JP2012220425A - Noise source estimating device and noise source estimating method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To solve such a problem that the detection omission of a noise source having a small absolute value is generated in near electromagnetic field distribution information, when the noise source is estimated from design information and the near electromagnetic field distribution information.SOLUTION: A noise source estimating device is used, which is provided with an analyzing portion 151 which analyzes an electromagnetic wave of an electronic apparatus on the basis of design information between a design information input portion 102 and an estimating portion 156; calculates current distributions in a current calculating portion 152 on the basis of the analysis result of the analyzing portion 151 and electromagnetic wave information from an electromagnetic wave information input portion 103; calculates a difference between both in a current difference calculating portion 153; and estimates a part having a large difference as a noise source in an estimating portion 156.

Description

  The present invention relates to a noise source estimation device and a noise source estimation method for estimating a noise source of an electronic device incorporating a circuit board.

  Conventionally, a noise source estimation device for estimating a noise source of a circuit board has a configuration characterized by including a design information input unit, an electromagnetic wave information input unit, and an estimation unit.

  Here, circuit board design information is input to the design information input unit. Also, electromagnetic wave information is input as measurement information of the circuit board to the electromagnetic wave information input unit. Furthermore, the estimation unit estimates which part of the circuit board is a noise source based on the design information and the electromagnetic wave information (see, for example, Patent Document 1).

JP 2003-302433 A

  However, in the conventional noise source estimation device and noise source estimation method, when the noise source is estimated from the design information and the nearby electromagnetic field distribution information, a noise source having a small absolute value in the nearby electromagnetic field distribution information is not detected. It had the problem of occurring.

  Here, noise having a small absolute value may greatly affect electromagnetic interference (EMI: Electro Magnetic Interference). In addition, since this electromagnetic interference affects other devices, it is regulated in each country, and products that do not satisfy this regulation cannot be shipped. For example, in Japan, it is regulated by VCCI (Voluntary Control Council for Interference by Data Processing Equipment and Electronic Office Machines). Accordingly, detection of a noise source having a small absolute value may increase the man-hours for EMI countermeasures and noise countermeasures, which may lead to a fatal problem for corporate activities such as product shipment delays.

  The present invention solves the above-described conventional problems, and an object thereof is to provide a noise source estimation device and a noise source estimation method that are free from detection failure of the noise source.

  In order to achieve the above object, a noise source estimation apparatus according to the present invention inputs a design information input unit for inputting design information of an electronic device including at least a circuit board and measurement information of electromagnetic waves radiated or conducted from the electronic device. In a noise source estimation device comprising an electromagnetic wave information input unit that performs an estimation unit that estimates which part of an electronic device is a noise source based on design information and measurement information, a design information input unit and an estimation unit An analysis unit that performs electromagnetic wave analysis of electronic equipment based on design information is provided between them, and the current distribution is calculated in the current calculation unit based on the analysis result of this analysis unit and the electromagnetic wave information from the electromagnetic wave information input unit. The difference between the currents calculated by the current calculation unit is calculated by the current difference calculation unit, and the estimation unit is configured to estimate the large difference as a noise source. To have.

  According to the noise source estimation apparatus of the present invention, the current distribution is calculated from the design information and the actual measurement information, and the portion where the difference is large is estimated as the noise source. It is possible to obtain an effect of extracting a noise source that cannot be extracted only by information and reducing detection omissions. Here, noise having a small absolute value has a great influence on electromagnetic interference. Since this electromagnetic interference affects other devices, it is regulated in each country, and products that do not meet this regulation cannot be shipped. Therefore, detecting the noise source reduces detection omissions at locations where EMI countermeasures should be implemented. That is, EMI countermeasures and noise countermeasures for electronic devices can be efficiently performed.

