JP2843572B2 - Equipment for testing circuits such as printed circuits - Google Patents

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a circuit test apparatus, and more particularly, to a test apparatus for an integrated circuit such as a printed circuit board.

[Problems to be solved by the prior art and the invention]

2. Description of the Related Art Devices are known which are configured to perform tests automatically, make contact with circuit connection points, input excitation signals to these connection points, and monitor response signals generated by a circuit under test. Such devices are often referred to in the art as "automated test equipment" (abbreviated ATE).

ATE is divided into two types: functional test equipment and device test equipment. In a functional test device, inputs are applied and outputs are received only at the normal input and output connection points of a circuit, such as, for example, a board mounted circuit edge connector.
The input signal is applied and the output signal is monitored to evaluate the functionality of the circuit assembly.

Functional test equipment has the problem that it requires a large number of input / output combinations (referred to in the art as test patterns) to make the circuit work well, especially when it involves complex circuits.
Furthermore, it is not possible to test certain circuit components, in particular digital memories and counters, since certain states of such devices cannot be controlled by externally accessible connection points alone. is there. For example, there is no external connection to the return input of the counter. Similarly, the state of an output of a device cannot be identified externally. For these reasons, some circuit assemblies can only be partially tested with functional test equipment.

On the other hand, the device test apparatus is installed so as to be in contact with an internal circuit connection point, that is, a connection point other than the external input unit and the output unit connection point. There is an advantage that you can. However, to take advantage of this advantage, a special piece of hardware consisting of a plurality of spring-loaded probes individually positioned to contact connection points in the circuit (in the art, a "nail bed" fixture) Cost). According to this means, the function of each component can be verified by driving and measuring the input and output of the component.

Disadvantages of the circuit-embedded (device) test equipment include:
Mainly, hardware is required to provide a circuit to be tested, for example, a fixture suitable for each substrate, and a means for holding the substrate in place and in contact with the fixture. A circuit for driving an output of a device through a nail to a state other than the state theoretically generated from a signal at the input thereof and testing another device having an input connected to the output. There are other less obvious disadvantages that can cause damage to the device. Such backdriving, which is referred to in the art as such, can damage a bag driven device, for example, by causing excessive heating.

Further, when the size of the circuit is reduced, it becomes difficult to access the internal connection of the circuit under test. Finally, built-in test equipment is inherently less well-suited for monitoring the full functionality of the circuit, especially the functionality that interferes with the respective timing of the signals generated by the various components. For these reasons, certain conditions lead to the use of built-in test equipment.

Nevertheless, the ATE provides information indicating the validity of the functionality of the individual components, and the validity of the full functionality of the circuit, not only by the functionality of each component, but also by the interactions between the components. It is desirable that the information shown be generated simultaneously.

The circuit-embedded test equipment gives priority to the former aspect over the latter aspect, and the functional test equipment is essentially the opposite,
The so-called diagnostic stage, in which the functional tester functions to pick up a possible cause of the failure, is a cumbersome operation. In fact, it is generally not possible to locate a fault simply by activating the external input and monitoring the external output. This is because there is no unique test pattern for each possible failure. This problem has been overcome to some extent in some functional test devices by providing a manual probe for use in diagnosing a board that failed the functional test.

During such a stage, the operator installs probes at various connection points of the circuit so that the connection points can be monitored in a manner somewhat similar to that of the built-in test equipment. However, such a procedure is slow and requires a skilled operator.

Accordingly, it is an object of the present invention to provide a functional test of a circuit, e.g., a printed circuit board, comprising a plurality of components mounted on a separate support and interconnected by a conductive network extending over at least one surface of the support. It is an object of the present invention to provide an apparatus that can not only do so, but also diagnose the proven cause of a functional failure without causing certain functional connection problems or damage problems arising in embedded test equipment.

[Means for solving the problem]

According to one aspect of the invention, the device of the invention comprises a layer of an electro-optic medium having dimensions substantially the same as the dimensions of the assembled circuit and being placed in electrical proximity to the conductor support surface. .

It is known from the prior art, for example from U.S. Pat. No. 4,618,819, to use electro-optical materials for remote monitoring of electrical signals without physical contact with the signal-carrying conductor. The device described in that patent uses an electro-optic crystal that is placed in physical proximity to the surface of an unencapsulated integrated circuit. The polarized beam is directed and reflected at a region of the crystal near the medium carrying the electrical signal to be measured. According to well-known rules, the properties of the reflected light are affected by the electric field around the conductor, and this electric field and the resulting signal are detected by suitable detection. Thus, an image of the electrical signal on the conductor can be obtained.

Although the present invention employs certain features of this conventional device,
The invention is, in many respects, in particular, in that known devices are dedicated to measuring signals circulating in an integrated circuit and not to measuring signals flowing between components of an assembled circuit, such as a printed circuit board. Is distinguished by the reason. Thus, this device does not demonstrate the technical feasibility of implementing the electro-optical method on circuits having large dimensions compared to the dimensions of the integrated circuit and significant height variations not present at all in the integrated circuit. Not only that, but above all, known devices that only collect analytical information that does not represent full functionality are not involved in the problems caused by the testing of assembled circuits and meet the requirements of functional and built-in tests Does not teach to apply the electro-optic method.

In one embodiment of the present invention, the electro-optical material may be composed of a polymer film having electro-optical properties. This film can be applied directly to the conductor support surface before mounting and assembling the components. The film may also be used to lie flat against the conductor support surface after component assembly. In this case, the film has openings for receiving these components or solder pads.
Also, the film can be installed in electrical proximity to the surface of the conductor of the substrate under test for testing and incorporated into the optical probe for testing. Such a film can be manufactured, for example, in the form of a transducer, which is dimensioned and configured in relation to the size and shape of the circuit under test, and which, on one side, is electrically connected to a conductor. Elements of such an electro-optic film adapted to be placed in close proximity to each other, at least one element for reflecting light reaching after passing through the layers of the film, and, on the other side, a conductive element serving as a reference electrode Transparent or translucent layer.

The device according to the invention adapted to constitute an ATE, in addition to the electro-optical layer, directs light to any area of the electro-optical medium and receives the light so as to be able to detect the electro-optical effect occurring on the medium. And means for applying a test electrical signal pattern to one or several external connection points so as to produce a response signal at at least one circuit connection point, wherein the light is connected to produce an analog. The analog can be directed to an area of the electro-optic medium that is in electrical proximity to the conductor making up the point, and this analog can be compared to the realized response of a circuit that has been found to be good, and an output signal representing this comparison generated. Means.

According to an embodiment of the device, the means for directing light comprises acousto-optical means for reflecting a light beam impinging on the electro-optical medium layer.

In an optical probe realized according to the above principle, it is preferable to provide an electric field lens having dimensions at least equal to the circuit under test. Advantageously, the lens is a plano-convex lens, the flat surface of which is close to the circuit under test. According to one embodiment, the electro-optic layer is applied to a flat surface of the lens.

For some applications, the electro-optic layer is selected so that the component of the electric field perpendicular to its plane can be detected by polarized light incident perpendicular to the plane of the layer. In the case of electro-optic polymer films, the electro-optic properties of the film can be determined by raising the film to a temperature well above the temperature at which its molecules gain mobility and by exposing the film to an electric field of a predetermined orientation. it can. Upon cooling of the film, if the electric field is maintained, the molecules become immobile and maintain their preferred orientation.
According to the invention, this orientation can be chosen, for example, to be perpendicular to the plane of the film or to lie in the plane of the film. It is also possible to operate in the case of polarized light incident on the layer, for example at 45 °.

