This application is related to and claims the benefit of priority from Japanese Patent Application No. 2007-17235, filed on Jan. 29, 2007, the entire contents of which are incorporated herein by reference.
1. Field
The present technique relates to prediction of attributive data of a product based on design information of the product and a simulation.
2. Description of the Related Art
A simulation technique using the finite-element method, for example, is utilized to predict attributive data of a product based on design information. A designer can obtain attributive data of a product by the simulation technique without actually producing the product.
The challenges dealt with in the simulation technique are to improve prediction accuracy of the simulation and to increase speed of the simulation.
Although typical simulation techniques may be utilized to predict attributive data, there is a need for an efficient and accurate attribute data prediction apparatus and method.
The disclosed apparatus includes a storage unit storing differential information between a simulated value of attributive data of a product and an actually measured value of the attributive data of the product, a prediction unit predicting attributive data of the new product by simulation of the new product using design information of the new product, and a correction unit correcting the simulated attributive data of the new product based on the differential information stored.
The disclosed method includes predicting attributive data of a new product by simulation using design information thereof and correcting the predicted attributive data of the new product based on differential information between a simulated value of attributive data of at least one other product and an actually measured value of the attributive data of said at least one other product.
Additional aspects and/or advantages will be set forth in part in the description which follows and, in part, will be apparent from the description, or may be learned by practice of the invention.
These and/or other aspects and advantages will become apparent and more readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings of which:
Reference will now be made in detail to the embodiments, examples of which are illustrated in the accompanying drawings, wherein like reference numerals refer to the like elements throughout. The embodiments are described below to explain the present invention by referring to the figures.
The circuit board deformation prediction apparatus 1 predicts a warp value of a circuit board undergoing a condition such as heating, which has a laminated structure and includes electronic parts mounted thereto. To execute the prediction, the circuit board deformation prediction apparatus 1 may include a design information database (DB) 10, a material characteristic information database (DB) 11, a warp measurement result database (DB) 12, a numerical analysis result database (DB) 13, a warp correction value database (DB) 14, a database management unit 15. As illustrated in
The design information database 10 manages design information of a circuit board which is a target of the prediction of a warp value. The design information may be prepared using, e.g., CAD and may be registered in the design information database 10.
The material characteristic information database 11 manages characteristic information of materials described in the design information. The characteristic information of the materials includes information of properties including those depending on temperature.
The warp measurement result database 12 manages warp value measurement results of the circuit board which has been subjected to measurement of a warp value. The warp value of the circuit board may be measured through operations of heating the circuit board while changing a temperature condition, and measuring the warp value of the circuit board under the heating at each temperature condition by the moiré interference method, the laser displacement measurement method, etc.
The numerical analysis result database 13 manages numerical analysis results of circuit board warp values, which are obtained by a numerical analysis.
The warp correction value database 14 manages, as a warp correction value, a differential value between a measurement result of the warp value, which is stored in the warp measurement result database 12, and the numerical analysis result of the warp value, which is stored in the numerical analysis result database 13.
The database management unit 15 manages the design information database 10, the material characteristic information database 11, the warp measurement result database 12, the numerical analysis result database 13, and the warp correction value database 14.
The warp measurement result collection unit 16 collects the measurement results of the circuit board warp values and registers the collected measurement results in the warp measurement result database 12.
The circuit board numerical analysis unit 17 performs a numerical analysis of the circuit board and registers the numerical analysis result in the numerical analysis result database 13. The circuit board numerical analysis unit 17 performs the numerical analysis based on design information and material characteristic information. One example of the numerical analysis performed by the circuit board numerical analysis unit 17 is described in a Non-Patent Document 1, i.e., Nobutaka Ito et al., “Studies on warp analysis technique for printed wiring board”, Proceedings of 12-th Symposium on Micro-Bonding and Packaging Techniques in Electronics held under the sponsorship of Micro-Bonding Research Committee, Japan Welding Society, Vol. 12, pp. 461-466, 2006.
The warp correction value registration unit 18 calculates a differential value between the measurement result of the circuit board warp value and the numerical analysis result of the circuit board warp value. Further, the warp correction value registration unit 18 registers the calculated differential value as the warp correction value in the warp correction value database 14.
When there is a request for predicting a warp value of a new circuit board, the numerical analysis result correction unit 19 outputs a corrected numerical analysis result as the numerical analysis result. In other words, the numerical analysis result correction unit 19 corrects a numerical analysis result of the new circuit board by using the warp correction value.
The warp measurement result collection unit 16, the circuit board numerical analysis unit 17, the warp correction value registration unit 18, and the numerical analysis result correction unit 19 can be implemented in the form of computer programs. The computer programs may be recorded on a computer-readable recording medium or transmitted via a network. Further, a unit, such as a CPU, etc., can implement the disclosed process or operation by executing the computer programs.
