Method of testing unloaded, large-area printed circuit boards with a finger tester

0001The present invention relates to a method for testing bare, large-area, printed circuit boards with a finger tester.

0002Devices for testing bare printed circuit boards can basically be divided into two groups, namely the finger tester and the parallel tester.

0003Parallel testers are test devices that use a contact arrangement, which is usually represented by an adapter, to contact all or at least most of the contact points on a circuit board to be tested at the same time. Such parallel testers are preferably used for testing bare printed circuit boards, since they can be used to scan a large number of circuit board test points quickly and reliably.

0004From the <patcit id="pcit0001" dnum="WO0140809A1"><text>WO 01/40809 A1</text></patcit> A method emerges with which areas of printed circuit boards are tested successively, the individual areas having to overlap in such a way that conductor tracks which extend from one area to another area have a contact point in the overlap area. This makes it possible to test such conductor tracks in sections for interruptions.

0005From the <patcit id="pcit0002" dnum="US5045783A"><text>US 5,045,783</text></patcit> shows the use of a scanning electron microscope for testing large circuit boards. The procedure can be used to detect interruptions on printed circuit boards. However, the scanning field of the scanning electron microscope is limited due to a deflection color error and the aberrations of the objective lens, which increase sharply with the deflection angle. The electron beam can therefore only apply and read potentials on an insulated printed circuit board in a narrow range. So that larger circuit boards can also be tested, they are subdivided into fields, the size of which corresponds approximately to the size of the maximum scanning range. Interruptions are detected by charging a contact point in a network and then measuring the associated contact points within the same network. If the contact points are not loaded, there is an interruption. A further test cycle is required to determine interruptions on conductor tracks that span several fields. In the case of two test areas, the first area is positioned in the scanning field of the electron beam measuring device and all the contact points contained in it are charged. The second area is then covered with the scanning field and the charge states of the contact points are tested with an electron probe. The measured state of charge is then compared with the expected state of charge based on the network geometry. If there is no load, there is an interruption. Other fields are tested analogously. However, the areas must first be unloaded.

0006A finger tester is for example in the <patcit id="pcit0003" dnum="EP0468153A1"><text>EP 0 468 153 A1</text></patcit> described. The advantages of the finger testers are that they are very flexible and that different types of circuit boards can be tested without the need for any mechanical modifications. In addition, any type of circuit board can be tested with a finger tester.

0007From the <patcit id="pcit0004" dnum="US3975680A"><text>US 3,975,680</text></patcit> and the <patcit id="pcit0005" dnum="EP0772054A2"><text>EP 0 772 054 A2</text></patcit> it is known to test conductor tracks for interruptions by detecting the electrical capacitance of the corresponding conductor tracks and by comparing them with a reference value. This results in a two-stage test procedure in which a field measurement is carried out in the first stage, with which complex conductance values ​​for the individual conductor tracks are determined on the basis of a first circuit board, and in the second stage only the conductance values ​​of other circuit boards are measured and these are measured with the by means of the guide values ​​determined by the field measurement. In the field measurement, electrical fields are created and the induced potentials or Voltage measured in the conductor tracks. Conductor tracks can be used as antennas for generating the electrical fields. Since these electrical fields only have a spatially limited effective range, the field measurement can be carried out in segments by using one or more conductor tracks of a specific area of ​​the circuit board as antennas for generating an electrical field and measuring the conductor tracks lying in the vicinity of these antennas.

0008There have long been test devices on the market in which the capacitance is measured at all circuit board test points of a conductor track and if these capacitance values ​​are all the same and still match a predetermined reference value, then these conductor tracks have no interruption.

0009From the <patcit id="pcit0006" dnum="US4565966A"><text>US 4,565,966 A</text></patcit> is a method and an apparatus for testing printed circuit boards, in which both the electrical resistance and the electrical capacitance of the individual conductor tracks are measured and compared with corresponding target values. If there are deviations of more than a predetermined amount, this is determined as an error.