The means for analyzing the electro-optical effect caused by the local voltage of the circuit under test in the electro-optical material can be analyzed by the polarization method just described or by interferometric methods as a function of the operating conditions.

If the analyzing means is of the polarization type, it is advantageous to increase the component of the electric field in the polarization direction of the initially incident light in the path of the light reaching the electro-optic medium layer. Furthermore, in the case of a polarization analyzer for analyzing the output light from the electro-optic layer along two axes, means are provided for phase shifting the output light from the electro-optic layer which can be controlled as a function of the output signal of the analysis means. Is advantageous.

The ATE according to the invention is preferably tested in a test phase in which only the external inputs and outputs are excited and monitored, and furthermore, if this test phase fails, further connection points, i.e. specific internal connection points It functions as a functional test device that operates at the diagnostic stage.

Advantageously, the connection points are checked sequentially and during each test many test patterns are entered. Preferably, all possible test patterns acting on the connection point being tested are entered continuously.

Thus, the device according to the invention allows a considerable embodiment and improvement of the diagnostic technique in a way not proposed by the prior art. In fact, with existing manual diagnostic probes, only non-amoigous ra.
It is necessary to use a complex test process at the circuit input to take into account the intervening time and the degree of intervention required to obtain a nit feature and allow the operator to make a diagnosis. Identifying many such vector test steps for each diagnostic location is complex and costly. In contrast, the device according to the invention makes it possible to examine a number of points very quickly during a diagnosis,
As a result, a simple test process can be used at each location, and measurement can be performed at many locations, so that the ambiguity can be easily resolved.

According to a preferred embodiment of the present invention, if the electro-optic medium does not consist of a polymer film adhered directly to the conductor support surface, the medium is placed in electrical proximity via an interface member. The interface member relays the potential appearing on the conductor supporting surface to the electro-optical member while maintaining the relative spatial arrangement of the two. The interface member may include a plurality of substantially parallel conductive posts separated from each other, and the assembly takes the form of a flexible sheet having dimensions substantially equal to the dimensions of the interface. Moreover, the sheet may have a contoured surface to match the shape of the assembly circuit if the assembly circuit is not flat.

According to another aspect of the invention, light may be directed in a general direction toward an electro-optic medium or a portion thereof. The electro-optic medium has a plurality of sites that are individually electrically biased along the path of light. In use, one site may be formed to function with a bias such that light detected at the output represents one region of the medium.

The invention relates equally and individually to printed circuit boards, optical probes and transducers for implementing the principles according to the invention described above.

〔Example〕

Specific embodiments of the present invention will now be described by way of example only with reference to the accompanying drawings.

In accordance with the present invention, an automatic test apparatus (FIG. 1) includes a layer 10 of electro-optic media having dimensions substantially equal to the dimensions of an assembled circuit, such as a printed circuit to be tested, wherein the assembly comprises The surface of the substrate 11 on which the electro-optical medium supports the electric circuit
Combined in electrical proximity to 12, the electrical circuit comprises components such as 15. Surface 12 is 16, 17
These conductors interconnect components and form connection points in the circuit. The medium 10 may be placed in electrical proximity to the surface 12 via an interface element 18, which transfers the potential appearing on its lower surface 19 to its upper surface 100 (in the configuration of FIG. 1). Function.

During use, the lower surface 19 is in contact with the printed circuit board,
The upper surface 100 is in contact with the medium 10.

Light from the source 101 is directed to any area of the substrate surface 12, from which the light is received via the positioning means 102,
The received light is converted into a beam splitter 104 such as a translucent plate.
To the detector 103.

The detector 103 is sensitive to changes in the optical properties of the medium induced by changes in the electric field in the vicinity of the light incident area of the medium 10 and produces an analog of this electric field.
5 takes the form of an electrical signal, which signal thus represents any electrical signal present on the conductor whose potential has been transmitted to the electro-optical medium by the interface member 18 in the area of light incidence.

The board 11 has longitudinal connectors such as 106 and 107, and these connectors are printed on the board to form external connection points of the circuit, and together with these connection points, a conventional functional test apparatus is used. It is possible to set up the physical connection with a suitable standard female connector that is well established for use. A schematically illustrated support or fixture 93 allows manipulation and / or support of the substrate during testing.

Actuate the circuit through these external connections by providing a response at the connection point that is now being electro-optically tested and providing the output 105 with a test pattern sufficient to generate an analog of this response. Can be.

Based on an analysis of the circuit to be tested, we anticipate the expected output for the applied test pattern, as is done with ATE technology, and compare this expectation 108 with the resulting output 105 in comparator 109. Use as a reference for As a result of the comparison, if the generated response is different from the expected response, the comparator 109 displays a test failure on the output unit 110 thereof.

FIG. 1 is a sectional view of a circuit supporting substrate. For clarity, a plan view of the substrate and its conductors (hereinafter referred to as tracks) and electrical components is shown, with FIG. 1 corresponding to a cross-sectional view along line II-II 'in FIG. I do. Typically, the circuit board is 300 m × 200 mm or more, and the electro-optic medium 10 and the relay member 18 have substantially the same dimensions as above. Electronic devices such as 15 on the substrate are the most common components of a dual in-line package, and a portion of the track network interconnecting these components is shown in FIG. Component to board
Components 15 that are not directly visible (in the first configuration) inserted into the underside of 11 are shown directly in FIG. 2 across the substrate to be soldered to the tracks supported by surface 12. Except for the connection pins that are present, they are shown virtually in FIG.

The solder pads between the ends of the component pins and the tracks constitute discontinuities in the surface of the substrate conductor that supports the surface. On the finished substrate, this surface is completely covered by a protective lacquer, which forms a few micrometer thick intermediate layer between the substrate conductor and the external test probe. Moreover, these conductors are often actually covered with oxides that form during manufacture and storage. The oxide itself is a substance that must be considered when realizing an industrial tester.

Finally, often modern printed circuit boards also support components on surfaces that support conductive track networks. Modern substrates support tracks and components on both sides.

In use, the interface member 18 is placed in contact with the conductor supporting surface 12 as shown partially in FIG. The interface member 18 has a plurality of struts, such as struts 111, which provide a potential for each conductor to the electro-optic medium 10 (partially shown) via at least one of the conductor struts. They are arranged in a matrix so as to be relayed. The nature and operation of the interface member 18 will be described later.

Each of the struts corresponds to a separately observable area of the circuit,
Light is directed to each column and received by reflection from each column. This light consists of a polarized light beam from a laser, and analysis of the received reflected light can be performed, for example, by using Wollaston (Wo) to separate the received light into two detectable beams.
(laton) detecting the rotation of the plane of polarization by means of a polarimetry assembly with a prism. In fact, the detector is 2
It consists of two detectors to measure the difference in intensity between the two beams and to compensate for variations in source intensity.

The electro-optic medium is preferably composed of a crystal. In a variant, the electro-optical material may comprise a polymer film provided with electro-optical properties. This film is the AT of Fig. 1.
E may be used instead of the electro-optic crystal layer.
Also, the film may be preformed to mate in direct contact with the conductors of the substrate between the components of the substrate to be tested without relying only on interface members or such thin members. . As a modification, before mounting the component, the surface in contact with the conductor was originally
Alternatively, it may be installed directly on a conductor supporting surface having light reflectivity by attaching silk screen printing of metal fine particles,
For example, the film may be integrated with the printed circuit board during its manufacture. In such a configuration, no interface members are required, and the electronic boards can be subjected to tests that, by their manufacture, would benefit from all the advantages of the present invention.

