JPH02310602A - Model synthesis type flow analysis system - Google Patents

Abstract

PURPOSE:To speedily change specification by synthesizing a flow analysis model which can analyze a whole channel from the characteristic of the channel form of a die, sequentially executing a prescribed program analysis so as to execute analysis and evaluation to the channel form, a flow control condition, a material and the like. CONSTITUTION:An input means 22 inputs the channel form of the die, the material physical property value of a forming material and the flow control condition in a system. A model synthesis means 26 extracts the characteristic of the form from the inputted channel form, judges compatibility with the analysis model from the characteristic and synthesizes the flow analysis model which can analyze the whole channel. A program execution means 28 takes out the program corresponding to respective synthesized flow analysis model from an analysis library, sets information required for executing the program from the inputted material physical property value and the flow control condition, sequentially executes the program so as to execute analysis and evaluation to the channel form, the flow control condition and the material.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a model synthesis type flow analysis system related to the evaluation and design of mold flow paths, resin properties, and molding conditions from the flow state of resin.

[Conventional technology]

Until now, as a system for plastic flow analysis, we have published Mold Technology, Volume 2, No. 11, Chapter 2, pages 16 to 19 (published by Nikkan Kogyo Shimbun (October 20, 1986)). Although it has been discussed, for a fixed analytical model,
It is not possible to change the analysis model according to the flow path shape, material property values, and flow control conditions by simply repeating various analyses.As a result, it is difficult to handle thermosetting materials whose material properties change during flow. It is something that cannot be done.

Furthermore, the conventional system evaluates the fluidity of a thermoplastic resin material when designing a mold for a housing of an electrical appliance or the like. When designing, what is important is the external appearance of the product and its design.

[Problem to be solved by the invention]

The above conventional technology does not take into account the following points,
It was not possible to apply it to the design of molds for semiconductors.

(1) Semiconductor plastic packages are made of thermosetting resin rather than thermoplastic resin used for housings of home appliances and the like. Thermosetting resin is a material that absorbs heat from the mold during molding, causes a curing reaction, and has a complex viscosity change. For this purpose, it is necessary to predict with high accuracy the viscosity change of the resin during molding. However, it is not possible to predict the viscosity change during flow of thermosetting resin.

(2) The cavity of a mold for semiconductors contains lead frames and chips that make up the semiconductor.

There are internal structures such as gold wires, and there is a high possibility that they will be deformed by resin flow during molding. Deformation of these internal structures will directly affect the quality of the semiconductor product. It is necessary to evaluate not only the resin filling situation but also the effect on the internal structure. However, when an internal structure is placed in a mold flow channel that has become flattened and narrowed, the flow of resin not only changes the internal structure.

The flow state of the resin also changes significantly, but we analyzed this change (
It is difficult to develop a program for simulation. In order to model the flow state of the resin due to its internal structure and reflect it in the program, improvements must be made while combining it with a large amount of experimental data, which requires a large amount of time and man-hours. It is virtually impossible to repeat such trial-and-error program development for each product.

(3) As mentioned in (2) above, in order to analyze the deformation of internal structures within a mold, 1. It is necessary to perform stress analysis, etc. not only from the flow state of the resin but also from information such as viscosity, flow rate, pressure, etc. obtained from the flow state j) 1=1. but,
An analysis program is created so that the data necessary for the analysis performed by the program is input in a specific form, and the output data is also output in a format suitable for expressing the results. Therefore, when performing analysis using multiple analysis programs, it is necessary to convert input and output data, but this requires considerable knowledge of program functions and input and output, which is difficult.

An object of the present invention is to provide a flow analysis system described below in order to solve the above problems and realize a flow analysis system that can be applied to the design of semiconductor molds.

(1) Provide a flow analysis system that can perform analysis and evaluation tailored to flow path shapes, flow control conditions, materials, etc.

(2) JIB product development provides a flow analysis system that can flexibly respond to changes in flow path shape, flow control conditions, materials, etc. in response to changes in design plans.

(3) Provide a flow analysis system that enables flow analysis of materials whose properties change depending on their history.

, (4) Provide a flow analysis system that can easily evaluate the validity of flow analysis results.

(5) Provide a flow analysis system that can successively expand the analysis target and analysis range.