Schematic block diagram of noise source estimation apparatus to which the present invention is applied Overall configuration diagram of a system including a noise source estimation apparatus in an embodiment of the present invention The flowchart of the noise source estimation apparatus in the embodiment of the present invention Explanatory drawing of the internal structural example of the near electromagnetic field table in embodiment of this invention Explanatory drawing which shows the intensity | strength of the near electromagnetic field distribution measured using the noise source estimation apparatus in embodiment of this invention Explanatory drawing which looked at the noise source in embodiment of this invention from the Z-axis direction Explanatory drawing which shows the intensity | strength of the near electromagnetic field distribution analyzed from the design information in embodiment of this invention Schematic diagram of a simple test substrate in an embodiment of the present invention Explanatory drawing which shows the intensity | strength of the near magnetic field distribution by actual measurement of the board | substrate for simple tests in embodiment of this invention Explanatory drawing which shows the intensity | strength of near magnetic field distribution by analysis of the design information of the board | substrate for simple tests in embodiment of this invention Explanatory drawing which calculates each electric current on to-be-measured object 1 from the intensity | strength of the near electromagnetic field distribution by actual measurement and the intensity | strength of the near electromagnetic field distribution by analysis of design information in embodiment of this invention, and shows the difference Explanatory drawing of the internal structural example of the board | substrate structure table in embodiment of this invention Explanatory drawing of the internal structural example of the interlayer table in embodiment of this invention Explanatory drawing of the internal structural example of the components table in embodiment of this invention Explanatory drawing of the example of an internal structure of the terminal table in embodiment of this invention Explanatory drawing of the internal structural example of the arrangement | positioning table in embodiment of this invention Explanatory drawing of the internal structure example of the net table in the embodiment of the present invention Explanatory drawing of the internal structural example of the wiring table in embodiment of this invention

(Embodiment)
Hereinafter, embodiments of the present invention will be described with reference to the drawings.

  FIG. 1 is a block diagram of a noise source estimation apparatus to which the present invention is applied. As this configuration, information input to the design information input unit 102 and the electromagnetic wave information input unit 103 is input to the storage unit 190. Information input to the command input unit 101 is output to the separation unit 105 via the command analysis unit 104, and the separation unit 105, the acquisition unit 106, the analysis unit 151, the current calculation unit 152, the current difference calculation unit 153, and the frequency error calculation. The unit 154 and the search unit 155 sequentially output and input information in series, and these also input and output information to and from the storage unit 190, respectively. Information output from the acquisition unit 106, analysis unit 151, current difference calculation unit 153, and search unit 155 is input to the estimation unit 156, and information output from the estimation unit 156 is input to the output unit 191. Information output from the storage unit 190 is also input to the output unit 191.

  FIG. 2 shows the entire system including the noise source estimation device 4. In this configuration, the near electromagnetic field of the DUT 1 is measured using the near electromagnetic field measuring device 2 and the noise measuring device 3, and the noise measuring device 3 is connected to the noise source estimating device 4. Here, the device under test 1 is an electronic device incorporating at least a circuit board, the near electromagnetic field measuring device 2 is a device that converts the near electromagnetic field of the device under test 1 as an electric signal, and the noise measuring device 3 is the frequency of the electric signal. A device that displays the characteristics. For example, there is a spectrum analyzer as one of the noise measuring devices 3.

  The configuration of the noise source estimation device 4 will be described below with reference to FIGS. 3 to 18 together with the operation, focusing on FIG. 3 showing the operation of the noise source estimation device according to the embodiment of the present invention.

  First, as shown in step S <b> 11 of FIG. 3, design information of the DUT 1 is input from the design information input unit 102 and stored in the storage unit 190. This design information is information created by, for example, a CAD (Computer Aided Design) apparatus, and is shown in the substrate configuration table 10 shown in FIG. 12, the interlayer table 11 shown in FIG. 13, the component table 12 shown in FIG. 14, and the FIG. The data is written in the terminal table 13, the arrangement table 14 shown in FIG. 16, the net table 15 shown in FIG. 17, and the wiring table 16 shown in FIG. The detailed contents of these tables will be described in detail in the last section of the present embodiment.

  Now, when the design information of the device under test 1 is stored in the storage unit 190 in this way, as shown in step S12 of FIG. 3, the electromagnetic wave of the near electromagnetic field of the device under test 1 measured by the noise measuring device 3 is obtained. Information is input from the electromagnetic wave information input unit 103 and stored in the storage unit 190. This electromagnetic wave information is written in the near electromagnetic field table 17 shown in FIG.

