Test method for a circuit board substrate and circuit board particularly suitable for use in such a test method

Abstract

Testing method for a circuit board substrate (4), in particular an RF substrate for a circuit board (2) of a radar system in a vehicle, wherein the circuit board substrate (4) has at least one material parameter to be tested, wherein in the testing method the material parameter is selected from a number of parameters, comprising a loss angle and a dielectric parameter, wherein - a circuit board (2) is provided with a test area (20) made from the circuit board substrate (4), in which a measuring structure (22) is formed, which has a measuring input (24) and a measuring output (26), - the measuring structure (22) is provided as a waveguide (28) having a waveguide space (30) arranged in and filled with the circuit board substrate (4), - an input signal with a predetermined measuring frequency is applied to the measuring input (24), - at least one value (52, 54) of a test parameter of the measuring structure (22) is determined, - at least one reference range (60, 60', 60") is specified for the value (52, 54), - and check whether the value (52, 54) is within the reference range (60, 60', 60").

Description

The invention relates to a test method for a printed circuit board substrate, in particular an RF substrate for a printed circuit board of a radar system in a vehicle, wherein the printed circuit board substrate has at least one material parameter to be tested, in which the material parameter is selected from a number of parameters comprising a loss angle and a dielectric parameter. The invention further relates to a circuit board which is particularly suitable for use in such a test method.Such a test method is given, for example, in the IPC-TM-650 Handbook for Test Methods in section 2.5.5.5.Radar systems for vehicles, for example passenger motor vehicles or construction machines and the like, typically comprise electronic components which transmit and / or process, in particular, high-frequency signals, so-called RF signals. Such electronic components are usually manufactured on a circuit board from a suitable circuit board substrate, i.e. an RF substrate. Lines are arranged on or in this circuit board substrate, which connect the electronic components to one another for guiding or transmitting the signals. In this case, RF substrate is understood to mean a substrate which is suitable for conducting RF signals.Since the electromagnetic field assigned to a signal is usually not limited to the respective line, but also extends into a space surrounding the line, the transmission of the signal is also influenced in particular by the circuit board substrate. In this case, the circuit board substrate is characterized in particular by material parameters which are often also frequency-dependent. Of particular interest in this connection are the dielectric parameter and the so-called loss angle. Dielectric parameters are understood in particular to mean the electrical permittivity of the corresponding material. The permittivity and generally the dielectric parameter are also referred to as ε r or Dk. Furthermore, the loss angle represents a measure of the absorption of the signal in the circuit board substrate. The loss angle is also referred to as tand or Df. An RF substrate advantageously has the smallest possible loss angle for high-frequency signals; a conventional circuit board substrate typically has a larger loss angle for such signals than an RF substrate. Usually, an RF substrate has a loss angle that is significantly less than 0.01. In contrast, a conventional, i.e. a non-RF substrate usually has a loss angle greater than 0.01. For example, the loss angle of the board substrate designated by the material identifier FR4 is significantly greater than 0.02 at a frequency of 76.5 GHz.During the production of a plurality of printed circuit boards, it is possible that the respectively provided printed circuit board substrate is nominally of the same type, but the material parameters actually differ, in particular in the case of different batches. In order to ensure the intended functionality of each of the printed circuit boards at least within a predefined tolerance range, the material parameters must, however, lie in predefined value ranges. Therefore, it is necessary to test the material parameters preferably of each board, but at least of each batch of board substrate.A possible test method for determining the actual material parameters of a circuit board substrate having the features of the preamble of claim 1 is specified, for example, in the IPC-TM-650 Handbook for test methods. There, in particular in section 2.5.5.5, a strip conductor test for determining the permittivity, i.e. the dielectric parameter and the loss angle, is described. The designation of the method therein is: stripline test for permittivity and loss tangent, also referred to below as IPC method.In this case, a test structure is used in which a resonator circuit having predetermined dimensions is clamped between two layers of a printed circuit substrate to be examined; this stack is in turn clamped between two ground layers. As a result, a strip conductor, so-called strip line, is formed, on which a number of measurements are carried out in the following.For this purpose, the resonator circuit has a measurement input which is supplied with an input signal with an adjustable frequency. Furthermore, the resonator circuit has a measurement output, by means of which the so-called S 2,1- parameter of the resonator circuit can be measured in combination with the measurement input. This S 2,1- parameter in particular indicates the forward transmission of the resonator circuit. In other words: the S 2,1- parameter has an amplitude which corresponds to the ratio of the output signal output at the measurement output to the input signal input at the measurement input. To determine the material parameters, it is necessary to determine this amplitude at specific measurement frequencies and likewise to record the associated measurement frequencies. These data points are then used to calculate the dielectric parameter and loss angle.In the presented IPC method, measurements of the amplitude of the S 2,1- parameter are carried out in particular close to or at the resonance frequency of the resonator circuit. However, its behavior near resonance is particularly dependent on the actual dimensions of the resonator circuit. In particular, these dimensions therefore also enter into the calculations for determining the material parameters. This also means, in particular, that the result is influenced accordingly by manufacturing inaccuracies of the test setup. In particular, inaccuracies which result from etching of conductor structures for forming the resonator circuit