(6) To provide a flow analysis system that can request further necessary information to perform flow analysis when analysis cannot be performed with only information such as the flow path shape, flow control conditions, materials, etc. required for analysis.

Another object of the present invention is to provide an apparatus capable of operating the flow analysis system described in the above object, and to operate the flow analysis system as described above.
The objective is to provide a device that can perform high-speed, high-precision numerical calculations on the host side and interactive processing on the workstation side.

Another object of the present invention is to provide a method for performing a composite analysis of analyzes related to flow analysis using the above-mentioned flow analysis system.

Another object of the present invention is to provide a mold evaluation method and a molding condition evaluation method that can flexibly respond to changes in mold flow path shape, molding conditions, and resin materials by using the above-mentioned flow analysis system. be.

Another object of the present invention is to provide a material design method for evaluating material components from viscosity evaluation in a fluid state using the above fluid analysis system.

[Means to solve problems]

In order to achieve the above objective, in a flow analysis system that calculates the flow state of the material and evaluates the flow path shape, flow control conditions, material, etc., the flow path shape of the mold to be analyzed,
An input means for inputting the physical properties of the material and flow control conditions, extracts the characteristics of the flow path shape, determines compatibility with the registered analysis model from the extracted features, and synthesizes a flow analysis model that can analyze the entire flow path. This system includes model synthesis means and program execution means for extracting programs corresponding to one synthesis model from an analysis library and sequentially executing them.

In addition, in a flow analysis system with the above means added,
A model reconstruction means is added that partially resynthesizes an already synthesized flow analysis model in response to changes in input data such as flow channel shape, flow control conditions, materials, etc.

In addition, in a flow analysis system with the above means added,
An experiment data registration means, a characteristic expression definition means for a material characteristic expression, a dissimilarity evaluation expression definition means for defining an evaluation expression for numerically evaluating the degree of difference between the experimental data and the value of the material characteristic expression, and the same evaluation expression. a difference evaluation means for calculating the difference between the experimental data and the value of the material property formula using A parameter correction amount calculation means for calculating from the difference degree calculated by the degree evaluation means, and a material property equation parameter estimation means are added.

In addition, in a flow analysis system with the above means added,
and analysis result registration means for registering analysis results.

A result display means is added for displaying the past analysis results registered in the analysis result registration means and the current analysis results at the same time so that they can be easily compared.

In addition, in a flow analysis system with the above means added,
A know-how registration means for registering the relationship between shape features, which are know-how for constructing a flow analysis model, and the analysis model; and a program registration means for registering a group of programs that can execute the analysis model on a computer, that is, an analysis library. , is added.

In addition, in a flow analysis system with the above means added,
A model dialogue synthesis means that divides a shape through dialogue processing, interactively defines feature quantities, and assigns an analytical model to the partial shape; a log storage means that retains input information during the dialogue processing; and a know-how registration means from the log information. a know-how generating means that registers the know-how in the know-how generating means and generates effective know-how for model synthesis; and merging the already registered know-how in the know-how registering means and the know-how generated from the know-how generating means.

This method also includes a know-how editing means to prevent conflicts between know-how.

In addition, in order to operate the flow analysis system equipped with the above means, we have created a bus that serves as a signal transmission path between connected devices, a bus control device that controls the bus, and a central processing unit that is connected to the bus. The workstation consists of a disk control device, a main storage device connected to the same central processing unit, a display device, a keyboard, and a disk connected to the same disk device.

In addition, in order to operate the flow analysis system equipped with the above means, the workstation is connected to the bus and has communication control that transmits data sent from the bus to other devices and receives data from other devices. A bus that attaches equipment and communicates data with the same workstation via a modem.

A host system consisting of a bus control device that controls the bus, a central processing unit, a disk control device, a communication control device connected to the bus, a main storage device connected to the central processing unit, and a disk connected to the disk device. This is a workstation/host AR calculation device connected to a computer.

In addition, in a flow analysis system with the above means added,
9. Register the relationship between an analysis model that allows more advanced analysis and output results in the know-how registration means, and store the program of the analysis model in the program registration means. By recording the flow analysis, it is possible to perform not only flow analysis but also a combination of analyzes related to flow analysis.