The numerical analysis result correction unit 19 reads the warp correction value from the warp correction value database 14. Then, the numerical analysis result correction unit 19 corrects the numerical analysis result, which has been analyzed by the circuit board numerical analysis unit 17, by using the warp correction value. The numerical analysis result correction unit 19 can read the warp correction value of the circuit board, which is similar to the structure of the circuit board having been subjected to the numerical analysis, by utilizing a data structure of the warp correction value database 14.
By utilizing the data structure of the warp correction value database 14 shown in
When an electronic device is improved, a structure of a circuit board for the electronic device after the improvement is often similar to the structure of the circuit board for the electronic device before the improvement.
The warp correction value registration unit 18 registers the warp correction value of the circuit board in the warp correction value database 14 in a corresponding relation to the model name of the circuit board.
Based on the data structure of the warp correction value database 14 shown in
The values of the structure items of the circuit board represent a size of the circuit board, a number of layers, a mounting type, a wiring ratio, a number of vias, a kind of material, etc. Examples of the circuit board include a circuit board for mounting an FBGA (Fine Pitch Ball Grid Array) LSI, a circuit board for mounting a CSP (Chip Scale Package) LSI, and a circuit board for mounting a POP (Package on Package) LSI. While
The warp correction value database 14 manages the warp correction value in accordance with the data structure shown in
The following exemplary description is given of process(es) executed by the circuit board deformation prediction apparatus 1.
The existing circuit board is heated while changing a temperature condition. The warp value of the circuit board undergoing the heating at each temperature condition is measured by the moiré interference method, the laser displacement measurement method, etc.
In operation ST10 of the flowchart of
In operation ST11, the circuit board deformation prediction apparatus 1 registers the measurement results, i.e., the collected warp values, in the warp measurement result database 12.
In operation ST12, the circuit board deformation prediction apparatus 1 reads, from the design information database 10, the design information of the circuit board for which the warp values have been measured.
In operation ST13, the circuit board deformation prediction apparatus 1 reads the characteristic information of the material, which is described in the design information, from the material characteristic information database 11.
In operation ST14, the circuit board deformation prediction apparatus 1 obtains a numerical analysis result for each warp value of the circuit board for which the warp values have been measured, based on the read-out of design information and material characteristic information. In other words, the circuit board deformation prediction apparatus 1 performs a numerical analysis of the warp value of the circuit board while changing the temperature condition.
In operation ST15, the circuit board deformation prediction apparatus 1 registers the numerical analysis result of the warp value, which has been obtained by the numerical analysis, in the numerical analysis result database 13.
In operation ST16, the circuit board deformation prediction apparatus 1 calculates a differential value between the measurement result of the warp value, which is stored in the warp measurement result database 12, and the numerical analysis result of the warp value, which is stored in the numerical analysis result database 13. The circuit board deformation prediction apparatus 1 registers the calculated differential value as the warp correction value in the warp correction value database 14.
Based on the measurement result and the numerical analysis result shown in
In the above disclosed process, when the warp correction value database 14 has the data structure shown in
Further, when the warp correction value database 14 has the data structure shown in
Thus, the circuit board deformation prediction apparatus 1 constructs the warp correction value database 14 containing the warp correction values which have been obtained from the measurement results and the analysis results of the existing circuit board.
Alternatively, the circuit board deformation prediction apparatus 1 can construct the warp correction value database 14 by using warp value measurement results of a circuit board having been produced, after performing a numerical analysis of a warp value of that circuit board prior to the production.
In operation ST22, based on the read-out design information and material characteristic information, the circuit board deformation prediction apparatus 1 obtains a numerical analysis result of the warp value of the circuit board by utilizing the numerical analysis method described in, e.g., the above-cited Non-Patent Document. In operation ST23, the circuit board deformation prediction apparatus 1 registers the numerical analysis result of the warp value of the circuit board in the numerical analysis result database 13.
Next, the circuit board having been subjected to the numerical analysis is heated while changing the temperature condition. The warp value of the circuit board under the heating at each temperature condition is measured by, e.g., the moiré interference method, the laser displacement measurement method, etc. In operation ST24, the circuit board deformation prediction apparatus 1 collects the warp value measurement results. In operation ST25, the circuit board deformation prediction apparatus 1 registers the measurement results in the warp measurement result database 12.
In operation ST26, the circuit board deformation prediction apparatus 1 calculates, as the warp correction value, a differential value between the measurement result of the warp value, which is stored in the warp measurement result database 12, and the numerical analysis result of the warp value, which is stored in the numerical analysis result database 13. Further, the circuit board deformation prediction apparatus 1 registers the calculated warp correction value in the warp correction value database 14.