influence the dimensions thereof and correspondingly also the measurement result. Therefore, the most accurate possible determination of the actual dimensions is necessary and, in particular, also a part of the IPC method; this is consequently correspondingly complicated to carry out.For testing at different frequencies, the dimensions of the above-described resonator circuit must be adjusted in particular in a frequency-dependent manner. The resonant frequency of the resonator circuit is to be adapted in particular to the frequency of the intended application. At high frequencies, resonator circuit dimensions are typically smaller and inaccuracies in fabrication result in larger measurement errors. Therefore, the IPC method is limited to determining the material parameters up to a specific frequency, in particular of about 12 GHz, i.e. for higher frequencies the IPC method becomes more inaccurate. However, radar systems are also known which are operated at frequencies, for example, in the vicinity of 24 GHz or 77 GHz. The frequency range covered by the IPC method for determining the material parameters is therefore not sufficient, in particular for testing circuit board substrates for radar systems with operating frequencies significantly above 12 GHz, for example 24 GHz or 77 GHz.Furthermore, the IPC method is complicated to carry out, since the test setup has to be assembled in accordance with the above description. For this purpose, in particular samples of the circuit board substrate must be provided without any coating. The method is therefore not suitable in particular as an online test method, that is to say for testing during the production of a printed circuit board.As an alternative to the stripline resonator circuit proposed in the IPC method, a resonator circuit can also be used which is based on a microstrip ring resonator, also called microstrip ring resonator. However, a microstrip conductor has higher losses than, for example, a strip conductor, since the microstrip conductor is typically applied to the upper side of a printed circuit board and can correspondingly radiate into the space lying above it. Furthermore, it also applies to the microstrip ring resonator that precise knowledge of the dimensions is necessary for interpreting the measurement results and inaccuracies have a corresponding effect, in particular during the etching of the microstrip ring resonator. In this case too, additional measurements must consequently be carried out in order to determine the actual dimensions of the conductor structures.Furthermore, EP 0 921 404 A2 discloses a measurement method for the high-frequency characteristics of dielectric materials, in which, for electrical materials, the amplitude and phase changes of the electric field component of an electromagnetic wave along its propagation direction on a high-frequency transmission line are measured. The high-frequency transmission line is designed as a microstrip line. In addition, a device whose refractive index changes according to the electric field is used to obtain an effective dielectric constant and a dielectric loss of the transmission line.Furthermore, DE 11 2009 000 784 T5 discloses a high-frequency module having a substrate, a circuit board and a waveguide. The substrate has an input-output section for high-frequency signals formed on a surface thereof. The circuit board has a dielectric waveguide line with its end face exposed, and is disposed on the one surface of the substrate such that a virtual plane that goes beyond the end face is cut by the one surface of the substrate. The waveguide has openings at the ends thereof, in which one of the openings is connected to the end face of the dielectric waveguide line and the other opening is connected to the input-output portion of the substrate.It is therefore an object of the invention to specify a test method for a circuit board substrate which is simple to carry out and in particular avoids at least one of the disadvantages mentioned above. The test method is intended to be suitable in particular for testing a printed circuit board substrate for a printed circuit board of a radar system in a vehicle. Furthermore, a circuit board suitable for such a test method is to be specified.The object set with regard to the test method is achieved according to the invention by a test method having the features of claim 1.In a test method for a printed circuit board substrate, in particular for an RF substrate for a printed circuit board of a radar system in a vehicle, the printed circuit board substrate has at least one material parameter to be tested. In this case, this material parameter is selected from a number of parameters, comprising a loss angle and a dielectric parameter. Furthermore, the circuit board is provided with a test region made of the circuit board substrate. Here, "produced from the circuit board substrate" is understood in particular to mean that the test region is in particular not exclusively produced from the circuit board substrate.The test method is suitable in particular for testing the circuit board at a frequency which is not accessible by means of previously known methods, for example the IPC method described at the beginning. Advantageously, it is possible to test the circuit board by means of the test method with reduced effort compared to the IPC method. Expediently, the test method is not limited to a specific frequency range by manufacturing inaccuracies of the measurement conductor, but manufacturing inaccuracies during the production of conductor structures applied for forming the measurement conductor have little or no effect on the test method. For example, inaccuracies in the etching of conductor structures provided for forming the measurement conductor are of minor importance in comparison with the influence of fluctuations in the material parameter. In particular, this advantageously eliminates the need for determining the exact dimensions of the measurement conductor.In the test region, a measurement structure, which is in particular embodied as a measurement conductor and has a measurement input and a measurement output, is furthermore embodied. The measuring structure is a fixed, integral component of the circuit board. In the following, without limiting generality, the measurement structure is referred to as a measurement conductor. In the test method, an input signal with a predetermined measurement frequency is applied to the measurement input and a test parameter, in particular an S parameter of the measurement conductor, is determined. At least one value of the test parameter is detected. In the case of an S parameter, it comprises a phase and an amplitude and at least one value of the phase or of the amplitude is determined. Furthermore, at least one reference range for the value is