0010There are special finger testers for testing bare, large-area printed circuit boards that have a correspondingly large test area. For example, the company MicroCraft KK offers a finger tester under the trade name EM12181, which has a working area of ​​800 mm x 1200 mm. The company Mania Technology AG offers under the trade name Speedy 580 XXL a finger tester that can test printed circuit boards up to a maximum size of 980 mm x 650 mm. Due to their small number, these extra-large finger testers are very expensive to develop, produce and operate. Such test devices are necessary for testing extra-large printed circuit boards with an edge length of more than 500 mm, since these printed circuit boards must also be tested reliably.

0011There is therefore a considerable need for an inexpensive solution for testing bare, large-area printed circuit boards.

0012The invention has for its object to provide a method for testing bare, large-area, printed circuit boards, which can be carried out with an inexpensive test device and with which the conductor tracks can be tested reliably.

0013The object is achieved by a method having the features of claim 1. Advantageous refinements are specified in the subclaims.

0014A finger tester is used in the method according to the invention for testing bare printed circuit boards with bare surfaces, wherein<ul id="ul0001" list-style="dash" compact="compact"><li>the circuit board is tested divided into several segments, the individual segments being tested in succession in a test area of ​​the finger tester until the complete circuit board has been tested, the test area being the entire area which can be covered by contact fingers of the finger tester and in the test electrodes the contact finger can contact the circuit board and the circuit board is larger than the test area, and</li><li>When testing a segment of the printed circuit board, conductor tracks that extend beyond the respective segment to be tested are tested for interruptions by capacitive measurement of the end points of these conductor tracks located in the segment, a capacitive measured value being recorded in each case after the final testing of the one Segment the circuit board is moved to position it with the next segment in the test area</li><li>all the capacitive measured values ​​belonging to a conductor track form a group and the capacitive measured values ​​within each group are compared with one another, and if at least one capacitive measured value differs from other capacitive measured values ​​of the respective group by a predetermined value, this is assessed as an interruption.</li></ul>

0015With the method according to the invention, conventional finger testers with a typical test area of ​​600 mm × 600 mm, which are produced in large numbers, can be used. The method according to the invention can also be used with finger testers that are already present. The method according to the invention can be used to test printed circuit boards which are larger than the test area of ​​the finger tester and, for example have a size of 1,200 mm x 600 mm, 800 mm x 1,200 mm, 800 mm x 1,500 mm or 600 mm x 800 mm.

0016In the method according to the invention, long conductor tracks, which extend over several segments, are tested for interruptions by means of a capacitive measurement. Since only a single conductor end point is to be contacted for a capacitive measurement, it is possible to test several segments of the circuit board independently of one another during the test procedure and then to evaluate the measured values ​​belonging to one conductor path together. The invention is also based on the finding that when there is an interruption and / or a high-resistance section in the conductor track, the capacitance of a long conductor track changes significantly, so that an interruption can be reliably detected with a capacitive measurement. Interruptions in the sense of the present invention also include sections in the conductor tracks with a resistance value of at least 1 megohm (1 MΩ).

0017With small conductor tracks, the capacitive measurement on interruptions is not always reliable. However, small conductor tracks are usually completely within the segments, which is why they can then be conventionally examined for interruptions using an ohmic measurement. The segments should preferably be arranged such that they overlap in such a way that all the small conductor tracks or sections of small conductor tracks that can be tapped off are arranged completely within a segment.

0018The invention is explained in more detail below using the drawings as an example. The drawings show:<dl id="dl0001"><dt>Fig. 1</dt><dd>schematically the structure of a finger tester,</dd><dt>Fig. 2</dt><dd>the division of a circuit board to be tested into individual test segments,</dd><dt>Fig. 3</dt><dd>the method according to the invention schematically in a flow chart, and</dd><dt>Fig. 4</dt><dd>schematically the structure of a test device with an automatic conveyor.</dd></dl>