The mechanism for detecting signals by electro-optical actuation will be described in more detail (FIG. 3), in particular in connection with an embodiment of the invention using an interface member. However, the above essential characteristics are the same when a polymer film is used.

Output light from a laser (eg, a 633 mm wavelength HeNe laser) is linearly deflected by a deflector 31 and
Is focused on the acousto-optic deflector 33 and directed via the lens 34 to the inspection location on the reflective surface 36 of the electro-optic crystal 37. Generally, intermediate strength, typically 5
A 100 mw continuous laser is used.

The known acousto-optical deflector 33 is controlled by a voltage signal continuously output by a control unit (not shown) of the ATE so as to deflect incident light to an arbitrary portion of the electro-optical medium desired to be inspected.

In a preferred embodiment, two acousto-optic deflectors may be installed directly. Then, the first control signal voltage controls the beam deflection along a first direction, e.g., corresponding to one of the dimensions of the substrate to be tested, and the other voltage signals are preferably perpendicular to the first direction. Yes, for example, controlling deflection along a second dimension corresponding to another substrate dimension.

When two deflectors are used in this way, the time required to move from one monitoring location to another monitoring location can be reduced to at most a few microseconds.

Opposite surface 36 is separated, at least in the sense that a potential applied to one point on this surface does not propagate to another. Thus, if the reflectivity of this surface is originally obtained by metal deposition, by means of a conductor, this surface is not composed of a uniform layer, but is separately deposited and not interconnected, but the interface member 300 Via (or directly in the case of a polymer film) the test location of the substrate 39
It consists of a matrix of reflective particles connected to the area under test, such as 38. According to one possible embodiment, the reflective surface 36 can be formed of an essentially separating material, for example a thin layer of a dielectric material such as an alternating structure of layers of titanium oxide TiO ₂ and silicon oxide SiO ₂ .

A transparent electrode, for example made of gold or aluminum, is deposited as a layer on the first surface of the electro-optic crystal 37 where the incident light strikes, this electrode being kept at ground potential as a reference and thus being inspected. If the electric potential at the test point is different from this reference, the polarization of the reflected light is different from the polarization of the incident light because the electric field is applied over the thickness of the electro-optic crystal 37, and this difference is detected.

Bismuth germanium oxide Bi ₄ Ge ₃ O _{12 with} crystal axis 100 optically oriented perpendicular to quarter-wave plate 302
In the case of such a three-dimensional crystal structure, the phase shift of light due to the continuous electro-optic action is proportional to the potential difference between the surfaces of the crystal 37 at the monitoring location,
It is independent of the distribution of the electric field in the crystal.

The reflected light is passed through a splitter 303 to a quarter-wave plate 302.
And downstream of the plate 302, the light is directed by a lens 304 to a Wollaston analyzer 305,
Two beams from the stem of the analyzer reaches the optoelectronic detector 308, 309 each, these detectors generate electrical signals in respective I _1, I ₂ at their output portion 306 and 307.

 According to this type of polarization detection, the following relationship is obtained.

I ₁ = I. (1 + m) and I ₂ = I. (1−m) (where I is a magnitude proportional to the luminous intensity of the laser 30, m is expressed in radians, and absolutely Do detection is lagging.) differential amplifier 310 for inputting a signal output section 306 and 307 signals I ₁ -I ₂ at its output 311 (and thus, outputs equal) to 2I _m. If, during the test, the total intensity change of the laser 30 corresponds to a frequency much lower than the frequency of the change of m, this value m can be obtained directly by appropriately filtering the available electrical signal at the output 311 of the amplifier 310. Can be. As a change example,
The signal I ₁ + I ₂ may be used to adjust the light intensity of the source.

In the case of non-stereostructured crystals, such as lithium nobelium oxide LiN ₆ O ₃ , the electro-optic effect is also proportional to the potential difference between the planes of the electro-optic crystal. However, the use of such crystals requires some care when the proportionality factor between the potential difference and the effect obtained is determined by the orientation of the crystal cut.

Desirable properties of electro-optic crystal are low absorption, low diffusion,
Low circular birefringence and high linearity. When a highly birefringent material such as lithium nobelium oxide is used, a slight change in the incident angle of light causes a change in the sensitivity of the electrostatic phase. Thermal changes that cause a change in crystal thickness have a similar effect, and both must be avoided.

In order to improve the performance of such a birefringent material, according to the present invention, the polarization of incident light may be controlled according to, for example, the incident angle or the temperature. Additionally or alternatively, the device of the present invention may exhibit a crystal structure that exhibits a perpendicular orientation, but substantially the same thickness, as compared to the electro-optic medium, so as to reduce or eliminate birefringence. Good to have.

Electro-optic media exhibit a resistivity of at least about 10 ohm.cm, especially for testing low frequency circuits, and a dielectric constant of at most about 100 to introduce only low capacitance (1 pf or less). .

Inside the interface member 18 (FIG. 4), conductive posts 111 are embedded in a flexible separation substrate 112. Prop
Each strut, such as 111, is for example cylindrical. Substrate 112 includes a collection of other mutually parallel struts that are longitudinally spaced and extend widthwise along interface member 18.

In use, these members are placed in contact with the conductor support surface 12 of the printed circuit board and the
, And the surface of the other strut is also close to the other conductor. The electric potential of the conductor of the support 111 is its upper surface 115
Can be monitored.

A conductive film 116 is attached to the upper surface 117, except for the region of the upper surface of the conductive column, which is grounded to give a sharp potential step to the upper surface of the column for the inspection of the electro-optical action as described above. .

Since the current does not need to flow along the conductive pillar 111, a high resistance can be allowed in a region close to the conductor 114. Thus, the device can be used to inspect circuits provided on a substrate, such as the substrate 11, on which the separating protective lacquer film 118 has been applied. This is a particularly important advantage of the ATE according to the invention. Thereby, the substrate can be tested after all the manufacturing steps including the lacquer application are completed. In prior art device testers, the nail must be in direct contact with the conductor, so that the lacquer must not be applied, and in functional tests, the lacquer must be penetrated if a manual probe is to be used. In the present invention, the finished substrate may be tested.

If the conductor support surface of the assembled circuit to be tested is not uniform, as is the case with the substrate 5 shown in FIG. 5, the corresponding surface of the interface member may be contoured, shaped or scraped appropriately. I do. The board 50 has, for example, circuitry using surface mounted components (eg, 52), and the connecting pins 50 of these components are mounted directly to the conductive tracks 53 without penetrating the board 50. ing. In such an arrangement, a flat conductive surface may not be present, and thus the interface member 51 is adapted to maintain the flat surface in close proximity to the electro-optic medium 54 to receive a surface mounted component. .

In the embodiment of the test apparatus according to the invention shown in FIG. 9, physical scanning as well as acousto-optical deflection are used to extend the surface of the scanning circuit. Laser 43 as in the example of FIG.
0 emits the test beam to the acousto-optic deflector 433 via the polarizer and lens 432. This device 433 comprises, for example, two acousto-optic deflectors of the type manufactured by Automate ET Automatismmus (Paris 78460, Chebreus, France).