specified and it is checked whether the value lies within the reference range.The reference range is used to test the material parameter on the basis of the measured value and to determine a test result. This expediently specifies a value range in which the value of the amplitude or of the phase is preferably intended to lie. In particular, the reference range also extends over a predefined frequency interval. In this case, the reference region is advantageously assigned a tolerance range of this material parameter on the basis of the dependence of the S parameter on the material parameter. If the measured value lies outside the reference range, the test result is negative, i.e. the circuit board substrate does not meet the requirements of the application in particular. If the measured value lies within the reference range, on the other hand, the test result is positive, that is to say the circuit board substrate is suitable in particular for the application. In this way, for example, a board provided with a negative test result can be removed from the further manufacturing process or generally pulled out of the traffic.The circuit board is frequently a multilayer circuit board and comprises a plurality of substrate layers each made of a dielectric material, between which a layer made of an electrically conductive material, for example copper, is arranged in each case. In this way, a stack is formed, wherein the substrate layers can be made of different materials. For high-frequency applications, such as radar systems, typically at least one substrate layer made of an RF substrate is provided. Since such an RF substrate is usually more expensive than a conventional circuit board substrate, for example, only one substrate layer is made from the RF substrate and further substrate layers from a conventional circuit board substrate. Any radio-frequency electronics provided and lines connecting them are then preferably arranged on the RF substrate or on a layer applied to the RF substrate. In this way, in particular high-frequency signals can be transmitted by means of the lines with particularly low loss.Suitable high-frequency conductors are used as lines, for example waveguides integrated into the circuit board substrate, so-called substrate integrated waveguides (SIW). When transmitting a signal by means of such a line, the associated electromagnetic field interacts in particular with the circuit board substrate surrounding the respective line. The transmission properties consequently also depend in particular on the material parameters of the circuit board substrate. The measurement conductor is therefore provided as a waveguide. This has a waveguide space which is arranged in the circuit board substrate and is filled with the latter. In particular, the waveguide is thus advantageously designed as a substrate integrated waveguide (SIW). The hollow conductor space is bounded laterally by hollow conductor walls made of conductive material. For example, an upper and a lower waveguide wall are each part of a ground layer applied above or below the circuit board substrate. Lateral hollow conductor walls are formed, for example, by means of plated-through holes, so-called vias, or also by means of trenches metallized on the inside, so-called grooves. A signal guided through the waveguide is thus guided in particular through the circuit board substrate to be tested and influenced as a function of its material parameter.Advantageously, the through-contacts of the respective waveguide wall are arranged at a predetermined distance from one another. This is selected in particular as a function of the measurement frequency. In the case of a higher measurement frequency, the distance is expediently selected to be smaller.The use of a waveguide as a measuring conductor has the particular advantage that, in comparison with the method described in the introduction, the S parameter is less dependent on manufacturing tolerances of the waveguide. In other words: When etching the conductor structures used to form the waveguide, i.e. in particular the waveguide walls, inaccuracies have a less pronounced effect on the S parameter than, for example, in one of the above-mentioned methods with a resonator circuit comprising a strip conductor or microstrip conductor. In addition, the S parameter and therefore also its amplitude and phase is substantially dependent on the material parameter. This simplifies the test method in a particularly simple manner, since in particular the precise knowledge of the dimensions of the waveguide and of its individual parts is of minor importance.A material parameter is, for example, the so-called loss angle, which represents a measure of an absorption caused by the circuit board substrate, i.e. describes an attenuation of the signal. Another material parameter is the dielectric parameter, which is in particular the relative permittivity or dielectric conductivity of the circuit board substrate. Both material parameters mentioned are typically frequency-dependent. In other words, the value of the material parameter is typically a function of the frequency of the signal. In particular, it is therefore advantageous to explicitly check the respective material parameter at the frequency of the corresponding application and to specify a test method suitable for this purpose.Since the circuit board is preferably provided for a specific application and is to be installed accordingly, a test region is provided on the circuit board for carrying out the test method. This is arranged, for example, in a corner of the circuit board and comprises a cutout of that substrate layer which is produced from the circuit board substrate to be tested. For a further embodiment of the test region, in particular, conductor structures made of a conductive material are also applied to the circuit board substrate and possibly additional components, for example mechanical or electronic components.The measuring conductor arranged in the test area is expediently designed as a so-called two-port, i.e. comprises an input port and an output port, which is referred to here as a measuring input or measuring output. The measuring conductor connects the measuring input and the measuring output for transmitting a signal. The input signal applied to the measurement input is then influenced in particular by the circuit board substrate when transmitted to the measurement output. The input signal has a predetermined measurement frequency, which preferably corresponds approximately to the frequency of the later application.For testing the material parameter of the circuit board substrate, also referred to below for short: for testing the circuit board, a so-called S parameter is preferably measured. In the case of a two-port, its transmission properties are characterized in