In response to a request for predicting a warp value of a newly designed circuit board, the circuit board deformation prediction apparatus 1 predicts the warp value of that circuit board.
In operation ST30, the circuit board deformation prediction apparatus 1 reads the design information of the circuit board from the design information database 10 in response to the prediction request.
In operation ST31, the circuit board deformation prediction apparatus 1 reads, from the material characteristic information database 11, the characteristic information of the material which is described in the read-out design information.
In operation ST32, the circuit board deformation prediction apparatus 1 performs a numerical analysis of the warp value of the circuit board. The warp value is calculated by utilizing the numerical analysis method, which is described in, e.g., the above-referred to Non-Patent Document, based on the read-out design information and material characteristic information. Further, the circuit board deformation prediction apparatus 1 performs the numerical analysis of the warp value at each of various temperature conditions,
In operation ST33, the circuit board deformation prediction apparatus 1 extracts a circuit board having a structure close to that of the target circuit board. The circuit board deformation prediction apparatus 1 reads the warp correction value of the extracted circuit board from the warp correction value database 14.
In operation ST34, the circuit board deformation prediction apparatus 1 adds the numerical analysis result of the warp value, which has been obtained in operation ST32, and the warp correction value which has been read out in operation ST33. Thus, the circuit board deformation prediction apparatus 1 corrects the warp value, which has been obtained as the numerical analysis result in operation ST32, to a value resulting from the addition.
In operation ST3S5, the circuit board deformation prediction apparatus 1 outputs the corrected warp value as the numerical analysis result.
As described above, the circuit board deformation prediction apparatus 1 performs the numerical analysis of the warp value of the circuit board. Further, the circuit board deformation prediction apparatus 1 corrects the warp value based on the differential information between the warp value of the circuit board, which has been obtained by the numerical analysis in the past, and the warp value of the circuit board, which has been actually measured in the past.
Generally, the structure of a circuit board for which a warp value is to be predicted is similar to the structure of a circuit board from which the former circuit board has been improved. Also, the structure of the circuit board affects the warp value of the circuit board to a large extent.
Accordingly, the circuit board deformation prediction apparatus 1 can correct the warp value of an improved circuit board, which is predicted at that time, by using the warp correction value obtained from the circuit board which is an original version of the improved circuit board. As a result, the circuit board deformation prediction apparatus 1 can increase the accuracy in predicting the warp value of the circuit board.
Meanwhile, according to the typical technique, the numerical analysis of the warp value of the circuit board is performed by using material characteristic information that is obtained by actually measuring the circuit board.
Material characteristic information of a circuit board including a resin contains an error at a temperature not lower than the glass transition point. Therefore, the warp value resulting from the numerical analysis based on such material characteristic information also contains a large error. For example, crosswise-braided glass fibers impregnated with a resin are generally used as material of a circuit board. The material of the circuit board is tested for stress relaxation by using the three-point bending method. In the test, the circuit board is irregularly warped due to a large drop in rigidity of the material of the circuit board at temperature not lower than the glass transition point. Hence the material characteristic information of the circuit board contains an error.
As a result, the prediction of the warp value of the circuit board by the numerical analysis based on such material characteristic information cannot have accuracy at a level demanded.
In contrast, according to the disclosed system and method, a warp deformation of the existing circuit board is actually measured in advance. Also, a warp deformation may be calculated by a numerical analysis. A differential value in the warp deformation between an actually measured result and a numerical analysis result is calculated. The differential value is stored as a warp correction value for the relevant circuit board. Further, warp correction values may be calculated for a plurality of circuit boards and the calculated warp correction values are stored in a database. When predicting a warp deformation of a new circuit board, according to the embodiment, the warp deformation of the new circuit board is calculated by the numerical analysis of that circuit board. The warp correction value of a circuit board, which is similar to the new circuit board, is obtained from the database. Further, according to the embodiment, the warp correction value is added to the numerical analysis result, thus calculating a prediction value of the warp deformation of the circuit board. As a result, a user can obtain the prediction value of the warp deformation, which is closer to an actual value.
As described above, the numerical analysis result correction unit 19 reads, from the warp correction value database 14, the warp correction value derived from data of a circuit board having a structure close to that of the circuit board for which the warp value is to be predicted.
In the above process, the numerical analysis result correction unit 19 may evaluate the number of matches by setting a weight for each of the structure items registered in the warp correction value database 14 and by calculating a total weight. For example, the numerical analysis result correction unit 19 adds 3 as the number of matches if a match is determined for the number of layers of the circuit board, and adds 1 as the number of matches if a match is determined for the size of the circuit board.