In addition, in a flow analysis system with the above means added,
By applying this to mold design, it is possible to quickly evaluate changes in specifications such as mold flow path shape, molding conditions, resin, etc.

In addition, in a flow analysis system with the above means added,
This method evaluates the viscosity of a resin with thermosetting properties in a fluid state within a mold, making it possible to determine the composition of the resin.

[Effect]

The functions of each of the above means are shown below.

First, the input means inputs into the system the flow path shape of the mold to be evaluated by flow analysis, the material property values of the material to be melted and flow-molded in the mold, and the flow control conditions of the material controlled by the molding machine. Enter within.

The model synthesis means extracts shape features from the flow path shape input by the input means, determines compatibility with a pre-registered analysis model from the extracted shape features, and analyzes the entire flow path. Synthesize a flow analysis model that can be used. The program execution means retrieves a program corresponding to each of the flow analysis models synthesized by the model synthesis means from a pre-registered analysis library, and executes the program from the material property values and flow control conditions input by the input means. Set the necessary information to run the program, run the program sequentially, and output the analysis results. Same input method.

By the model synthesis means and the program execution means,
As with semiconductor molds, it is easy to reflect the internal structure, material, and control conditions inside the mold cavity in the model, so analysis and evaluation can be performed in accordance with the flow path shape, flow control conditions, and materials. be able to.

In addition, in response to a partial change in the flow path shape, etc. inputted by the inputting means, the model reconstruction means partially changes the flow analysis model that has been synthesized by the model synthesizing means, thereby adjusting the flow rate according to the input change. Reconfigure the analytical model. The input means, model synthesis means, program execution means, and model reconstruction means described in (1) above also respond to changes in the flow path shape, flow control conditions, and materials based on changes in the design plan.

Since the flow analysis model can be resynthesized, the flow analysis can be performed according to the designer's trial and error.

Further, the material property formula parameter estimating means includes an experimental data registration means, a characteristic formula definition means, a dissimilarity evaluation formula definition means, a dissimilarity evaluation means, and a parameter correction amount calculation means. The experimental data 'U8 means has the role of inputting and registering experimental data of materials used for molding into the system. The characteristic expression definition means has the role of defining and inputting into the system a material characteristic expression that expresses the unique characteristics of the material.The dissimilarity evaluation expression definition means has the role of defining and inputting a material characteristic expression expressing the material-specific characteristics into the system. It has the role of defining an evaluation formula for numerically evaluating the degree of difference from the value of the material property formula defined by the definition means, and inputting it into the system.

The dissimilarity evaluation means uses the evaluation formula defined by the dissimilarity evaluation formula definition means to determine the difference between the experimental data registered by the experiment data registration means and the value of the material characteristic formula defined by the characteristic formula definition means. It has the role of calculating. The parameter correction amount calculation means calculates the correction amount for the material-specific parameter determined from the experimental data and is a parameter of the material property formula from the dissimilarity calculated by the same dissimilarity evaluation means, and 0 or more, which has the role of By using the material characteristic equation, it is possible to accurately predict the viscosity of a thermosetting resin whose viscosity characteristics change during flow, so it is possible to perform flow analysis on semiconductor molds.

Moreover, the analysis result registration means registers the results obtained by the flow analysis. The result display means simultaneously displays the analysis results registered by the analysis result registration means and the results of the current analysis with different conditions, etc., so that they can be easily compared. The analysis result registration means and result display means allow various analysis results to be compared and evaluated from many viewpoints, so the validity of the analysis results can be easily determined.

Further, the know-how registration means registers the relationship between the shape feature, which is the know-how for constructing the flow analysis model, and the analysis model. The program registration means registers a group of programs (analysis library) that can execute an analysis model on a computer. By using these means, it is possible to gradually expand the analysis target and the analysis range, and additional know-how and analysis programs can be registered, so that it is possible to flexibly deal with changes in the analysis target and the sophistication of the analysis contents.

Further, the model interaction synthesis means decomposes the shape through interactive processing, defines feature quantities interactively, and assigns an analytical model to one partial shape. The log holding means holds input information during interaction processing. The know-how generating means generates know-how effective for model synthesis using know-how registered in the know-how 518 means from log information. The know-how editing means merges the know-how already registered in the same know-how registration means and the know-how generated from the same know-how generating means, and prevents inconsistency between the know-hows. By using these means, the model synthesis means can synthesize a flow analysis model by performing interactive processing once, even for objects for which a flow analysis model cannot be synthesized based on the relationship (know-how) between the registered shape features and the analysis model. I can do it.