particular by a scattering matrix having four elements which are each referred to as S parameters. These describe in particular in each case the ratio of the signals input and output at the measurement input and measurement output with respect to the input signal. Each S parameter typically comprises an amplitude and a phase for quantifying the amplitude ratio and the phase relationship of the respective signal with respect to the input signal.According to the test method, at least one value of the amplitude or the phase is measured. Expediently, a plurality of values are measured, for example at different frequencies, as a result of which the printed circuit board can be checked in particular in a specific frequency range. Furthermore, values of both the amplitude and the phase are advantageously measured. For this purpose, for example, a so-called network analyzer is used, which is connected to the measurement input and the measurement output by means of suitable high-frequency cables.The amplitude or the phase are usually dependent on the material parameter, as a result of which it is expediently possible to indirectly determine the material parameter. Here, indirect is understood to mean that instead of a value of the material parameter, a value of a parameter dependent thereon, namely for example the amplitude or the phase of the S parameter, is measured. That is, depending on a change in the material parameter, for example, due to variations in the mixing ratio of the materials used in the fabrication of the board substrate, the measured value of the amplitude or the phase also changes.In particular, on account of the test region arranged on the printed circuit board, the test method described above is particularly suitable for testing printed circuit boards which are already either designed or at least provided for a specific application or use. For example, a printed circuit board is installed in a radar system or is already in operation and is intended to be subjected to a subsequent inspection. With regard to the specific application or use, in an advantageous embodiment, the circuit board comprises a functional region in addition to the test region. Functional elements assigned to the application of the circuit board are expediently arranged in the latter. In the case of a printed circuit board for a radar system of a vehicle, these are, for example, RF transmitters and receivers, suitable antennas, amplifier components and suitable lines for connecting these elements.Preferably, none of the elements of the functional region is connected to the measurement conductor, whereby in particular an impairment of the measured value by the functional elements is reduced, preferably avoided. The functional region and the test region may nevertheless have one or more common ground layers.In an alternative embodiment, however, the test region is part of the functional region and the measurement conductor is a line arranged in the functional region. This allows a particularly compact design of the circuit board.For simple and uncomplicated testing of the circuit board, in a preferred development, the measurement input and the measurement output are each connected to a measurement connection arranged in the test region. This measurement terminal is, for example, a suitable high-frequency plug connector for a coaxial cable or a measurement contact for contacting by means of a test tip. As the high-frequency connector, for example, a coaxial connector of SMA type is used. In contrast, in order to form a measuring contact, a suitably dimensioned conductor surface is preferably connected to the measuring input and the measuring output. The configuration of the measuring terminal as a measuring contact is particularly suitable if no suitable high-frequency plug connector is available for the frequency used.An output signal is preferably measured at the measurement output and the output signal is compared with the input signal in order to determine the S parameter. In other words: the S parameter is expediently the S 2,1- parameter or else forward transmission parameter. The S 2,1- parameter thus indicates the ratio of the signal coming out at the measurement output, i.e. the output signal to the input signal coming into the measurement input. In particular, in this case the phase of the S parameter indicates the phase shift which the input signal experiences when passing through the measurement conductor; the amplitude indicates in particular the attenuation of the input signal.The test method is expediently used as an online test method during a production process, in particular during the production of a radar system for a vehicle. For this purpose, the measurement conductor is advantageously applied together with the conductor structures provided for the application of the circuit board. This in particular saves a process step in such a way that a conductor structure suitable for testing the circuit board does not have to be applied separately, but is already integrated into the circuit board. All the boards within a production process therefore preferably have the test area.Advantageously, the test method is carried out automatically. At a suitable point in the course of the production process, for example before the application of electronic components, it is then possible to test the board and, in particular, remove it from the production process. In particular, the production costs can thereby be reduced. In addition, it is advantageously possible to test each circuit board separately, instead of testing a sample of the circuit board substrate representative of a plurality of circuit boards produced therefrom. The use as an online test method also eliminates the need, in particular, for a separate test method, as a result of which the production costs are expediently reduced.Advantageously, the test method is used for quality control of a number of printed circuit boards, in particular printed circuit boards of a radar system for a vehicle. In particular, due to the possibility of separately testing each board on the basis of the test area arranged thereon, the test method for targeted quality control can be used, wherein targeted is understood to mean that a specific board is being tested and, in particular, not a representative test setup.Preferably, a subsequent testing of the circuit board is possible, for example for fault finding in the case of defective systems which have a corresponding circuit board. Alternatively, the quality control can also be suitably used in combination with the above-mentioned embodiment as an online test method.The measurement frequency of the input signal is expediently selected from a frequency range from 20 GHz to 120 GHz, as a result of which in particular the material parameter can be checked in an