Further, the numerical analysis result correction unit 19 is not always required to evaluate a match/mismatch for the size of the circuit board. The reason is that, when a match is determined for the structure except for the size, the numerical analysis result correction unit 19 can obtain the warp correction value with substantially satisfactory accuracy through a correction operation of multiplying a certain ratio.
The process of selecting the warp correction value will be described below in connection with several practical examples.
When a target for prediction of a warp deformation is an LSI package, for example, the numerical analysis result correction unit 19 executes processing as follows. The numerical analysis result correction unit 19 makes a determination based on chip dimensions (thickness, width, length, etc) and package dimensions (thickness, width, length, etc) of a product which is the prediction target. More specifically, the numerical analysis result correction unit 19 sets upper and lower allowable error limits for each dimension in terms of percentage and selects the warp correction value that is managed in a corresponding relation to a product among a plurality of products registered in the warp correction value database 14, which has dimensions within the respective allowable ranges.
When a target for prediction of a warp deformation is a resin-molded LSI package, for example, the numerical analysis result correction unit 19 executes processing as follows. From among products registered in the warp correction value database 14, the numerical analysis result correction unit 19 selects a plurality of products each having a structure close to that of the target product. The selection is performed based on characteristic data (i.e., a rate of thermal expansion, a glass transition point temperature, and a bending elastic modulus) regarding a resin material of the package. More specifically, the numerical analysis result correction unit 19 sets upper and lower allowable error limits for each item of the characteristic data in terms of percentage and selects the warp correction value that is managed in a corresponding relation to a product among the plurality of selected products, which has data values within the respective allowable ranges.
When a target for prediction of a warp deformation is a POP LSI part including a chip mounted to a substrate, for example, the numerical analysis result correction unit 19 executes processing as follows. From among the products registered in the warp correction value database 14, the numerical analysis result correction unit 19 selects a plurality of products each having a structure close to that of the target product. The selection is performed based on condition data (i.e., maximum pressure, temperature, and time) regarding a process of mounting the chip to the substrate. More specifically, the numerical analysis result correction unit 19 sets upper and lower allowable error limits for each item of the condition data in terms of percentage and selects the warp correction value that is managed in a corresponding relation to a product among the plurality of selected products, which has data values within the respective allowable ranges.
When a target for prediction of a warp deformation is a circuit board, for example the numerical analysis result correction unit 19 executes processing as follows. The numerical analysis result correction unit 19 makes a determination based on the size (width and length), the number of layers, the thickness of a solder resist, the thickness of each insulating layer and each conductor layer, and the wiring density of each conductor layer of a product which is the prediction target. More specifically, the numerical analysis result correction unit 19 makes the determination by setting upper and lower allowable error limits for each item in terms of a numerical value or percentage and selects the warp correction value that is managed in a corresponding relation to a product which has data values within the respective allowable ranges.
When a target for prediction of a warp deformation is a circuit board, for example, the numerical analysis result correction unit 19 executes processing as follows. The numerical analysis result correction unit 19 makes a determination based on a wiring material proportion on each of both sides as viewed from the center of the circuit board in the direction of thickness. Specifically, the numerical analysis result correction unit 19 selects the warp correction value that is managed in a corresponding relation to a product having a ratio between the wiring material proportions on both the sides, which is closest to that of the target product.
When a target for prediction of a warp deformation is a circuit board, for example, the numerical analysis result correction unit 19 executes processing as follows. The numerical analysis result correction unit 19 makes a determination based on a wiring material proportion of one or plural layers among wiring layers, which are positioned nearest to the front or rear surface of the circuit board, on both sides as viewed from the center of the circuit board in the direction of thickness. Specifically, the numerical analysis result correction unit 19 selects the warp correction value that is managed in a corresponding relation to a product having a ratio between the wiring material proportions on both the sides, which is closest to that of the target product.
When a target for prediction of a warp deformation is a circuit board, for example, the numerical analysis result correction unit 19 executes processing as follows. The numerical analysis result correction unit 19 makes a determination based on the material name of a main insulating layer or the model name. Specifically, the numerical analysis result correction unit 19 selects the warp correction value that is managed in a corresponding relation to a product made of a material which is closest to that of the target circuit board.
When a target for prediction of a warp deformation is a circuit board, for example, the numerical analysis result correction unit 19 executes processing as follows. The numerical analysis result correction unit 19 selects the warp correction value that is managed in a corresponding relation to a product which is matched in a parameter having a relatively large weight with the target product.
Although a few embodiments have been shown and described, it would be appreciated by those skilled in the art that changes may be made in these embodiments without departing from the principles and spirit of the invention, the scope of which is defined in the claims and their equivalents.
Number | Date | Country | Kind |
---|---|---|---|
JP2007-017235 | Jan 2007 | JP | national |