0019The method according to the invention for testing bare, large-area printed circuit boards having conductor tracks is carried out in a finger tester. A finger tester has a plurality of test electrodes 1 which can be contacted with circuit board test points of the conductor tracks 2 of a circuit board 3 (<figref idref="f0001">Fig. 1</figref>; <figref idref="f0002">2</figref>). A test electrode 1 is integrated in a contact finger 4. The present finger tester is a multi-finger system which has a total of twelve contact fingers 4, six contact fingers being arranged on one side of the circuit board 3 to be tested. A finger tester in which the contact fingers can be moved independently of one another is in the<patcit id="pcit0007" dnum="EP0468135A1"><text>EP 0 468 135 A1</text></patcit> described. The contact fingers 4 can be moved parallel to the surface of the printed circuit board 3, so that the test electrodes 1 can be contacted with certain printed circuit board test points of the conductor tracks. The contact fingers 4 are each fastened to a carriage 6 which is controlled by a position control device 5 and which can be moved in one plane parallel to the printed circuit board 3. The carriages 6 are each provided with a vertically aligned actuating cylinder 7 with which the contact fingers 4 can be rotated about the vertical axis. Furthermore, a tilting device is integrated on the contact fingers 4, so that the fingers can be lowered onto the printed circuit board 3 with the test electrodes 1 arranged at their tips.

0020The position control device 5 receives its signals from a central control unit 8 to control the movement of the contact fingers 4. To generate the measurement signal, a function generator 9 is provided, which generates a measurement voltage with a certain amplitude U.₀ and certain frequency f₀ to the test electrode 1 of a contact finger 4 for feeding into a conductor track of the circuit board 3 to be tested. In the present embodiment, the frequency is f₀ between 0 and 2000 Hz.

0021The finger tester has a first, second and third evaluation device 10, 11, 12. The first evaluation device 10 is used to carry out an interruption test by means of an ohmic measurement. Here, two circuit board test points of a conductor track are contacted by two test electrodes 1 at the same time and a direct current generated by the function generator 9 is applied between these two test electrodes 1. The first evaluation device 10 detects the voltage falling between the two test electrodes 1 and uses this to determine the electrical resistance of these conductor tracks or this conductor track section. If the resistance is above a predetermined threshold value of, for example, 100 ohms, the conductor track is assessed as having an interruption.

0022The second evaluation device 11 is used to perform short-circuit tests between two conductor tracks, an ohmic measurement being carried out here between two circuit board test points of two different conductor tracks. If the resistance determined is below a predetermined threshold value, for example 100 MΩ, this is assessed as a short circuit between the two conductor tracks.

0023Instead of an ohmic measurement, the short-circuit test can also be determined by means of a so-called field measurement, as described in the <patcit id="pcit0008" dnum="US3975680A"><text>US 3,975,680</text></patcit> or the <patcit id="pcit0009" dnum="EP0772054A2"><text>EP 0 772 054 A2</text></patcit> is described, in which the measured complex conductance values ​​of the conductor tracks can be used to determine whether there is a short circuit between two conductor tracks.

0024The third evaluation device 12 is used to carry out the interruption test by means of capacitive measurement. Here, a conductor track is only electrically contacted with a single test electrode 1 at one of its circuit board test points. A measurement signal with a predetermined frequency is fed in. Another test electrode 1 is either contacted with a flat test electrode arranged parallel to the circuit board 3 or contacted with a further conductor track of the circuit board 3. Both the test electrode and the further conductor track 3 act as a kind of antenna or Counter electrode to the conductor track to be tested, into which a measurement signal to be detected which is fed in by the measurement signal fed into the conductor tracks to be tested is induced and which is detected by the third evaluation device 12. On the basis of this detected measurement signal, the respective capacitance C_i be determined. In the context of the present invention, a capacitance is also understood to mean a complex conductance and a capacitive measurement is also a measurement of the complex conductance, provided that the complex proportion of the conductance does not approach zero.

0025In the method according to the invention for testing bare, large-area printed circuit boards 3, the printed circuit boards are divided into several “virtual” segments. In<figref idref="f0002">Fig. 2</figref> A circuit board 3 is shown, in which a few conductor tracks 2 are shown in a schematically simplified manner. This circuit board has a length L of 900 mm and a width B of 440 mm. This printed circuit board is divided into 3 segments I, II, III, the first segment I extending 320 mm to the right from the left boundary edge and the third segment III extending 320 mm to the left from the right boundary edge. The second or middle segment II is arranged in the central area and overlaps with the two outer segments I and III in each case in an overlap area of ​​40 mm. The second segment II thus has a width of 340 mm. The overlap regions 13 thus have a width b of 40 mm.