application-relevant frequency range, but cannot be checked using the method described at the beginning. It is generally conceivable that the test method can also be used for frequencies below 20 GHz and above 120 GHz. In this case, the IPC method described at the beginning can be replaced by the test method described here. This test method is thus improved in particular with respect to the covered frequency range compared to the known methods.The measuring conductor preferably has an operating frequency range which is characterized in particular by a lower and an upper limit frequency. If the measurement frequency is selected from the operating frequency range, the loss angle is advantageously substantially proportional to the amplitude of the S parameter. For checking the loss angle, the measurement frequency of the input signal is therefore advantageously selected from the operating frequency range and a value of the amplitude, also called an amplitude value, is measured.Expediently, the phase of the S parameter is proportional to the dielectric parameter for measurement frequencies from the operating frequency range. Therefore, in an advantageous development for checking the dielectric parameter, the measurement frequency of the input signal is selected from the operating frequency range and a value of the phase, also called phase value, is measured.In an advantageous development, the measurement conductor has a cut-off frequency range. This does not correspond in particular to the operating frequency range and also does not overlap with it. That is, a particular frequency is associated with either the cut-off frequency range, the operating frequency range, or neither. Suitably, in the cut-off frequency range, the dielectric parameter and the amplitude are each substantially proportional to the measurement frequency. For checking the dielectric parameter, the measurement frequency of the input signal is therefore expediently selected from the cut-off frequency range and a value of the amplitude is measured.In a preferred development, the measurement frequency of the input signal is changed, in order to measure a plurality of values at different measurement frequencies. In particular, the amplitude or the phase is measured as a function of the measurement frequency. From the resulting curve, the cut-off frequency range and the working frequency range can then advantageously be identified. In particular, the amplitude value measured in decibels is substantially constant in the operating frequency range and falls in the cut-off frequency range for decreasing frequencies.The reference range is expediently selected as a function of the tested material parameter. In this case, the reference range when checking the dielectric parameter is a value range with an upper and a lower limit value. In other words, the reference range extends from a lower limit to an upper limit, and thus defines a band. Accordingly, the tolerance range for the dielectric parameter is a band with an upper and a lower limit value for the dielectric parameter.Since the loss angle should preferably be as small as possible, the reference range is expediently a value range above a limit value when checking the loss angle. The reference range thus has in particular only a lower limit which defines a maximally tolerated loss angle. In particular, the lower limit applies to the measured values of the S parameter.Suitably, the reference range is characterized by at least one threshold value which is frequency dependent. Since, for example, in the case of checking the dielectric parameter, the measured phase or amplitude is frequency-dependent, the limit value is also expediently frequency-dependent.The object relating to the circuit board is achieved according to the invention by a circuit board having the features of claim 14. The circuit board is suitable for use in a test method according to one of the above-mentioned embodiments. The circuit board is manufactured from a circuit board substrate and comprises a functional region and a test region. The latter is manufactured from the circuit board substrate and comprises a measurement structure designed as a hollow conductor. This has a measurement input and a measurement output for measuring a value of a test parameter, in particular a value of a phase or a value of an amplitude of an S parameter for the purpose of testing a material parameter of the printed circuit board substrate. Furthermore, the functional region and the measurement structure or the waveguide are not connected, i.e. none of the elements of the functional region is connected to the measurement conductor. The advantages of such a circuit board are thus obtained analogously from the above-mentioned advantages for the test method. The above-mentioned developments and their explanations also apply accordingly.An exemplary embodiment of the invention is explained in more detail below with reference to a drawing. Shown therein are: FIG. 1 shows a schematic illustration of a section of a circuit board having a test region and a functional region, FIG. 2 shows the test region of the circuit board according to FIG. 1 with a measurement conductor, FIG. 3 shows an amplitude measurement for testing the circuit board according to FIG. 1, and FIG. 4 shows a phase measurement for testing the circuit board according to FIG. 1.A board 2 partially made of a board substrate 4 to be inspected is shown in FIG. 1. In the exemplary embodiment shown here, the circuit board 2 is designed as a multilayer circuit board and comprises a plurality of substrate layers, in particular an upper substrate layer 6 and a lower substrate layer 8. The circuit board 2 furthermore comprises a plurality of layers, in particular a first layer 10 and a second layer 12. The layers 10, 12 are in particular not embodied over the full surface and continuously, but rather have conductor structures not shown in more detail here, for realizing a suitable circuit layout.The upper substrate layer 6 is in particular enclosed by the first and the second layer 10, 12. The second layer 12 is arranged between the two substrate layers 6, 8. A third layer can be arranged on the underside 14 of the second substrate layer 8. Furthermore, the circuit board 2 can generally also have further substrate layers and layers.The upper substrate layer 6 is made of a board substrate 4 to be tested in the board 2 shown in FIG. 1. In particular, the board 2 is designed for use in a radar system and the board substrate 4 is an RF substrate, while the lower substrate layer 8 is made of another conventional board substrate. Since the circuit board 2 is provided for a specific application, the circuit board 2 has a functional region 16, in which components 18 for producing the corresponding functionality with respect to the application are