Also, buses, bus control devices, central processing units, disk control devices, main storage devices, and display devices.

A workstation consisting of a keyboard and a disk stores the flow analysis system on the main memory, stores know-how (compatibility relationships between shape features and analysis models) and analysis libraries on the disk, and stores the flow analysis system on the main memory. By operating the flow analysis system on the main storage device, the processing device can evaluate the specifications necessary for mold design.

Further, in the workstation/host computer device, the workstation side executes interactive processing, and the host computer side executes numerical calculations. As a result, the flow path shape,
It is easier to input material property values and flow control conditions, and
Calculations can be performed quickly and results can be displayed in a very easy-to-read manner.

The relationship between the analysis model for more advanced analysis and the output results can be registered in the know-how registration means, and the same analysis model can be programmed into the program H8 means 311a, so it can be used not only for flow analysis but also for flow analysis. Composite analysis related to can be performed.

This system can quickly evaluate specifications such as mold flow path shape, molding conditions, resin, etc., making it easy to evaluate molds and molding conditions during mold design. .

This system can evaluate the viscosity of the resin in the fluid state inside the mold as a material property equation, so it can be applied to material design where the composition is determined based on the moldability of the resin.

〔Example〕

The present invention will be explained below with reference to FIGS. 1 to 10.

First, to explain the flow analysis system according to the present invention, FIG. 3 illustrates its hardware configuration, and FIGS. 4(α) to (d) illustrate its software configuration. It is. As shown in FIG. 3, the hardware configuration consists of a workstation 1 and a host computer 2, and the workstation 1 and host computer 2 are connected by modems 20α and 20b. The workstation 1 has a central processing unit in the multipath 1loL which is under the control of the bus controller 12α! ! 13α, a disk device 18α via a disk control device 14cL, and a modem 20α via a communication control device 19α, and a main storage device 15α, a display device 16, and a keyboard 17 are also connected to the central processing unit 13a. It has a contained configuration. The host computer 2 connects the central processing unit 136 to the multipath 116 under the control of the bus control unit 1126, the disk device 18/ via the disk control unit gi 146, and the modem via the communication control unit 196. 2ob is connected to the central processing unit 13b, and the main storage device 15 is also connected to the central processing unit 13b.
b is accommodated. As a result, data from the keyboard 17 on the workstation 1 side is stored in the main storage device GL 15a via the central processing unit 13α, and simultaneously displayed on the display device 16, and data on the main storage device 15α is central processing unit 13α,
It is stored in the disk device 18α via the multipath 11cL and the disk control device 14α, and is also stored in the communication control device 19.
Data is transferred arbitrarily, such as by transferring data to the host computer 2 through the modem 20a and receiving data from the host computer 2 side. Further, data from the communication control device 196 via the modem 206 on the host computer 2 side is stored in the main storage device 156 via the central processing unit 136, and data on the main storage device 15b is stored in the central processing device IAB+multipath 116 and stored in the disk device 186 via the disk controller 146.

In addition, from the software configuration point of view, as shown in FIG. Program registration section 25, model synthesis section 26
, a material property equation parameter estimation section 27, and a program execution section 28. Further, as shown in FIG. 4(6), the know-how registration unit 24 includes a model registration unit 240, a shape registration unit 241, a shape feature registration unit 242, a shape division rule registration unit 243, a divided part position relationship registration unit 244, a part It is composed of a shape registration section 245 and a shape feature determination rule registration section 246.

Furthermore, the modem synthesis section 26, as shown in FIG. 4(c),
Shape feature calculation section 260, analysis program allocation section 261
, modem creation unit 262. Input/output item adjustment unit 263, shape division unit 264, partial shape feature determination unit 265, analysis modem integration unit 266, model dialogue synthesis unit 267, log storage unit 2
68, a know-how generating section 269, and a know-how editing section 270. Further, the material characteristic equation parameter estimating section 27 includes an experimental data registering section 271. as shown in FIG. It is composed of a characteristic expression definition section 272, a dissimilarity evaluation expression definition section 273, a dissimilarity evaluation section 274, and a parameter correction amount calculation section 275.