A first deflector is provided in the housing 433 to deflect the laser beam in a first direction a, and a second deflector deflects the beam so shifted in a right-angle direction b to combine two deflectors. Thereby, it is possible to sweep a 50 × 50 mm square surface at a distance of 1000 mm (corresponding to the focal length) from the output section of the lens 432.

The beam exiting device 433 falls on a turning mirror 441 of a first physical deflector 443, which deflects the beam to a turning mirror 442 of a second physical deflector 444. Deflector 44
3, 444 is displaced in two directions A, B parallel to the directions a, b, so that the emergent beam 445 is 500 at the focal plane of the lens.
Sweeps a rectangular surface of × 500mm.

The beam 445 is, for example, 500 x 500 mm made of BK7 glass
It falls into the convex 449 of the plano-convex lens of the size.

A rectangular layer of mosaic of BGO (bismuth germanium oxide, Bi ₄ G ₃₃ O ₁₂ ) crystals is attached to the flat surface of the lens. The dimensions of the lens 450 and the electro-optic crystal layer 451 are deflectors
This roughly corresponds to the total area to be swept by the beam 445 exiting the system 443.

BGO layer 451 has a thickness in the range of 1 mm. In this example, by aligning the BGO layer 451 with the flat surface of the lens 450, the mechanical vibration naturally occurring in the crystal as a result of the piezoelectric resonance phenomenon can be made annular. Such oscillations are evident by the parasitic optical signals that cause these oscillations by photoelastic effects on the crystal.

The assembly of laser 430, polarizer 431, acousto-optic and physical polarizer 433, 443, 444, and lens 450 with BGO layer 451 are integrated by layer 451 with the polarized light collection / detection device. This assembly (not shown in FIG. 9)
Shown in the figure as 319, a splitter 303, a quarter-wave plate 302 of lens 304, and a Wollaston prism 305
And an optoelectronic detector 308, 309, and an operational amplifier 310 constituting an optical probe incorporated in the ATE head.

Interface with anisotropic conductivity when used
An interface member 462 similar to 18 (FIG. 1) is first placed in contact with the surface under test of the printed circuit board 460 (FIG. 9) and the card 460 in contact with one surface thereof.
An analog of the conductor voltage is formed on the other side 463. This plane 463 and the electro-optic crystal 4 in the center of the optical probe
The free surfaces of 51 are in testable contact and are pressed firmly together by pressing means (not shown) to eliminate any parasitic spacing between BGO layer 451, interface member 462 and the contact surface of the substrate to be tested. Or minimize it. In this example, the JSC at Federal Republic (West Germany)
An elastomer sheet sold by Technics under the name "Zebra Vidimentionel" was used for the interface members. Its thickness is, for example, 0.1 to 5 mm depending on the type of substrate and its surface discontinuity.

In FIG. 9, the inspection area 470 corresponds to the surface of an element to be subjected to acousto-optic scanning of a 50 × 50 mm crystal. The displacement by the physical deflection device allows for acousto-optic scanning of 100 adjacent element surfaces on the surface of the electro-optic crystal 451.

Typically, an acousto-optic scanning speed that can be obtained is 10
0 KHz (frequency from one test point to another test point). This physical scan looks at a change of about 50 milliseconds from one element area 470 to another element area.

Referring to FIG. 9, the polarization of the linearly polarized light beam 445 at the input of the crystal layer 451 will be measured when the conductor of the substrate under test has no voltage load in the area where the beam is incident. It remains unchanged while crossing. By applying a voltage, the phase shift P between the two components of the electric field of light
(U). Then, the state of the beam polarization returning through the lens 450 at the output of the BGO layer 451 is elliptical.

The emerging beam is redirected by the splitter 303 (FIG. 3) to the detection lens 304 and then split by the Wollaston prism 305 into two components, the intensity of which is represented by the following equation:

I ₁ = 1 / 2I ₀ (1 + cos (P + P ₀ )) I ₂ = 1 / 2I ₀ (1-cos (P + P ₀ )) (In the above equation, I ₀ is the intensity of incident light, and P ₀ is four. This is the stationary phase lag introduced by the one-wavelength plate 302.) The differential amplifier 310 outputs the signal S.

S = I ₁ −I ₂ = I ₀ (p + P ₀ ) If the phase shift P is zero, the signal S is often the layer 451
Is very small because the amount of polarization change that occurs in

As an example, for a laser wavelength of 647 mm (krypton laser), BGO with an index of 2 and an electro-optic coefficient equal to 1 pm / V
For crystals, a value of P = 8.10 ^-4 radial / V. Is obtained.

Quarter wave plate, P ₀ = Pi / 2, and S = I ₀ × P
In the response in the case of (2), the signal S changes linearly with V, and a value of P of about 10 ^-4 is easily detected. This is so even though the intensity of the laser source is subject to relatively slow fluctuations in time (up to about 10 percent) according to the differential method used.

By fixing the BGO crystal, the induced deformation results in different parasitic birefringence between one location and another. As a result, P ₀ varies between one location and another location. After each positioning of the beam and before electrical measurements by inserting a Kerr cell 480 between the quarter-wave plate 302 and the Wollaston prism, P ₁ is biased.
Return to P _1/2 . The device is constructed by placing a plate of electro-optic material between two transparent electrodes in a variable electric field supplied from the output 311 of a differential amplifier 310 by a feedback loop 482 consisting of a switch 484 and a variable amplifier 486. ing. Then the phase shift can be performed on plate 31
By introducing between 1 and the electric field component of the beam passing therethrough, it is possible to adjust to the level required to eliminate the background component of the signal S corresponding to the test location. For a test signal of a moderately high frequency, the high frequency component of the signal s contains the desired test information.

The modulation depth of this configuration can be further improved by introducing an imperfect polarizer 340 between the crystal layer output and the quarter wave plate 302.

In FIG. 10, the polarizer is a cube 340 that blocks the beam 335 emanating from the lens 34 of FIG.
Is used in place of the divider 303.

Axis system 350 represents the polarization of light exiting the lens, where P is the angle of polarization of the linearly polarized light along the Y axis relative to the initial direction. X of sector 352 representing the polarization of beam 335
The component and the Y component represent the principal axes of the polarization ellipse, and the Wollaston prism 305 (FIG. 3) is oriented to split the light component along both of these axes.

Divider cube 340 separates beam 335 into two beams, one beam 334 is transmitted and the other beam 336 is reflected. This cubic interface 355 serves as a defective polarizer for the reflected light and is configured to increase its polarization ellipticity. The component along the Y axis is based on the principle described in French Patent Application No. 88/04177 (filed on Mar. 30, 1988 in the name of the present applicant) with respect to the component X as shown in diagram 360 in FIG. Greatly reduced.

In contrast, for the transmitted portion of beam 334, it is linearly polarized along Y (line 370).

The components I ₁ and I ₂ received after separation of the beam 336 by the Wollaston prism are expressed as follows:

I ₁ = (I + A _p ) I ₀ / 2A ² I ₂ = (I + A _p ) I ₀ / 2A ² A is a coefficient larger than 1 indicated by the characteristics of the polarizer cube 340. These relationships indicate that more powerful lasers can be used to increase the modulation depth without saturating the detector at the output of the Wollaston prism.

The interface between the printed circuit board 460, the interface member 462, and the crystal layer 451 is schematically illustrated in the eleventh. Two conductors 467 and 468 constituting the 2 test points, one is the lower voltage V _1, the other is shown in the ground potential. These conductors are often covered with oxide. When the substrate is completed, these conductors are covered with lacquer. The oxide or lacquer layer is between the top surface of the substrate and the interface
An interval 469 is formed between the 462 and the opposing surface.