arranged. Such components 18 are, for example, electronic components and lines connecting these. In the case of a radar system, the functional region 16 has, for example, suitable RF transmitters and receivers, antennas or other radio-frequency components.The circuit board substrate 4 is characterized by at least one material parameter. In the exemplary embodiment described here, in particular by means of two material parameters, namely the dielectric parameter and the loss angle. These two material parameters are of particular importance for a radar application. A suitable test method checks whether the material parameters of the circuit board 2 meet the requirements of the application and, in particular, whether the values of the material parameters are within a predetermined tolerance range.For carrying out the test method, the circuit board 2 comprises a test region 20 shown in more detail in FIG. 2, which is arranged in a corner of the circuit board 2 in FIG. 1 ; a dashed line marks the boundary with the functional region 16. The upper layer 10 is largely made of conductive material in the test region 20 and only has recesses in the corners of the test region 20. A measuring conductor 22 is arranged in the test region 20, which in turn is used for testing the material parameters. For this purpose, the measuring conductor 22 has a measuring input 24 and a measuring output 26, which is connected in each case to a measuring connection 27. This measuring connection 27 is here each designed as a conductor surface, for example, and is formed by recesses in the conductive material in the corners of the upper layer 10. The measuring connections 27 serve here in particular for placing or placing a test tip, not shown here. This can then in turn be connected to a measuring device, for example a network analyzer, by means of a suitable high-frequency cable. In general, other suitable embodiments of the measuring terminal 27 are also conceivable, for example as coaxial plug connectors in SMA embodiments.In the embodiment shown in FIGS. 1 and 2, the measuring conductor 22 is designed as a waveguide 28, more precisely as a substrate integrated waveguide (SIW). This includes a waveguide space 30, which is filled in particular with the circuit board substrate 4 to be tested. The waveguide space 30 is surrounded by a number of horizontal waveguide walls 32 and by vertical waveguide walls 34, which are manufactured in particular from a conductive material, preferably from the same material as the layers 10, 12.To form the lateral waveguide walls 34, the two layers 10, 12 are electrically conductively connected by means of a plurality of plated-through holes 36 in the exemplary embodiment shown here. The through-contacts 36 are embodied here as so-called vias, i.e. bores metallized on the inside. Alternatively, a configuration of the lateral waveguide walls 34 as inwardly metallized trenches, so-called grooves, is conceivable.Furthermore, the waveguide 28 is preferably straight, i.e. has no bends or deflections. The vias 36 are then arranged in particular in a row and at a specific distance 38 from one another. The number of vias 36 of a row and the distance 38 also define a length 40 of the waveguide. Furthermore, the two rows are spaced apart from one another and in this way define a width 42 of the waveguide 28.The through-contacts 36 also each have a specific diameter 46. The distance 38 and the diameter 46 are expediently selected as a function of frequency; that is to say selected with regard to the frequency or the frequency range which is used or examined in the test method. The length 40, the width 42 and the height 44 of the waveguide 28 are also selected taking into account the frequency or frequencies used in the test method and in particular also in the intended application.To carry out the test method, a measurement signal having a predefined measurement frequency is fed or coupled into the measurement conductor 22 by means of the measurement input 24. The measurement signal propagates substantially along the measurement conductor 22, the propagation being influenced accordingly by the board substrate 4. In particular in the exemplary embodiment shown here, the measurement signal is essentially guided through the waveguide space 30 filled with the circuit board substrate 4 to be tested. In this case, the measurement signal experiences in particular a phase shift and an attenuation, the extent of which is dependent in each case on the dielectric parameter and / or the loss angle of the printed circuit board substrate 4.For checking the material parameters, therefore, the phase shift and / or the attenuation of the input signal is expediently determined. For this purpose, in the exemplary embodiment shown here, the so-called S 2,1- parameter of the measurement conductor 22 is measured. This has an amplitude and a phase that quantify the attenuation or phase shift. The S 2,1- parameter is determined in particular by comparing the so-called output signal present at the measurement output 26 with the input signal. This is carried out, for example, by means of a network analyzer, not shown here, which is connected to the measurement input 24 and the measurement output 26. In particular, the amplitude of the S 2,1- parameter then corresponds to the amplitude ratio of output signal to input signal and the phase of the S 2,1- parameter corresponds to the phase difference between output signal and input signal.To explain the test method further, FIG. 3 shows an amplitude measurement 48 and FIG. 4 shows a phase measurement 50 of the S 2,1- parameter. The amplitude A, also denoted |S 2,1| and the phase P, also denoted φ, are in each case represented as functions of the frequency F, in particular of the measurement frequency of the input signal. For this purpose, the frequency F is shown in each case rising from left to right on the abscissa, and the values 52 of the amplitude A and the values 54 of the phase P rising from bottom to top on the ordinate. The measuring conductor 22 has a so-called working frequency range 56 and a so-called cut-off frequency range 58, which are identified in FIGS. 3 and 4 by vertical dashed lines. In this case, the operating frequency range 56 denotes, in particular, an operating range provided on the basis of the selected dimensions of the measurement conductor 22. In contrast, the cut-off frequency range 58 corresponds to a frequency range on which the functionality of the measurement conductor 22 is reduced in comparison to the working frequency range 56.In order to test one of the material parameters, at least one value 52, 54 of the phase P or of the amplitude A is now measured and compared with a respective reference range 60, 60', 60". In the exemplary embodiment, a reference range 60 is assigned to the value 52 of the