0026Large circuit boards have a typical size of e.g. 1200 mm x 600 mm, 800 mm x 1200 mm, 800 mm x 1500 mm or 600 mm x 800 mm. Printed circuit boards with an area of ​​less than 600 mm x 500 mm are not considered to be large-surface printed circuit boards.

0027The method according to the invention is described below using the method shown in <figref idref="f0003">Fig. 3</figref> shown flowchart explained in more detail.

0028The method begins with step S1.

0029The individual segments are determined in step S2. For this purpose, the data describing the conductor tracks of the printed circuit board 3 (Gerber data) are used, from which it is determined where the boundary regions between adjacent segments can be arranged. The relevant criteria are explained in more detail below.

0030In step S3, the circuit board 3 to be tested is arranged with a segment I, H, III in a test area of ​​the finger tester. The test area is the entire area that can be covered by the contact fingers 4 and in which the test electrodes 1 can contact the printed circuit board 3. The test area typically has a size of 600 mm x 600 mm.

0031In step S4, all conductor tracks that lie completely within this segment or whose sections lie entirely within the respective segment are tested for interruption by means of an ohmic measurement (S4). In segment I from<figref idref="f0002">Fig. 2</figref> the tracks 2/1, 2/2 and 2/5 are completely within the segment. Furthermore, a section of the conductor track 2/4 lies completely within the segment I. This section extends from the left end point or circuit board test point of the conductor track 2/4 to the circuit board test point of the conductor tracks 2/4, which lies within the overlap region 13.

0032Then the conductor tracks within the segment are tested for short circuit in a conventional manner by means of an ohmic measurement (S5).

0033In step S6, the conductor tracks, which extend beyond the segment to be tested and cannot be scanned in sections, such as conductor track 2/4, are tested for interruption by means of capacitive measurement. The capacitance values ​​of the conductor tracks are measured at all circuit board test points, i.e. the contact points connected to the respective conductor track. At the in<figref idref="f0002">Fig. 2</figref> Printed circuit board 3 shown extends the conductor track 2/3 over all segments. The capacitance values ​​are recorded from the circuit board test points connected to this circuit 2/3.

0034If all measurements have been carried out on a segment, a check is carried out in step S7 as to whether another segment is to be tested. If this is the case, the process sequence goes to step S3 and the circuit board with the further segment is arranged in the test area of ​​the finger tester. The operations in steps S4, S5, S6 are carried out accordingly.

0035If it is determined in step S7 that the measurements have been carried out in all segments, the method sequence goes to step S8, in which the measurement results are evaluated. The capacitive measured values ​​are evaluated here in particular. The capacitive measured values ​​of a conductor track each form a group. The measured values ​​within a group are compared with one another and if at least one measured value differs from other measured values ​​of the respective group by a predetermined value by, for example, more than 0.5% of the electrical capacity, this is assessed as an interruption. It is also possible for a further comparison to be carried out with a predetermined reference value, which is carried out beforehand by means of a faultless printed circuit board (“golden board”) or using a method as described in the<patcit id="pcit0010" dnum="EP0772054A2"><text>EP 0 772 054 A2</text></patcit> is explained, has been determined.

0036In the example above, the conductor tracks 2/3 and 2/6 are tested for interruption by means of capacitive measurement. The remaining conductor tracks 2/1, 2/2, 2/4, 2/5, 2/7 are tested with an ohmic measurement. Conductor 2/4 is tested section by section during testing within segments I and II.

0037The minimum length of a conductor track, which is capacitively tested for interruptions, is typically 30 mm-60 mm for the segments arranged in the exemplary embodiment, which corresponds to the width b of the overlap regions 13. The minimum length of a capacitively divisible circuit board depends on the electrical capacity of the respective conductor track, which in turn depends on the materials used and the design of the conductor track.

