3 3 0 3 ~

least one abnormal result in the examination of the circuit output nodes, be used as follows. First of all an internal node is selected and examined, and all possible or desirable state combinations are applied to the circuit input nodes whilst the corresponding node states are examined and recorded. From the results data obtained, as great a number as possible of possible fault hypotheses are eliminated.

Then another node is chosen and the same procedure repeated. The internal nodes are thus examined one by one until evidence of thP cause of the fault is revealed.

Those skilled in the art will undPrstand that the present invention is not limited to the single embodiment described and put forward above, that is recourse to the prior art technique of backtracking which involves the use of expert systems.

Those skilled in the art will also understand that although the embodiments described make use of but one light beam, use of several beams is with the scope of the present invention.

Lastly, the arrangement which has just been described for circuit testing may be combined with a spectrum analyser of the signals provided by such a circuit. In the system described thus far the signals are recorded in real time and the test takes account only of the presence or absence of pulses at predetermined instants. The pass band of the system, limited by that of the detectors used, may be in the region of 100 MHz for example.

A detailed analysis of waveforms at a much greater frequency may howevèr be obtained by operating as described herein after with reference to Figure 21, in acaordance ~ 44 ~ ~ 330360 with a sampling (stroboscope)technique.

A CW laser of the type already described 901 is arranged to project the test light toward an optical system 902. This system is of the type previously describad. It produces and sweeps an analysis light beam 903 projected in the direction of a circuit 907 to be tested across an electro-optical arrangement comprising for example a field lens and an electro-optical transducer of the type described with reference to Figure 9. The reflected optical signals are divided by a splitter and transmitted to a detector 914.

A second laser 920 of the pulsed type is provided which emits at a predetermined repetition rate pulses which are very short with respect to the duration of signals produced at the circuit nodes in response to test excitations applied to the inputs 922 of the circuit 907.
. ' The light pulses of laser 920 are directed on the one hand to a photodetector 924 and on the other, via a splitter 925, across a variable optical delay line 926, toward a reflector 932 which directs the pulses toward the input of the optical system 902 along the same axis as the light output of the laser 900. The delay linè 926 comprises two mova~le reflectors 927 and 928 allowing two right angle bends in the direction of the beam output of the splitter 925, and a fixed reflector 929 to realign the beam in the direction of reflector 932. The optical delay imposed on the beam between the pulsed laser 920 and the optical system 902 may be modulated by bringing together or moving apart the two reflectors 927 and 928 and the fixed mirror 929 (arrow 930).

The electrical output of the photo detector 924 is connected to synchronisation apparatus 940 which controls 45 _ 1 330360 the test signal generator 942 connected to the inputs 922 of the circuit board 907 such that the electric pulses are applied to the inputs with a repetition rate equal to that of the pulsed laser.
.
The duration OI the pulses o~ laser 920 is very æhort compared with the pulses at the inputs 22. The signal provided by the detection system 914, of low bandwidth compared with the length of the laser pulses, is integrated over a large number of samples. Each point of a waveform or a circuit node may be examined by a suitable adjustment of the optical path between the pulsed laser 920 and the optical system 902.

Alternatively, an electrical adjustment of the synchronisation pulses may be made. In use in testing the continuous laser 900 emits: the pulsed laser is off. For a detailed timing analysis, the laser 900 is switched ofî and the pulsed laser 920 excited.

.::

Claims (48)

1. Apparatus for circuit testing, such as a printed circuit board, the circuit comprising a plurality of electronic components arranged on an isolating support and interconnected by a network of conductors formed on at least one surface of the support, the apparatus including a layer of an electro-optical medium of dimensions substantially equal to those of the circuit placed, in use, electrically proximate a conductor bearing surface such that an analysis of changes in incident light under the effect of voltages appearing-in said conductors as a response to the application of test signals to the tested circuit may be made, the apparatus further including means to direct the light toward any region of the electro-optical medium and for receiving light emanating from this region, such that the electro-optical effects appearing in the medium may be detected, and means for applying a pattern of electrical test signals to one or more external nodes such that a response signal is produced at at least one circuit node, the light being directable toward a region of the electro-optical medium which is electrically proximate a conductor constituting the node, so as to produce an analogue.

2. Apparatus in accordance with claim 1 and in which the light maybe generally directed toward an electro-optical medium, or toward a portion thereof, the electro-optical medium comprising a plurality of distinct regions, subjectable to electrical bias, along the path of the light, these distinct regions being excit-able by polarisation such that the light detected at the output is representative of a single region of the medium.