amplitude A in the working frequency range 56, a reference range 60' in the cut-off frequency range 58, and a reference range 60" to the value 54 of the phase P. The reference range 60, 60', 60" generally indicates a value range which is assigned in particular to the tolerance range of the material parameter. In this case, the reference range 60, 60', 60" can be specified either only at a specific frequency F, i.e. one-dimensionally, or for a plurality of frequencies F, i.e. two-dimensionally. In the exemplary embodiment described here, the reference range 60, 60', 60" is two-dimensional, i.e. comprises, in the graphs illustrated in FIGS. 3 and 4, an area which is restricted in particular in the horizontal direction by a frequency interval 62 and in the vertical direction by suitable limit values 64, 66.In order to determine the reference regions 60, 60', 60", a measurement similar to the test method is preferably carried out on a reference substrate, not shown here. This is selected in such a way that the material parameter has a preferred, in particular ideal value. With respect to this material parameter, at least one value 52, 54, in particular reference value of the amplitude A or of the phase P, is then measured. The reference range 60, 60', 60" is then determined as a value range around the reference value. For example, a reference value for checking the loss angle is determined by means of an amplitude measurement 48 of the S 2,1- parameter, and the reference range 60 is defined in such a way that it comprises all values which amount to at least 95% of the reference value 60.Referring to FIG. 3, the inspection of the loss angle of the board substrate 4 will be explained in detail first. The loss angle, which is also referred to as Df, is characterized in particular in that it is substantially proportional to the amplitude A of the S 2,1- parameter for frequencies F from the working frequency range 56 of the measurement conductor 22. Therefore, for checking the loss angle, an amplitude measurement 48 is advantageously carried out on a frequency interval 62, which here corresponds to the operating frequency range 56. In other words, values 52 of the amplitude A at different frequencies F are measured from the working frequency range 56, as shown in FIG. 3.Also shown in FIG. 3 is the reference range 60 with which the amplitude values 52 are compared. The reference range 60 here comprises a lower limit value 64 and all values which are greater than this limit value 64 An upper limit value 66 results in this case in particular automatically in that the amplitude A of the S 2,1- parameter can assume at most the value 0 dB, that is to say there is no attenuation of the measurement signal by the measurement conductor 22. Therefore, only one limit value, in particular a lower limit value 64, is necessary for defining the reference range 60. In particular, the lower limit value 64 applies to the S 2,1- parameter. The lower limit value 64 is also assigned a lower limit value, denoted Dk ref,unten for the loss angle Df. As is clearly evident from the amplitude measurement 48 shown in FIG. 3, the amplitude values 52 are within the reference range 60, and the checking of the loss angle is thus positively evident. In other words, the result of the test method for the loss angle of the circuit board substrate 4 is positive; the circuit board 2 is therefore suitable for the intended application at least with respect to the loss angle of the examined circuit board substrate 4.FIG. 3 also shows an amplitude measurement 48 in the cut-off frequency range 58 of the measurement conductor 22 for testing the dielectric parameter, which is also referred to as Dk. In the cut-off frequency range 58, this is expediently substantially proportional to the measurement frequency, i.e. frequency F, as a result of which in particular an increasing profile of the values 52 with an increasing measurement frequency results. This measurement is assigned the reference range 60' which, unlike the reference range 60 for the loss angle, is defined by an upper and a lower limit value 64', 66'. The limit value for the dielectric parameter assigned to the upper limit value 66' is also referred to as Dk ref,oben and the limit value for the dielectric parameter assigned to the lower limit value 64' is referred to as Dk ref,unten. When measuring the amplitude A at different frequencies F, a corresponding band results therefrom. As can be seen from FIG. 3, the measured values 52 are in the range defined by the limit values 64' and 66'. The tested printed circuit board 2 thus fulfills the requirements with respect to its dielectric parameter Dk. Thus, according to FIG. 3, Dk ref,unten< Dk< ref,oben. applies to this.Furthermore, the upper and the lower limit value 64', 66' here are each frequency-dependent, and the reference range 60' therefore extends correspondingly obliquely and upwards for ascending frequencies F.The phase measurement 50 illustrated in FIG. 4 illustrates the testing of the circuit board substrate 4 with respect to the dielectric parameter by measuring values 54 of the phase P of the S 2,1- parameter, in particular at different frequencies F. Advantageously, the phase P in the working frequency range 56 of the measurement conductor 22 is substantially proportional to the dielectric parameter. In addition, the phase P is frequency-dependent. Accordingly, for the associated reference range 60'', an oblique profile results similar to the case of the amplitude A measured in the cut-off frequency range 58, but decreasing for increasing frequencies F. Also in the case of the phase measurement 50 shown in FIG. 4, the reference range 60'' has an upper and a lower limit value 64'', 66''. The limit value for the dielectric parameter assigned to the upper limit value 66'' is also referred to as Dk ref2,oben and the limit value for the dielectric parameter assigned to the lower limit value 64'' is referred to as Dk ref2,unten. If the value 54 of the phase P lies within the reference range 60", the test result is positive; the dielectric parameter is then consequently within the predefined tolerance range and in particular meets the condition Dk ref2,unten< Dk<K ref2,oben. on the working frequency range 56.In general, in the test method described by means of FIGS. 3 and 4, the respective value of the material parameter is not measured directly in associated units, but rather the test method uses a respective dependence of the S parameter on the material parameter. In particular, the frequency range 62 used during the test is important here, since the behavior of the measuring conductor 22 is different in the working frequency range 56 and in the cut-off frequency range 58. This fact is also used in particular in the test method in order to test different material parameters.