0038The electrical capacitance of typical conductor tracks is in the range of a few pF. Therefore, at the commonly used measuring frequencies of 2 kHz to 64 kHz, resistance values ​​of at least 100 KΩ can be detected with a capacitive measurement. The relationship between the electrical capacitance and the minimum detectable resistance is described by the following formula:<maths id="math0001"><math display="block"><mi>R</mi><mo>≈</mo><mfrac><mn>1</mn><mrow><mi>ω</mi><mo>⋅</mo><mi>C.</mi></mrow></mfrac><mo>⁢</mo><msqrt><mfrac><mn>1</mn><mi>k</mi></mfrac><mo>-</mo><mn>1</mn></msqrt><mo>,</mo></math><img file="EP1920263B1_D0001.tif" /></maths>where R is the resistance value, C is the capacitance, w is the angular frequency and k = 0.07.

0039<figref idref="f0004">Fig. 4</figref> shows schematically simplified a device for automatic testing of large-area printed circuit boards, which comprises a finger tester with test electrodes 1, which are each arranged on a carriage 6. The test electrodes 1 can be moved towards the printed circuit board 3 and away from the printed circuit board 3 in order to contact individual printed circuit board test points. The carriages 6 are arranged horizontally displaceably on cross members 14. A test unit with crossmember 14, slide 6 and test electrodes 1 is provided for the upper and lower side of a printed circuit board 3 to be tested.

0040A conveyor for transporting the printed circuit board 3 to be tested is arranged between these two test units. This conveyor device has two narrow conveyor belts 15 which are arranged running parallel to one another. The circuit board 3 to be tested is placed on the conveyor belts 15.

0041Clamping devices 16 are arranged in the test area between the two test units, with which the circuit board 3 to be tested can be fixed in the test area or between the two test units. The clamping devices are automatically opening and closing clamps, so that the clamping can be carried out automatically by means of the control device. If the circuit board 3 is fixed in the test area by means of the clamping device 16, the conveyor belts 15 are lowered somewhat and laterally out of the test area (in the direction perpendicular to the plane of the drawing) <figref idref="f0004">Fig. 4</figref>) moved out so that the test area is free of the conveyor belts 15. Then the circuit board test points of a segment of the circuit board 3 can be tested. After the final testing of the segment, the conveyor belts 15 are moved again under the printed circuit board 3, the clamping device is released, the printed circuit board 3 is moved further in the conveying direction (arrow 17) in order to position it with the next segment in the test area. It is thus possible to automatically test the several segments of a printed circuit board 3 in succession.

0042The invention can be briefly summarized as follows:<ul id="ul0002" list-style="none" compact="compact"><li>The invention relates to a method for testing bare, large-area printed circuit boards with a finger tester.</li></ul>

0043According to the method according to the invention, the printed circuit boards are tested divided into several segments, conductor tracks which extend beyond a segment being tested by capacitive measurement of the end points located in the respective segment, an interruption of the conductor track being determined if a measured value of the capacitive measured values ​​belonging to a conductor track and which differ significantly from other measured values.

0044The method according to the invention can thus be used to test printed circuit boards which are larger than the test area of ​​the finger tester, the individual segments being arranged in succession in the test area of ​​the finger tester. The maximum size of the individual segments corresponds to the test area of ​​the respective finger tester. As a result, it is not necessary to provide a special finger tester for large-area printed circuit boards, which is considerably more expensive than finger testers in a test area with a standard size.

0045In the method according to the invention, a circuit board to be tested with the segments that are not tested can protrude from the test area of ​​the finger tester.

Reference list

0046<dl id="dl0002" compact="compact"><dt>1</dt><dd>Test electrode</dd><dt>2</dt><dd>Conductor track</dd><dt>3</dt><dd>Circuit board</dd><dt>4</dt><dd>Contact finger</dd><dt>5</dt><dd>Position control device</dd><dt>6</dt><dd>Sledge</dd><dt>7</dt><dd>Actuating cylinder</dd><dt>8</dt><dd>central control unit</dd><dt>9</dt><dd>Function generator</dd><dt>10</dt><dd>first evaluation device</dd><dt>11</dt><dd>second evaluation device</dd><dt>12</dt><dd>third evaluation device</dd><dt>13</dt><dd>Overlap area</dd><dt>14</dt><dd>traverse</dd><dt>15</dt><dd>Conveyor belt</dd><dt>16</dt><dd>Clamping device</dd><dt>17</dt><dd>arrow</dd></dl>