3. Apparatus in accordance with claim 2 and in which the light is detected by a series of detectors each having its own output.

4. Apparatus in accordance with claim 1 including means to compare said analogue with the response that a known to be good circuit would have and to provide an output signal represent-ing the result of the comparison.

5. Apparatus in accordance with claim 1 and in which the electro-optical layer has a structure such that it is sensitive to an electric field component which is perpendicular to its plane.

6. Apparatus in accordance with claim 1 arranged to operate as part of a functional tester, in a phase of the test in which only the external inputs and outputs of the circuit are excited and monitored, useable in the case of a fault established during the test phase, in a diagnostic phase in which a large number of nodes, including internal nodes, are monitored.

7. Apparatus in accordance with claim 1 and in which the nodes are examined one by one, a set of test patterns being applied at each examination.

8. Apparatus in accordance with claim 7 and in which all possible test patterns affecting a node during an examination are applied consecutively.

9. Apparatus in accordance with claim 1 and in which the electro-optical medium is placed electrically proximate the conductor bearing surface through the agency of an interface member relaying to the electro-optical medium the electrical potentials present on said surface.

10. Apparatus in accordance with claim 9 and in which said interface member comprises a plurality of mutually isolated substantially parallel columns.

11. Apparatus in accordance with claim 9 in which the interface member has a profiled surface to mate with the contour of an assembled circuit.

12. Apparatus in accordance with claim 1 and in which the electro-optical medium is formed by a polymer film endowed with electro-optical properties and applied directly to the conductors before the mounting of components.

13. Apparatus in accordance with claim 1 and in which the means for directing light toward any region of the electro-optical medium includes two electro-acoustic deflectors mounted in series.

14. Apparatus in accordance with claim 1 and in which the electro-optical medium has a reflective surface formed by a set of metallic mirrors mutually electrically separate.

15. Apparatus in accordance with claim 1 including means for imposing polarisation on the incident light.

16. Apparatus in accordance with claim 1 and wherein the electro-optical medium includes substantially identical crystal thicknesses presenting mutually perpendicular orientations.

17. Apparatus in accordance with claim 6 arranged, in a diagnostic phase, to backtrack.

18. Apparatus in accordance with claim 1 and in which the light includes a set of individually directable beams.

19. Apparatus in accordance with claim 1 characterized in that the face of the electro-optical layer opposite its face electrically proximate the conductors is provided with a conduc-tive layer connectable to a reference potential.

20. Apparatus in accordance with claim 1 characterized in that the face of the electro-optical layer electrically proximate the conductor bearing surface is provided with at least a covering reflecting the incident light having passed through the electro-optical layer.

21. Apparatus in accordance with claim 1 characterized in that the electro-optical layer is constituted by an electro-optical polymer film.

22. Apparatus in accordance with claim 21 characterized in that the electro-optical molecules of the polymer film are orientated along a preferred direction perpendicular to the plane of the film.

23. Apparatus in accordance with claim 21 characterized in that the electro-optical molecules of the polymer film are orientated along a preferred direction parallel to the plane of the film.

24. Apparatus in accordance with claim 21 characterized in that the polymer film is permanently associated with the surface of the circuit support.

25. Apparatus in accordance with claim 21 characterized in that the polymer film is adapted to the configuration of the circuit to be tested and is applied to the surface of the support after the mounting of electronic components.

26. Apparatus in accordance with claim 25 characterized in that the polymer film is adapted to be placed in contact with the circuit at the moment of its assembly to allow testing with an optical probe including means for transmitting light toward a predetermined region of the polymer film and for receiving the response of the light from that region such that the electro-optical effects appearing in the film may be detected.

27. An optical probe for a circuit tester, such as printed circuit boards, such a circuit comprising a plurality of elec-tronic components arranged on an isolated support and inter-connected by a network of conductors present on at least one surface of the support, a layer of an electro-optical medium having one face covered with a reflecting layer placed, at least during the test, electrically proximate test points on the conductors on the circuit to be tested, and said circuit being supplied with electrical test signals so as to give in response a voltage signal at at least one of the test points of the circuit, this optical probe being characterized in that it includes:
means to produce and direct a light beam in the direction of a predetermined circuit test point of the circuit mounted in the tester;
means to analyse light obtained in response from the electro-optical medium in proximity of said test point; and means to control the displacement of said beam from one test point to another at the surface of said circuit to be tested.

28. An optical probe in accordance with claim 27 char-acterized in that said means for producing the light beam produce a polarised light beam and said analysis means detect changes in the polarisation of the light obtained in response from the electro-optical medium electrically proximate the test point.