An electrical equivalent circuit diagram of this configuration is schematically shown in FIG. Crystals, by the capacitance of the conductors and the spacing 469, the voltage V ₂ to reach the crystal in response to V ₁ was when greater is the capacitance of the lower and crystals capacitance interval 469 is weaker than all. Moreover, the predetermined interface member 462 having a thickness defined by the components provided on the surface of the card 460, the greater the spacing 469, crosstalk represented by the ratio V ₃ / V ₂ increases.

 In one example, these parameters were as follows:

Interface member 462 thickness: e = 2 mm BGO crystal 451 thickness: = 1 mm Crystal dielectric constant: e '= 16 Distance between two test points: d = 400 microns Maximum allowable crosstalk is about 10% And the height h of the gap 469 should not exceed 1.5 microns.

It is therefore desirable to deposit an electro-optic material that has as low a dielectric constant as possible and at the same time exhibits as high an electro-optic coefficient as possible.

The relationship between the crystal's electro-optic sensitivity index n ³ r (n is the refractive index of the medium and r is the electro-optic coefficient) and the dielectric constant e ′ is a standard for the selection of materials for use as the electro-optic layer 451. (Demonstration number).

The table below illustrates this relationship for different crystals.

In devices with BGO layers as intended, reliable detection of 1 volt amplitude test bits at a frequency of 5 MHz, output at test points 2 mm apart and achieved by a ground conductor, was achieved. The thickness of the elastomer interface member is
2 mm and the contact surface was oxide-free and covered with a 5 micron thick lacquer layer.

Given the above table, it is shown the number of organic compounds such as MNA that form such substances that are of most interest for the intended use.

Such materials can be used not only in crystalline form, but also in combination with a support incorporating molecules of such active electro-optic material. Thus, Perspex (PMMA: polymethyl methacrylate) can be used as a support at an MNA density of about 15%. Thus, MNA is used as a dopant with the molecules carried in the PMMA matrix to make the composition electro-optical. Other possible dopants with electro-optical properties include, for example:

DNA [4- (N, N-dimethylamino) -3-acetamidonitrobenzene] COANP [2-cyclo-octylamino-5-nitropyridine] PAN [4-N-pyrrolidino-3-acetaminonitrobenzene] MBANP [2- (Α-Methylbenzylamino) -5-nitropyridine] This type of polymeric material has been used in some fields such as Rockhead Missiles and Space, Hoechst Theraneses and other large companies in the chemicals field. Companies, universities and research institutes are the topics of many development plans currently underway. [For example, "Polymers for Nonlinear Optical Devices" (Sophia Antipolis, by the conference report of the symposium called "Nato Advanced Workshop",
(June 19-24, 1985).

Film on a known substrate using the obtained polymer,
Fibers or thin layers can be formed.

A particularly interesting property of these polymers is that they can be used in the form of films that can be produced in large quantities and at a reasonable cost. These films can be used on transparent supports, such as glass, with a thickness of several microns (eg, 10 microns). Also, these films can be supplied directly in the form of an elastic film made of several adjacent layers of the initial film (eg, 500 microns thick). After drying, the molecules of the electro-optically active component are fixed to an amorphous support material without a specific orientation. Making the support material electro-optical requires heating the active molecule to a temperature sufficient to restore some mobility within the matrix. The value of the transition temperature varies depending on the polymer, but is typically around 100 to 200 ° C. In this state, these active molecules are oriented with respect to the support under the action of an electric field. These molecules tend to align themselves in the direction of the excitation field. The higher the excitation electric field, the higher the proportion of active molecules oriented in the direction of the electric field.

When the temperature is reduced again while the molecules are oriented under the action of the electric field, they retain this orientation. Thus,
This material retains the oriented structure exhibited by anisotropic optical behavior (Pockels action) in the presence of an electric field.
Thus, when the material is exposed to linearly polarized incident light in the absence of an electric field, the material does not cause any change in polarization.

In contrast, in the presence of an electric field, the transmitted light maintains elliptically polarized light that is linearly related to the strength of the applied electric field.

Figure 12 shows that polymer molecules are oriented by the application of an electric field, called the matching electric field, while the temperature is reduced below the transition region (beyond which the electro-optical dopant molecules lose their mobility). The exponential distribution of the electro-optic film 500 realized on the polymer of FIG. The elliptical surface represented as 502 depicts the refractive index distribution of the material while applying an electric field perpendicular to the plane (called the detection electric field). This ellipsoid shows the spatial refractive index change of the material. This ellipsoid has a rotational symmetry about the normal 504 in the plane of the film 500, indicating that the material is optically isotropic and parallel to the plane of the film.

When the film 500 is illuminated with a polarized incident beam 506 perpendicular thereto, light collected from the film (by transmission or subsequent reflection from the opposite side 508 of the film) does not show a change in polarization. This explains why if such a film is used instead of the BGO layer 451 of FIG. 9, for example, the voltage at the connection of the scanning circuit cannot be detected. On the other hand, the incident beam 510 inclined with respect to the plane of the film changes its polarization.

Therefore, in this case, it is necessary to use a polarized incident light beam inclined with respect to the electro-optic layer in order to exhibit the Pockels effect caused by the voltage in the test circuit.

FIG. 12b shows the ellipsoid of the index obtained with the same material under the action of a detection electric field perpendicular to the plane of the film, when the molecules of the film 500 are pre-oriented in the plane of the film by the matching electric field. In this case, the ellipsoid 51 of these indices
2 has a rotational symmetry with respect to an axis 513 in the plane of the film 500. Moreover, the normal incident beam 506 changes its polarization along the principal axis in the plane of the film as a function of the difference in indices γ ₂ , γ ₃ . Under these conditions, the assembly of FIG. 9, which has a polarimetric analyzer as shown in FIG. 3, which involves normal incidence, incorporates the Pockels effect created by the circuit to be tested into the optical probe or substrate. It can be used for polymer films to be produced.

A device for orienting the molecules of the active composition in the plane of the polymer film is shown schematically in FIGS. 17 and 18. The polymer stop 800 is charged to a tunnel oven 801 which has been heated to a temperature T of about 100 ° C. (slightly above the transition temperature).

This strip was kept at the same potential + V continuously.
To the interface between the two opposing metal plates 802, 804. At the exit 805 of the tunnel oven, the strip enters a second tunnel 810 formed, for example, by two opposed metal plates 812, 814 kept at zero potential. As a result, the spacing 815 between the heating tunnel exit 801 and the second tunnel 810 is susceptible to a substantially uniform and parallel electric field in the plane of the strip 800 located at this spacing. The electro-optically active molecules of the film tend to orient themselves parallel to this direction due to the local action of the electric field at the exit 805 of the tunnel 801. The two nozzles 818, 819 on either side of the film 800 open at a spacing 815, at which interval an inert gas (eg, argon or hexafluoride ion SF ₆ ) cooled to −40 ° C.
Is sent out to both sides of the film 800. This gas is discharged from between the plates of the second tunnel by means not shown. The temperature of the film 800 passes through a transition region in the space 815. The active molecules maintain the preferred orientation in the plane of the film that cools in the second tunnel.

Prior data on the orientation of the film structure and the incidence of the test light allow this electro-optic polymer film 500 to be used to test signals in circuits by means of a polarimetry.