7(cL) to (f) show the contents of each of the main registration sections shown in FIG. 4''(α) to (d) as follows.

That is, the program registration unit 25 (see FIG. 7(f)) includes the analysis program Pil Pl, p3, the modeling program M, M, and the platform and data conversion program H, H3.
However, the shape feature registration unit 241 of the know-how generation unit 26
(See Figure 7 (α)) Shape features T1 to Tn and their calculation formulas (calculation formulas that use shape specifications to be analyzed as input data)
However, the shape division rule registration section 243 (see FIG. 7(B)) stores division rules for shape features, and the shape feature determination rule registration section 243 (see FIG. 7(Q)) stores analysis program determination. Shape feature estimation rules as factors are registered in advance by a skilled model analyst.

Now, assuming that the operator specifies the shape of the object A to be analyzed and the analysis item α from the keyboard 17, the shape and dimensions of the object A to be analyzed are stored in the shape registration section 241 of the know-how registration section 24. It is first registered as a value. Thereafter, the shape feature calculation section 260 of the model synthesis section 26 calculates each feature value according to the calculation formula corresponding to the shape features T8 to Tn indicated by the shape feature registration section 242 of the know-how registration section 24, and Registration 242
6 Next, the shape division rule registration unit 243 is registered in the IF-THEN format.
is registered in. In accordance with division rules that indicate how to divide the object shape into analyzable units, the partial shape registration unit 241 stores a portion A□ defined from the shapes αL, 61, and GI, and a portion A□ defined from the shapes αII, 6II, and cJ. Two partial groups of the defined portion A2 are to be registered. Also, at this time, it is determined whether or not there is a surface that is in contact with each of the parts A, A, and if there is a face that is in contact with each other, the positional relationship with each of the parts A, A, that is, Contact surface information is registered in the divided portion positional relationship registration section 244. After this, part A1. Based on the rules recorded in the shape feature determination rule registration unit 246 for each of A2, the partial shape feature determination unit 265 of the model synthesis unit 26 calculates the shape features that serve as standards for specifying the analysis program. Shape registration section 245
On the shape feature side of, for example, part A0 has features S, ,
The feature S2 is registered for the portion A2, and so on. After that, the analysis item α entered from the keyboard 17 and the part A0. A, Features for each S, ,S3. Using S as a reference value for the search, the program registration unit 25 is searched for an analysis program name that matches each of the parts A, A, and A. As a result, there is a program master for part A, and part A
The program P2 is assigned to the program P2, respectively, and is registered in the compatible program column in the partial shape registration section 245. At this time, the analysis programs p, , p2 each have their own modeling methods M, , M, and these are also registered in the program registration section 25, so the names of these modeling method programs are also the same as the partial shape. It is registered according to the R8 part 245. In response to this, the model creation unit 262 of the model synthesis unit 26 executes the program Mi for the part ○, and the partial model is further
Similarly, a partial model is created for portion A2 as well.

As mentioned above, the analysis program and its model for each divided portion have been determined, and the entire modeling will be completed once the data output items between each model are unified. To unify input and output items, first, the divided portion positional relationship registration unit 244 is referred to, and it is checked whether portions A, , A, are adjacent to each other. If part A,,A.

If it is confirmed that the input/output item adjustment unit 263
searches for a program that converts the output item 0UTl of program P from the input and output items of programs pX and p2 to the input item IN2 of program Pj, but as a result, the conversion program H1 is found, so parts A□, A2 During the analysis for each, the procedure for executing the program master is to be determined.

Through the above-described procedure, an analysis model can be generated by combining different analysis programs.

The model synthesis of the present invention has been outlined above, and next, the mathematical model of the analysis program will be explained.

Regarding the analysis program registered in the main part program registration section 25 (see FIG. 7(f)) shown in FIG. 4(α),
The explanation using FIG. 6 is as follows.

First, we will explain the basics of the model reflected in the analysis program.1 The flow analysis system to which the present invention is applied is a mold for semiconductor plastic packaging, and it is necessary to handle thermosetting resin, which is the packaging material. There is.

The isothermal viscosity formula for thermosetting resin is expressed as follows.