29. An optical probe in accordance with claim 28 character-ized in that the analysis means include means for analysing the polarisation of the light along two axes and apparatus for the transmission of light from the electro-optical layer toward the said analysis means, this apparatus including means to augment the ellipticity of said light by reducing the electric field component in the direction parallel to the initial polarisation of the incident light on the electro-optical layer.

30. An optical probe in accordance with claim 28 char-acterized in that the analysis means further includes, interposed along the path of the light toward the electro-optical layer, means for providing a controllable phase lag as a function of the signal produced by-said analysis means in the absence of a voltage at the test point.

31. An optical probe in accordance with claim 27 character-ized in that said analysis means detects changes in light inten-sity resulting from interferences by reflection from both faces of the layer of electro-optical medium in response to voltages at each test point of the examined circuit.

32. An optical probe in accordance with claim 31 char-acterized in that it comprises moreover means for positioning the analysis means at a point of optical sensitivity in the absence of a voltage at the analysed test point.

33. An optical probe in accordance with claim 27 char-acterized in that the means for controlling the displacement of the incident light beam includes acousto-optical deflection means in two transverse directions with respect to the general direction of the beam.

34. An optical probe in accordance with claim 33, char-acterized in that the means for controlling the displacement of the incident light beam further includes second deflection means for deflecting the light beam issued from the acoustic-optical deflection means.

35. An optical probe in accordance with claim 27 char-acterized in that the means for directing the light beam in the direction of the electro-optical layer includes a convergent field lens the dimensions of which are at least equal to the surface of the electro-optical layer and which has a plano-convex profile the plane face of which is toward the electro-optical layer.

36. An optical probe in accordance with claim 35 char-acterized in that the plane face of said lens is, during the test, in contact with a transducer in which the electro-optical layer is incorporated.

37. An optical probe in accordance with claim 35 char-acterized in that it includes a transducer bonded to the plane face of said lens in which the electro-optical layer is incorpor-ated.

38. A transducer for a testing circuit, such as a circuit on as a printed circuit board, the circuit being constituted by an assembly of electronic components mounted on an isolating sup-port and interconnected by a network of conductors present on the surface of the support, and of which certain regions at least are intended to be subject to a voltage test characterized in that it includes:
at least an electro-optical polymer film element having dimensions and configuration allowing one of its faces to be placed electrically proximate one of a plurality of test points on the network of conductors of said circuit at the support surface;
a reflective covering on said face of the film place-able in electrical proximity to at least one test point and able to deflect incident light which it receives across said film toward the other face; and at least one conductive covering transparent to the incident test light connectable to a reference potential.

39. A transducer in accordance with claim 38 characterized in that said reflective covering is discontinuous and formed of reflective particles mutually electrically isolated.

40. A transducer in accordance with claim 38 characterized in that the same film element is arranged to be associated with several test points and is shaped to provide openings in the interior of which discontinuities at the surface of the support of the circuit carrying the conductors may be accommodated, when said film element is brought into physical proximity with the test points of said conductors.

41. A transducer in accordance with claim 39 characterized in that it is adapted to be integrated with a printed circuit board.

42. A transducer in accordance with claim 39 characterized in that it is adapted to be positioned with the film element in electrical proximity with the test points of said circuit when placed in a printed circuit board tester for testing.

43. A transducer in accordance with claim 42 characterized in that said film element is associated with a transparent support on the opposite face to the reflecting coating face.

44. A printed circuit board comprising an isolating support on a surface of which is present a network of conductors allowing inter-connection of electronic components mounted on the support characterized in that it includes above each of a plural-ity of test points at the surface of said conductors at least an electro-optic polymer film element, said polymer film comprising on its surface proximate the test point a layer reflecting incident light toward the other face after passing through the thickness of the electro-optical polymer film.

45. A printed circuit board in accordance with claim 44 characterized in that the polymer film element comprises on its face opposite to the face having the reflecting layer a conductive layer connectable to a reference potential said conductive layer having notches to make level the output connections and/or solder connections to certain components.

46. A printed circuit board-in accordance with claim 44 characterized in that the same electro-optical film is associated with several test points and presents openings to level certain connections and/or certain components of the circuit.

47. A board in accordance with claim 44 characterized in that the electro-optical molecules of the polymer film are orientated in a preferred direction perpendicular to the plane of said film.

48. A board in accordance with claim 44 characterized in that the electro-optical molecules of the polymer film are orientated in a preferred direction parallel to the plane of said film.