As a variant, according to another aspect of the invention, Fabry-Perot interferometry can be used instead of polarimetry to detect the Pockels effect due to the voltage to be tested.

FIG. 19 shows a circuit board 840 which is a typical example of an assembled circuit.
The substrate is provided with a support 841 of separating material, on the surface of which the electronic components are provided, which are electrically conductive links such as 848. Interconnected by

A layer or plate of electro-optic material 842 is placed in contact with the top surface of conductor 848 and in proximity to circuit support substrate 841. This layer receives a predetermined normal incidence of the laser beam emitted by the laser source. The surface of the film 842 facing the laser source is a layer 84 of conductive translucent material forming a translucent mirror of average reflection coefficient value.
Coated with 5. The other surface of the film 842 (opposite the circuit board under test) is coated with a conductive and reflective layer of reflective material 846, such as an aluminum layer that is thicker than the layer 845 to allow a substantial amount of light to reach this layer. The part is to be reflected. A layer 847 is provided to absorb that portion of the light transmitted by the reflective surface 846.
Layer 846 may be formed of particles or fragments that are separated from each other to prevent short circuits between adjacent conductive tracks on substrate under test 840.

Interference occurs between the light reflected by the parallel mirrors, and the light beam diverging from the film (indicated by a double arrow in the drawing) is reflected by the translucent mirror 851 and travels in a common path of the emission beam 843 and the output beam 849. The superimposed signal is directed to analysis means for measuring the intensity of the optical signal 849. As is well known, the interference phenomenon is the thickness (e) of the film, the distance separating the mirror, and the wavelength [lander (lander = 1/1000 cm ³ )].
And the refractive index (a) of the material of the film. The index of refraction "n" varies with the electric field experienced by the film, ie, as a function of the potential V to be measured.

Thus, by measuring the luminosity of the signal at the receiver 850, a measurement of the potential V at a predetermined location of the circuit 800 is achieved. If the layer of electro-optic material 842 is a polymer film of the type described above (and can be integrated with the substrate 840), even if the electro-optic molecules of the film were originally oriented perpendicular to the plane of the film However, it can be seen that the Pockels effect can be measured by this interferometry at normal incidence.

To mitigate the consequence of thickness differences in the electro-optic film or layer, the light source emits a light beam of variable wavelength. This means that: That is, the operating point of the device is determined by the characteristic curve of the strength as a function of PHI = [2 pi / lumbar (1/1000 cm ³ )] (20th.
From above, from point A, located on the first part of the curve where I = maximum for any PHI (between the maximum and zero)
I can be shifted to a point B located in a second portion where I varies significantly as a function of PHI, and the sensitivity is a function of the change in I in the second portion of the curve near point B, preferably about It corresponds to the maximum value / 2. With point B fixed, the measurement is affected by the wavelength corresponding to point B.

In FIG. 13, the printed circuit board 600 has a substrate 602 of a material conventionally comprised of epoxy and fiberglass, which substrate has a metal 605, 606 on each of its faces 603, 604. A covering or conductive track is provided to form a network connecting the components provided on the substrate.

The surface 604 of the substrate is coated with a layer of an electro-optic polymer film 608, which extends over substantially all of the surface above the metal tracks 606 to be formed. The film 608 is assembled to the card after formation of the metal coating, but prior to installation of the components, for example, by bonding. Face 604
Is covered with a reflective aluminum layer 609 with the inspection substrate pattern of the elementary mirror. The size of the particles or pieces of this pattern is such that each piece does not cause a short circuit between two adjacent metal tracks. On the opposite side, the film 608 is coated with another aluminum layer 610 that is thin enough to be transparent to test light projected from the direction 612 onto the substrate. This layer 610
Constitutes a reference electrode for the electric field generated in the layer of the film 608 by the voltage applied to the conductor 606.

Substrate 600 supports discrete components or components, such as integrated circuit package 616, on surface 603 thereof. The component has connection pins 618 which penetrate the substrate substrate in holes 620 provided at right angles to the metal tracks 606 on the other side of the substrate, these pins being connected via the solder pads 622 to the metal tracks. Connected to 606. Also shown is a component 624, which is provided on the surface 604 of the substrate and whose two outer surfaces 625, 626 are connected to two metal tracks 606 by two beads of solder 628, respectively. . To avoid any electrical contact between the metal solder and the reference electrode 610, anticipate openings such as 630 before assembly of the components 614, 616, 624 solder beads 622, 62
Around 8 formed a polymer film 608. These openings may be formed before the components are placed on the guard, or may be formed in the film 608 prior to bonding the film 608 to the card surface 604. Alternatively, a pre-formed film or film element as a function of the substrate area supporting the conductor to be tested may be placed on the substrate before mounting. Top surface of metal cover 606 and film 6
The contact area, such as 632 between 08 and 08, forms the test site.

These areas can be monitored by projecting the incident laser beam and analyzing the beam reflected by the corresponding metal mirror 609. This analysis is performed with a test probe similar to the device of FIG. 9 with the BGO layer 451 and interface layer 462 removed. In fact, a plano-convex lens 45
Light leaving the zero impinges on a card provided immediately adjacent to the flat surface of the lens. Depending on each part of the test board 600,
The beam changed back by the voltage action driving this point is detected and analyzed by a configuration such as 319 in FIG. Due to the close contact between the film 608 and the test conductor 606,
Excellent optical storage of the signal to be monitored and good spatial resolution are ensured. In a certain substrate structure (13b) in which the conductive region 606A to be inspected is provided close to the ground connection region 606B, an electric field is formed parallel to the plane of the electro-optic film (see FIG. 13b). Lens 640, 64
1). The presence of this electric field can be tested directly without using any electrodes of the opposite polarity that might otherwise be necessary. In this case, the Pockels effect is manifested by a substantially parallel, no longer perpendicular electric field in the plane of the film.

According to another embodiment of the present invention, an electro-optic polymer film can be directly used for an optical probe of a test apparatus. In fact, the good properties of these materials are very suitable for this application because of their high level of electro-optic coefficient and their low dielectric constant. These materials are BG
Instead of the O crystal layer 451, for example, a plano-convex lens 450 (ninth
2) can be formed in the form of a layer which is bonded to the flat surface.

According to another advantageous technique, for each type of substrate, an electro-optical transducer specific to this type of substrate can be used. In this connection, there is the good result that the costs for the above-mentioned types of polymers are not very high. Thus, for each new type of substrate under test, it is possible to produce a transducer that is designed to be associated with an optical probe during testing, but is suitable for that form of substrate.

Transducer 700 (FIGS. 14 and 15), when connected for testing, provides an internal opening or cavity in which components, solder pads and other discontinuities on the top surface of printed circuit card 705 are located. In the form of a plate 702 made of glass or transparent plastic material with low photoelasticity and easy to process. Converter 700
An electro-optical polymer film of the type already described with reference to FIG. The film 712 is coated with a transparent aluminum electrode where it meets the support plate 702. Film 712
Is also coated with a reflective layer of aluminum. As mentioned earlier, this layer is not continuous,
It is formed of particles that are spaced apart from each other, so that particles in contact with a conductor on the surface of the substrate do not come into contact with an adjacent conductor or their potential is not affected by this adjacent conductor. It has become.

Apertures 720, 721 provide components, such as 724, for example.
Into the solder beads such as
Transducer 700 is arranged such that pins 726 of component 727 through 5 are fixedly received on conductors on surface 730 of the contacting substrate of transducer 700.