  As shown in FIG. 4, the near electromagnetic field table 17 has an electric field when measured in a state where the DUT 1 does not operate for each of “frequency”, “X coordinate”, “Y coordinate”, and “Z coordinate”. A “noise floor” field indicating strength (or magnetic field strength) and a “noise amount” field indicating electric field strength (or magnetic field strength) when the DUT 1 is operated are provided. Since the “current value (analysis)”, “current value (actual measurement)”, “electromagnetic field noise determination”, and “current difference determination” fields are fields related to later operations, they will be described later. Here, for example, in order from the left column with respect to the second row in FIG. 4, the frequency is 30 [MHz] with respect to the magnetic field strength, and the near electromagnetic field measurement when measuring and analyzing the noise amount and noise floor of the DUT 1. The X, Y, and Z coordinates in the apparatus 2 are 0.0 [mm], 0.0 [mm], and 5.0 [mm], the noise floor at the coordinates is 10.0 [dBuV], and the noise amount is 10 0 [dBuV].

  FIG. 5 shows the noise amount distribution in the near electromagnetic field table 17 of FIG. In FIG. 5, the horizontal direction and the vertical direction represent the X axis and the Y axis, respectively, and the Z coordinate at this time is assumed to be constant at an arbitrary value. The pattern density of each segment 5 represents the amount of noise or noise floor measured in that region, and the higher the pattern density, that is, the darker the color, the higher the noise amount or noise floor value. Here, the segment 53 has the largest amount of noise, and the amount of noise decreases in the order of the segment 52, the segment 51, and the segment 5. In FIG. 5, the value is high because the upper side is dark. The contents shown in FIG. 5 can be output to an output unit 191 such as a display as necessary.

  Now, when the electromagnetic wave information measured by the noise measuring apparatus 3 is stored in the storage unit 190 in this way, the user inputs a command using the command input unit 101 such as a keyboard or a mouse. There are types of commands such as design information read command, electromagnetic wave information read command, noise analysis command, electromagnetic field analysis command, current calculation command, difference calculation command, noise source estimation command, etc., but here the noise source estimation command is input Assuming that When a command is input in this way, the command analysis unit 104 analyzes the type of the command, and when it recognizes that a noise source estimation command has been input, calls the separation unit 105.

  As a result, in step S13 of FIG. 3, the separation unit 105 refers to the near electromagnetic field table 17 stored in the storage unit 190, and the value obtained by subtracting the noise floor from the noise amount exceeds a predetermined value such as 3 dB or more. The frequency at the time shown is specified, and a noise is separated by setting a flag in the determination field of the nearby electromagnetic field table 17. In FIG. 4, the “◯” mark in the “electromagnetic field noise determination” column indicates that the flag is set, while the “x” mark indicates that the flag is not set.

  Next, in step S <b> 14 of FIG. 3, the acquisition unit 106 acquires frequency information used in the DUT 1 from the net table 15 stored in the storage unit 190. 3, the member information of the DUT 1 from the board configuration table 10, the interlayer table 11, the component table 12, the terminal table 13, the arrangement table 14, and the wiring table 16 stored in the storage unit 190. To get. Further, in step S <b> 16 of FIG. 3, electromagnetic wave information of the DUT 1 is acquired from the near electromagnetic field table 17 stored in the storage unit 190.

  Thereby, in step S17 of FIG. 3, the analysis unit 151 analyzes the nearby electromagnetic field for each member of the DUT 1 based on the member information acquired by the acquisition unit 106 and the like. For the analysis method of the near electromagnetic field, a simple analysis method may be used, or a detailed analysis method such as a moment method or an FDTD (Finite Difference Time Domain) method may be used. Since these analysis techniques are known techniques, a detailed description thereof is omitted here.

  Here, in step S <b> 18 of FIG. 3, the current calculation unit 152 measures the near electromagnetic field information of the device under test 1 analyzed by the analysis unit 151 and the measured object of the near electromagnetic field table 17 stored in the storage unit 190. The respective currents on the object to be measured 1 are calculated with respect to the noise amount of the near electromagnetic field information of the object 1 minus the noise floor. For this calculation, an analysis technique similar to that of the analysis unit 151 is used.