Claims

Test method for a printed circuit board substrate (4), in particular an RF substrate for a printed circuit board (2) of a radar system in a vehicle, wherein the printed circuit board substrate (4) has at least one material parameter to be tested, wherein in the test method the material parameter is selected from a number of parameters comprising a loss angle and a dielectric parameter, wherein - a printed circuit board (2) is provided with a test region (20) manufactured from the printed circuit board substrate (4), in which test region a measurement structure (22) is formed, which has a measurement input (24) and a measurement output (26), - the measurement structure (22) is provided as a waveguide (28), which has a waveguide space (30) which is arranged in the printed circuit board substrate (4) and is filled with the latter, - an input signal having a predefined measurement frequency is applied to the measurement input (24), at least one value (52, 54) of a test parameter of the measurement structure (22) is determined, at least one reference range (60, 60', 60") is predefined for the value (52, 54), and a test is carried out to determine whether the value (52, 54) lies within the reference range (60, 60', 60").Test method according to the preceding claim, characterized in that the test parameter is an S parameter and the S parameter comprises a phase (P) and an amplitude (A) and in that at least one value (52, 54) of the phase (P) or of the amplitude (A) is determined.Test method according to one of the preceding claims, characterized in that the circuit board (2) comprises a functional region (18) in addition to the test region (20).Test method according to one of the preceding claims, characterized in that the measurement input (24) and the measurement output (26) are each connected to a measurement connection (27) arranged in the test region (20).Test method according to one of the preceding claims, characterized in that an output signal is measured at the measurement output (26), and the output signal is compared with the input signal in order to determine the test parameter.Test method according to one of the preceding claims, characterized in that it is used as an online test method during a production process, in particular during production of a radar system for a vehicle.Test method according to one of the preceding claims, characterized in that it is used for quality control of a number of printed circuit boards (2), in particular printed circuit boards (2) for radar systems for a vehicle.Test method according to one of the preceding claims, characterized in that the measurement frequency of the input signal is selected from a frequency range from 20 GHz to 120 GHz.Test method according to one of the preceding claims and according to Claim 2, characterized in that the measurement structure (22) has an operating frequency range (56), and the measurement frequency of the input signal is selected from the operating frequency range (56) and a value (52) of the amplitude (A) is measured in order to test the loss angle.Test method according to one of the preceding claims and according to Claim 2, characterized in that the measurement structure (22) has an operating frequency range (56), and the measurement frequency of the input signal is selected from the operating frequency range (56) and a value (54) of the phase (P) is measured in order to test the dielectric parameter.Test method according to one of the preceding claims and according to Claim 2, characterized in that the measurement structure (22) has a cut-off frequency range (58), and the measurement frequency of the input signal is selected from the cut-off frequency range (58) and a value (52) of the amplitude (A) is measured in order to test the dielectric parameter.Test method according to one of the preceding claims, characterized in that the measurement frequency of the input signal is changed, for measuring a plurality of values at different measurement frequencies.Test method according to one of the preceding claims, characterized in that the reference range (60, 60', 60") is selected as a function of the tested material parameter, wherein the reference range (60, 60', 60") is a value range with an upper and a lower limit value (66', 64', 66", 64") when testing the dielectric parameter and is a value range above a limit value (64) when testing the loss angle.Circuit board (2) which is suitable for use in a test method according to one of the preceding claims, wherein the circuit board (2) is produced from a circuit board substrate (4) and comprises a functional region (18) and a test region (20), wherein the latter is produced from the circuit board substrate (4) and comprises a measurement structure (22), which is designed as a waveguide (28), having a measurement input (24) and a measurement output (26) for measuring a value (52, 54) of a test parameter, and wherein the functional region (18) is not connected to the measurement structure (22) or the waveguide (28).

Images