When the substrate 705 and the transducer 700 are mated by an operating device (not shown), the transducer is a plate sufficient to bring the electro-optic polymer film 712 into contact with a conductor such as 728, 732 on the surface 730 of the substrate 705. 702 makes it more rigid. The parasitic capacitance between the sensitive film 712 and the contact is the minimum level that can achieve good spatial resolution, for example, 0.1 mm for a 10 micron thick lacquer. With relatively thick interface elastomers such as those described with reference to FIGS. 1 to 3, the result that disassembly is possible is substantially better. The low thickness of this lacquer keeps the crosstalk between test points very close together at an acceptable value.

The support plate can be manufactured by routine mechanical means.
The sensitive film (FIG. 15) can itself be formed by laser. The transducer assembly 700 is in direct contact with (or on) the flat surface of the optical probe's electric lens (see 45 in FIG. 9) instead of the structure constituted by the BGO lacquer 451 and the elastomer interface 462. (Immediately adjacent). Alternatively, the transducer assembly 700 may be operated separately from the lens while the substrate under test is inserted. As in the case of the electro-optical substrate of FIG. 13, the film portion or electro-optically sensitive film may be separate and may be deformed into a predetermined number of test points or areas distributed over the substrate surface. Good. The transducer 700 has an area without mirrors or apertures 728 so that the reference symbol 728 that calibrate the beam reflection system to accurately direct the incident beam to the conductor for the connection under test of the circuit can be positioned. Is provided. The controller of the reflective assemblies 433, 434, 444 (FIG. 9) is then programmed to selectively send signals to the connections of the substrate corresponding to the area of the transducer comprising the electro-optic film or material. Can be

FIG. 16 is a schematic sectional view of the electro-optical film 740,
This film 740 has a transparent reference electrode 742 provided on one surface thereof. The other side of the film is covered with a test pattern of small mirrors 744 formed by depositing a thin layer of aluminum. To further increase the contrast of the measurement, the polymer is etched (eg, chemically) between the mirrors to form a dimple 706 of the required resolution (eg, 0.1 mm). Thus, 2
The capacitance between two adjacent mirrors 744 is reduced. This film can be used in the transducer 700 of FIGS. 14 and 15.

In addition to the above-mentioned point that the present invention makes it possible to test a finished substrate, that is, a substrate coated with lacquer, another advantage of the present invention is that the activity of internal connection points can be observed, Conventional test equipment is almost always interrupted. In fact, any contact with the probe, the current carrying element, interrupts the normal operation of the monitoring circuit, and with conventional test equipment, the monitoring circuit, whether or not it is testing, It should not work.

In conventional test equipment, the maximum speed at which the monitoring circuit can function may in some cases be limited by test requirements. Thus, in the same operating situation in which the circuit should function properly, no test is performed. As a result, the conventional test apparatus has a problem that the manual diagnostic probe has a high capacitance (about 100 pf). On the other hand, the interface member 18 only has a capacitance of 1 pf, so that the circuit can operate at full speed during the test.

The electro-optic materials mentioned above are the angles of polarization of the beams passing through these materials in the region traversed in response to this electric field under the action of the electric field. A substantially linear electro-optic effect (Pockels effect) occurs at the intended wavelength. Other materials exhibit a secondary action (Kell action), and according to this secondary action,
The change in the polarization angle is the area unit of the electric field [100 square feet (=
9.29cm ² )]. This property is used in another embodiment of the invention (FIG. 6), in which light is brought into close electrical contact with a conductor, such as conductor 61 printed on a substrate 62 of the electrical circuit to be tested. It is directed laterally as a whole to the provided secondary electro-optical medium 60.

Light is received by the detector 63 at the opposite edge of the medium 50, which outputs according to the principles outlined above.
Yields 64. This output 64 can be used in ATE as described above.

On the upper surface of the electro-optical medium 60, an electrode such as the electrode 65 is attached along the light path. Electrical connection (not shown)
Can bias each electrode to a predetermined potential, eg, zero or 20 volts relative to a reference potential.

Under these conditions, the nature of the transmitted light received by detector 63 is determined by the potential appearing before and after the electro-optic medium. When the circuit below the medium is excited, these potentials are relayed to the lower surface of the medium 60 via the relay member 66. Taking the example of a digital circuit having a logic 0 at 0 volts and a logic 1 at 5 volts, assuming that all bias electrodes 65 are biased to 0 volts, then either 0 volts or 5 volts of electro-optic Applied to the medium 60.

The characteristics (FIG. 7) of the electro-optic medium selected for this embodiment of the invention, for example, a PLZT ceramic composition of the type used for controlling the optical gate of a laser beam, are shown in FIG. It is.

Thus, at an applied potential of 0-5 volts, the electro-optic effect is detected between a and b along the ordinate. Now one
Assume that one electrode, for example, electrode 65, is biased to a potential of 20V. Then, the effective potential appearing before and after the medium 60 in the area of the conductor 61 is 20 volts (logic 0) or
15 volts (logic 6). These potentials act between c and d.

Therefore, the dynamic signal appearing in the region of the biased electrode is distinguished from the cumulative effect occurring in the unbiased region,
-D is much larger than swing ab. Therefore, by selectively biasing the electrodes, the activity of an arbitrary region along the light path can be inspected.

In order to check the activity of an arbitrary area, a detector row 80 (FIG. 8) extending in the edge direction is required. Where parts of the embodiment of FIG. 6 correspond to those of FIG. 8, joint reference numerals have been used.

A bias electrode, such as electrode 65, extends laterally across the medium 60 in a direction substantially perpendicular to the direction of light propagation in the medium in a strip-like manner to reduce the electrical activity of any region.
The electrode that biases the region is energized and can be examined by selecting a detector output signal, such as 81, that receives light through the region.

A method for configuring an ATE according to the present invention to operate is described below.

Obviously, in an assembled circuit, such as a printed circuit board, the state of each output is determined by the prior state of the input. For example, if a good circuit is known, K
The state SK of the third output is a function FK of the sequence [E] of input states, which can be expressed as SK = FK ([E] 0).

The purpose of the prior art functional test apparatus is that for a sequence of input states [E] assigned to the circuit, the state of each output, such as the state SK of the Kth output, characterizes the correct operation of the circuit.
The goal is to prove that a function like FK can be successfully found on the sequence [E] of input states.

Otherwise, ie, if the state of at least one output, such as the Kth output, is incorrect, for example,
If S'K takes the place of SK, it is clear that the monitoring circuit has failed.

However, it will be appreciated that this information does not aid in circuit repair and that the SK and [E] requirements are very complex.

Further examination shows that the state of each output is determined by the state of the input through the state taken by the internal circuit connection point. In the case of a good circuit, the state SK of the K-th output is not only directly, but also the state I ₁ of the first internal connection point, the state I ₂ of the second internal connection point, etc. Also depends on the sequence of input states [E], which is n
In the case of a circuit having the number of connection points, it is expressed by the following equation.

_{SK = GK (I 1, I} 2 ···, I n) state on behalf of the K-th output takes a state SK S'k
, The first connection point takes the state I ′ ₁ instead of taking the state I ₁ , and / or the third connection point takes the state I ′ ₃ instead of taking the state I _3. etc··
・ There is a series of possibilities as shown in the following equation.

_{S'K = GK (I '1,} I' 2 ··· I n) or _{S'K = GK (I 1, I} 2, I 3 ··· I n) or S'K = GK (I _'1 and _{I 2, I '3 ·· I} n), etc .. in general, the appearance of the abnormal state S'K the conventional test device detects the actual cause without resorting to long and difficult additional functions It is a priori due to many possible causes that may not be possible.