  Thereby, in step S19 of FIG. 3, the current difference calculation unit 153 measures the current distribution calculated from the member information of the device under test 1 calculated by the current calculation unit 152 and the device measured by the noise measurement device. The difference of the current distribution calculated from the result of 1 is calculated. At this time, in the current distribution calculated from the member information and the measurement information, the “X coordinate”, “Y coordinate”, “Z coordinate”, and “frequency” respectively calculate the difference between the current values at the same coordinates. The difference calculation is performed for all locations. Then, in a certain coordinate (for example, X coordinate = X1, Y coordinate = Y1, Z coordinate = Z1, frequency = f1), when the difference value indicates a predetermined value or more such as 3 dB or more, the frequency is specified, A flag is set in the determination field of the near electromagnetic field table 17 (“○” mark shown in the “Current difference determination” column in FIG. 4 indicates that the flag is set, while “×” mark indicates that the flag is not set. Represents).

  Next, in step S20 of FIG. 3, the frequency error calculation unit 154 identifies a record flagged in the determination field of the near electromagnetic field table 17, and for each frequency written in the “frequency” field of the record. The frequency error described later is calculated. As described above, the frequency error calculation unit 154 does not directly input the frequency error, but the user may directly input the frequency error, or a preset constant may be used as the frequency error. Good.

  Next, in step S21 of FIG. 3, the search unit 155 refers to the near electromagnetic field table 17, and for each frequency written in the “frequency” field of the record flagged in the determination column, the harmonics thereof. It is checked whether a noise component exists in the component, that is, whether a flag is set in the determination column of the frequency corresponding to the harmonic.

  For example, when a flag is set in the determination column of a record in which information indicating 30 MHz is written in the frequency field (hereinafter, simply referred to as “frequency“ 30 MHz ””), the frequencies corresponding to the harmonics of 30 MHz are 60 MHz and 90 MHz. Therefore, the harmonic components are searched until a predetermined frequency such as 1 GHz, and a flag is set in the determination column of the frequencies corresponding to the harmonics “60 MHz”, “90 MHz”, “120 MHz”, “990 MHz”. Find out if you have.

  Thus, when searching for harmonic components, it is preferable to consider a frequency error of a predetermined width. For example, when the frequency corresponding to the harmonic is 60 MHz, 0.6 MHz is set as the frequency error by calculating 60 MHz × 0.01. Then, in consideration of the frequency error of 0.6 MHz, it is checked whether or not a flag is set in the determination column of frequencies “59.4 MHz” to “60.6 MHz”.

  Thus, the coefficient used when calculating the frequency error is not limited to 0.01. That is, you may make it change the said coefficient for every frequency (60 MHz, 90 MHz, 120 MHz, ...) which corresponds to a harmonic.

  Accordingly, in step S22 of FIG. 3, the estimation unit 156 analyzes the noise component searched by the search unit 155, the frequency information acquired by the acquisition unit 106, the member information, the electromagnetic wave information, and the analysis unit 151. As a result, based on the result calculated by the current difference calculation unit 153, it is estimated which part of the DUT 1 is the noise source.

  That is, when it is found in step S21 in FIG. 3 that a noise component is present, the frequency of the noise component can be estimated as the frequency of the noise source. For example, as shown in FIG. 6, when the near electromagnetic field of an electronic device in which a noise source is arranged is measured, it is assumed that the intensity of the electromagnetic field distribution at a frequency of 30 MHz is as shown in FIG. Here, in FIG. 6, “×” is attached to the segment 54 corresponding to the noise source. Further, it is assumed that the strength of the electromagnetic field distribution obtained by analyzing the near electromagnetic field from the design information of the electronic device is as shown in FIG. In this case, in the conventional method, only the upper side of FIG. 5 is extracted as a noise source, and the lower center portion having a small absolute value is not extracted as a noise source. However, when this method is used, the measurement of the device under test 1 is compared with the data analyzed from the design information of the device under test 1, that is, the lower center portion where the difference between FIG. 5 and FIG. It is possible to extract the segment 52).