In contrast, ATE according to the invention is diagnostic, i.e.
After proving the functional problem of the test circuit by showing at least one abnormal result in the test of the output connection point of the circuit,
It can be used as follows. First, the internal connection points are selected and examined, and all possible states or combinations of desired states are applied to the input connection points of the circuit, while the states of the corresponding connection points are examined and recorded. Remove as many possible fault estimates from the resulting data as possible.

Next, another connection point is selected, and the same procedure is repeated. Thus, the internal connection points are examined one at a time until evidence of the cause of the failure appears.

Those skilled in the art will appreciate that the present invention is not limited to the one embodiment described above that relies on conventional turnaround techniques, including the use of specialized equipment.

Also, those skilled in the art will recognize that although the embodiments described above use only one light beam, the use of several beams is also within the scope of the invention.

Finally, the structure described for testing circuits can be combined with a spectrum analyzer of the signals provided by such circuits. Thus, the device records the signal in real time and the test only considers the presence or absence of a pulse at a given moment. The band of the device which is limited by the band of the used detector is, for example, in the region of 100 MHz.

However, detailed analysis of wavelengths at very large frequencies requires sampling (stroboscope) technology.
This can be achieved by operating as described below with reference to FIG.

A CW laser 901 of all the mentioned types converts the test light into an optical device 9
It is arranged to project toward 02. This device is of the type described above. The apparatus produces and sweeps an analytical light beam 903 that is projected in the direction of the circuit under test 907 across an electro-optic structure comprising, for example, an electro-optic lens and an electro-optic transducer of the type described with reference to FIG. . The reflected light signal is split by a splitter and the detector 914
Conveyed to.

A pulsed second laser 920 is provided which, in response to a test excitation applied to the input 922 of the circuit 907, generates a very short pulse for the duration of the signal occurring at the connection of the circuit. Emit at a given repetition rate.

The light pulse of laser 920 is directed on the one hand to a photodetector 924 and, on the other hand, via a splitter 925 to a reflector 932 through a variable optical delay path, which reflects the pulse of laser 900 light. The output is directed along the same axis to the input of the optical device. Delay path 926 comprises two movable reflectors 927, 928 which bend at right angles to the direction of the beam output of splitter 925.
And a fixed reflector 929 that realigns the beam in the direction of the reflector 932. Pulse laser 920 and optical device 90
The light delay imposed on the beam between 2 and 2
This can be alleviated by moving the 27, 928 and fixed mirrors 929 closer and further away (arrow 930).

The electrical output of the photodetector 924 is connected to a synchronizer 940, which controls a test signal generator 942 connected to the input of the circuit board 907 to convert the electrical pulses to the pulse laser repetition rate. Input is made to the input unit at an equal repetition rate.

The pulse duration of the laser 920 is very short compared to the pulse at the input 22. The signal emitted by the low bandwidth detector 914 compared to the length of the laser pulse is integrated over a number of samples. Appropriate adjustment of the optical path between the pulsed laser 920 and the optical device 902 can be used to inspect each waveform or circuit connection point.

As a modification, an electrical adjustment of the synchronization pulse may be performed. During use in the test, the continuous laser 900 emits light and the pulsed laser is off. In the detailed timing analysis, the laser 900 is turned off and the pulse laser 920 is excited.

[Brief description of the drawings]

1 shows an embodiment of an ATE according to the invention; FIG. 2 shows parts of the device of FIG. 1; FIG. 3 shows a view of the optical assembly; FIG. 5 is a diagram of another interface element; FIG. 6 is another diagram of the present invention.
FIG. 7 is a graph showing characteristics of the electro-optical action with respect to an applied voltage; FIG. 8 is a perspective view showing elements of the ATE of FIG. 6; FIG. 10 is a schematic diagram of an embodiment of a reflector used for inspection; FIG. 10 is a diagram of an embodiment of a light separator; FIG. 11 is a diagram showing an arrangement of a contact element under test; FIGS. 12a and 12b. FIG. 13 shows an anisotropic representation of an electro-optic film; FIG. 13 is a cross-sectional view of a printed circuit board incorporating an electro-optic polymer for testing; FIG. 13b shows a modification of FIG. 13 in some detail. FIG. 14 is a view of an embodiment in which a preformed electro-optical film is arranged on a substrate; FIG. 15 is a plan view of the film used in FIG. 14;
Figure shows another embodiment of the film used in FIG. 14;
FIG. 18 and FIG. 18 are schematic diagrams showing a method for producing a polymer film useful for practicing the present invention; FIG. 19 is a diagram showing a structure for detecting Pockels action by the interferometric Fabry-Perot technique; FIG. 21 is a diagram showing a modification of the technique in FIG. 19; FIG. 21 is a diagram showing a structure for analyzing the waveform of a signal at a circuit connection point in detail. 10 ... electro-optical medium layer, 11 ... printed circuit board, 15 ...
Components, 16, 17 ... conductor, 18 ... interface member, 30 ... laser, 32, 34 ... lens, 33 ... acousto-optic reflector, 36 ... reflective surface, 38 ... test point, 101 ... …
Source 102, positioning means 103, detector 104, beam splitter 111, support, 114 support, 116, conductive film 118, separation protection lacquer film

 ──────────────────────────────────────────────────続 き Continuation of the front page (56) References JP-A-64-18071 (JP, A) JP-A-60-73366 (JP, A) JP-A-63-204172 (JP, A)

Claims (4)

(57) [Claims] An apparatus for testing a circuit, such as a printed circuit, which interconnects electronic components by means of a conductive network formed on at least one surface of an insulating support, the test on the circuit being tested. The size and substantially the size of the circuit, disposed in electrical proximity to the surface supporting the conductive network in use so that incident light can be analyzed by the effect of voltage appearing on the conductor in response to the application of a signal. Electro-optic media layers of the same size, and means for sequentially projecting polarized light onto corresponding individual areas of the electro-optic media layer for individually sampling the surface of the printed circuit, further comprising: A conductive network including an interface member having a plurality of isolated, substantially parallel conductive struts, in electrical proximity of said electro-optic media layer via said interface member Placed on the surface that supports
An apparatus wherein an electric potential present on the surface is relayed to the electro-optical medium layer. 2. The apparatus of claim 1, wherein the surface of the interface member is shaped to fit the surface contour of the circuit to be mated. 3. A transducer for use in voltage testing of at least one area of a circuit, such as a printed circuit, interconnecting electronic components by means of a conductive network formed on at least one surface of an insulating support, An electro-optic polymer film element of a geometry capable of placing a surface in electrical proximity to said at least one area in said conductive network; and said one side of said electro-optic polymer film element. A reflective film overlying, passing through the electro-optic polymer film element to the voltage test zone, and reflecting incident light representing the voltage of the voltage test zone toward the other surface of the electro-optic polymer film element; And a conductive film capable of transmitting test incident light and being maintained at a reference potential. 4. A printed circuit connecting electronic components to one another by means of a conductive network formed on at least one surface of an insulating support, wherein an electrical test is carried out on each of the individual test areas of said conductive network. An optical polymer film element is arranged, and an incidence on the surface of the electro-optic polymer film element close to the test area, passing through the electro-optical polymer film element to the voltage test area and representing the voltage of the voltage test area A printed circuit board provided with a layer capable of reflecting light toward an opposite surface.