  Here, an example of the noise source having a small absolute value will be described. A substrate for simple testing as shown in FIG. In this substrate, one signal line 6 is arranged on the first layer, a ground plane 7 is arranged on the second layer, and an insulating layer is sandwiched between the first layer and the second layer. However, it is assumed that the ground plane 7 is divided into two parts on the left and right sides at the central part 8 of the substrate. That is, the ground plane 7 does not exist immediately below the first-layer signal line 6 in the central portion 8 of the substrate. FIG. 9 shows the measurement result of the near magnetic field when a current is passed through the signal line of the substrate. In FIG. 9, details of each unit are not described here, but the vertical direction represents the amount of noise, and the value increases as it goes upward. Therefore, the amount of noise is relatively large in the portion corresponding to the central portion 8 of the substrate. In this case, if the absolute value of the noise amount in the portion corresponding to the central portion 8 of the substrate is small, this portion is not extracted as a noise source in the conventional method. FIG. 10 shows the intensity of the magnetic field distribution obtained by analyzing the near magnetic field from the design information of the substrate. In FIG. 10, the portion corresponding to the substrate central portion 8 shows the same amount of noise as the other portions, that is, the portion where the ground plane 7 is located immediately below the signal line 6. Here, by using this method, it is possible to extract a portion corresponding to the substrate central portion 8 where the difference between FIG. 9 and FIG. 10 occurs.

  If the noise source can be estimated, the estimation result is output to the output unit 191 such as a display. This output form is such that the user can visually determine which part of the DUT 1 is a noise source. For example, as shown in FIG. 6, the estimated noise source itself is output. In this case, when the user inputs a display switching command, FIG. 5 showing the intensity of the measured electromagnetic field distribution, FIG. 7 showing the intensity of the electromagnetic field distribution analyzed from the design information, or the result of both measurements It is also possible to output the currents on the object 1 as shown in FIG. Although a certain frequency (for example, 30 MHz) has been described here, an arbitrary electromagnetic field distribution can be displayed by freely setting a frequency range displayed by the user.

  If a plurality of noise sources are found, all of them may be output as noise source candidates. Alternatively, the signals flowing through the members may be sorted in descending order of the power supply voltage, and only the higher ranks may be output as noise source candidates. Further, the current value calculated by the current calculation unit 152 may be sorted in descending order, and only the higher ranks may be output as noise source candidates.

  As described above, according to the present invention, it is possible to estimate which part of the device under test 1 is a noise source based on the design information of the device under test 1, so that EMI countermeasures can be taken promptly. Can do.

  In the above description, harmonic components are searched until a predetermined frequency is obtained such as 60 MHz, 90 MHz, 120 MHz,..., 990 MHz, but the present invention is not limited to this.

  Further, the amount of noise emission depends on the rise time or fall time of a signal flowing through the member in addition to the fundamental frequency. That is, even with a signal having a low fundamental frequency, if the rise time or fall time is short, strong electromagnetic waves are radiated to a high frequency. Therefore, only the member whose rise time is a predetermined value (for example, 5 ns) or the member whose fall time is the predetermined value (for example, 5 ns) or less may be set as a noise source candidate. That is, the estimation unit 156 selects only a specific member (for example, a member having the rise time of 5 ns or less) as a candidate for a noise source.

  Furthermore, when a plurality of noise components are listed as noise source candidates (for example, when noise components of 10 MHz, 15 MHz, 20 MHz, 30 MHz, 40 MHz, and 45 MHz are measured), the frequency of these noise components is the net table 15. There are times when it is not written in the “frequency” field. At this time, the frequencies corresponding to the lower harmonics (for example, the frequencies 20 MHz, 30 MHz, and 40 MHz corresponding to the harmonics of 10 MHz) may be excluded from the noise source candidates. Alternatively, the priority order as a noise source candidate may be lowered. Further, when there are a plurality of noise source candidates, the priority order of the noise source candidates having a high frequency may be increased.

  Here, the contents of each table from 10 to 16 will be described in detail. These tables are related to the circuit board. However, even in the case of an electronic device incorporating a circuit board, the present invention can be applied by extending these tables to information about the entire electronic device.

  As shown in FIG. 12, the board configuration table 10 includes “layer name” fields that constitute the circuit board, and “signal type” fields that flow through these layers. Further, the interlayer table 11 has a “layer thickness” field in the circuit board as shown in FIG. Information written in the tables 10 and 11 is input to the noise source estimation device 4 together with information on the board outline. Here, it is assumed that a printed wiring board having four layers is designed, and the first layer and the fourth layer are for signal lines, the second layer is for 0V ground, and the third layer is for 5V power supply. The case where it is for is illustrated.

  As shown in FIG. 14, the component table 12 includes “component number”, “component name”, “number of terminals”, “shape”, “length”, and “width” for each component included in the circuit board. Each field is provided. “SOP” exemplified in the “shape” field is an abbreviation for Small Outline Package.

  As shown in FIG. 15, the terminal table 13 includes fields of “terminal number”, “attribute”, “rise time”, and “fall time” for each component included in the circuit board. “IN” illustrated in the “attribute” field indicates that the terminal is an input terminal, while “OUT” indicates that the terminal is an output terminal.

  As shown in FIG. 16, the arrangement table 14 includes a “coordinate” field, an “arrangement angle” field, and an “arrangement plane” field that indicate the arrangement position for each component arranged on the circuit board. A general method such as manual placement may be used as a method for placing components on the circuit board at the design stage.

  As shown in FIG. 17, the net table 15 includes a “net name” field indicating the name of each signal line (hereinafter referred to as “net”) included in the circuit board, and names of terminals connected to the net. "Connection terminal name" field, a "frequency" field of a signal flowing through the net, and a "potential" field. In this example, the net named “net2” is a 33 MHz signal line, the net named “net100” is a 0V ground, and the net named “net200” is a 5V power supply. Yes.

  As shown in FIG. 18, the wiring table 16 includes a “type” field for each net and a “configuration coordinate” field indicating the configuration coordinates of the foil or plane of the net. The plane here is a ground plane or a power plane. The composition coordinates of the net foil or plane are the coordinates of the points constituting the net foil or plane (for example, if the net foil can be represented by a line segment, both end points of the line segment). is there.

  The noise source estimation device and the noise source estimation method of the present invention can be effectively used at least for EMI countermeasures and noise countermeasures of electronic devices with a built-in circuit board.

DESCRIPTION OF SYMBOLS 1 Measured object 2 Near electromagnetic field measuring device 3 Noise measuring device 4 Noise source estimating device 5 Segment 6 Signal line 7 Ground plane 8 Substrate center part 51 Segment 52 Segment 53 Segment 54 Segment 101 Command input part 102 Design information input part 103 Electromagnetic wave Information input unit 104 Command analysis unit 105 Separation unit 106 Acquisition unit 151 Analysis unit 152 Current calculation unit 153 Current difference calculation unit 154 Frequency error calculation unit 155 Search unit 156 Estimation unit 190 Storage unit 191 Output unit

Claims (2)

At least a design information input unit for inputting design information of an electronic device incorporating a circuit board, an electromagnetic wave information input unit for inputting measurement information of an electromagnetic wave radiated or conducted from the electronic device, and the design information and the electromagnetic wave information. And an estimation unit for estimating which part of the electronic device is a noise source based on the design information between the design information input unit and the estimation unit. An analysis unit that performs analysis is provided, and the current calculation unit calculates the current distribution of the result of analysis by the analysis unit and the electromagnetic wave information from the electromagnetic wave information input unit, and calculates the difference between the currents calculated by the current calculation unit. A noise source estimation device characterized in that the current difference calculation unit calculates a current difference and the estimation unit estimates a portion having a large difference as a noise source. Design information input processing for inputting design information of at least an electronic device incorporating a circuit board, electromagnetic wave information input processing for inputting measurement information of electromagnetic waves radiated or conducted from the electronic device, the design information and electromagnetic wave information, And a noise source estimation method for estimating which part of the electronic device is a noise source based on the electromagnetic wave analysis process for performing electromagnetic wave analysis based on the design information. Noise obtained by calculating each current distribution using the obtained value and electromagnetic wave information input from the electromagnetic wave information input process, calculating a difference between the two, and estimating a portion having the large difference as a noise source Source estimation method.

Images