11 o (T) = a axp (6/T)
...-(2)tll(T)=dexp(e
/T) --(3)c (T)=f
/T −2 (4) When 1=0, η=η.

(T) ・・・・・・(5) When t=t, (T) η=Q ・・・・・・(6) Here, η:
Viscosity, η. : initial viscosity, t, gelling time, o: viscosity increase coefficient, T: absolute temperature, t: time αv b+ d* e* f+ t is a resin-specific parameter that is not affected by molding conditions. The characteristics of equation (1) at an arbitrary temperature T are shown in FIG.

In the mold, the resin flows while receiving heat from the tube wall, so it is under a non-isothermal state, and the viscosity is predicted from the isothermal viscosity equation as shown in the first order. From equation (1), here, μ=(ηc.(T)
door/llll'(T) ......(8)τ =
1 / 111 (cho) ...
...(9) is obtained, and equation (7) is as shown in Figure 6,
τ=0 and μ=1. This is a curve having characteristics such that τ=1 and μ=Q. As shown in FIG. 6, the state (τ8.μ,).

That is, τ=τ0. When the time ΔL and the temperature ΔT increase from μ=μ, the state shifts to (τ2.μ2). From equation (9), τ is a function of t and T, and the increment Δτ of τ due to a state change can be determined by the following equation.

jt jT Also, from equation (7), the value of μ when changing from τ to →τ2 is μ2=
μ, +Δμ; μm+□Δτ ...(12), and (
8) Formula η; η obtained from formula. (T)μ0(T) ・・・・・・(
13), T=T2. By substituting μ=μ2, the viscosity η2 in a new state can be found as follows.

η4=η. (T2)μ2C(T2)...
(14) By repeating the above procedure from τ=0 to 1, it is possible to calculate the viscosity change from the initial non-isothermal state to the gelled state.

Furthermore, in order to analyze the state in which the resin flows in the mold flow path, it is necessary to simultaneously solve the above-mentioned viscosity calculation method and the basic equations of continuity equation, momentum conservation equation, and energy conservation equation.

The model formula for a circular pipe flow path is as shown below.

Continuity formula: Q= 2 π0:uz rcjz ・
-=(15) Momentum conservation equation: Energy conservation equation: Two points, Q: flow rate, R: circular pipe radius, 1jz: pipe axial flow velocity, ■: pipe radial distance, z: pipe axial distance, P : Pressure, η: Viscosity, ρ: Density, T: Temperature, 2 hours, λ: Thermal conductivity Based on the above model formula, program registration part 2
A basic analysis program for flow analysis registered in 5 will be developed. In the explanation, only the circular pipe flow path has been described, but the same applies to the case of a flat plate flow path, a diffusion flow path, etc.

Now, 1. When the present invention is applied to plastic flow analysis of a mold for semiconductor plastic packages, design of mold flow path specifications and molding conditions, and selection of plastic material (resin) based on the analysis results. Taking this as an example, the processing procedure is as shown in FIG.

First, to explain the background, a semiconductor plastic package is a semiconductor plastic package in which a semiconductor chip with a lead frame attached is placed in a mold cavity, and a plastic encapsulant is inserted from the gate (inflow port) of the mold cavity. It is made by injecting and curing resin material. At this time, by conducting a flow analysis of how the viscous resin material flows within the mold cavity, it becomes possible to evaluate the specifications of the mold to be designed. However, resin materials have a thermosetting property that absorbs heat from the mold wall surface during flow, causing a significant change in viscosity, and its properties cannot be expressed using clear theoretical formulas. Experimentally, a flow analysis model program has been developed for a simple mold flow path shape in which viscosity can be measured, but the actual mold cavity structure is complex, and its complexity It is impossible to create a flow analysis model that takes this into account using only one type of calculation method. Therefore, it is necessary to create an analysis model that takes into account the complexity of the flow path shape by combining a plurality of analysis programs (models). Now, when an analyst generates a model to analyze the flow inside the mold cavity, the disk device 18
The program registration unit 25 stored therein includes resin flow analysis programs P, P2. p,, these resin flow analysis programs P
).

H(21), H(32), H(13) (f((i
, j) A program for converting output data of the resin flow analysis program PL into input data of the resin flow analysis program Pj) is registered in advance.

Here, the contents of the resin flow analysis programs P□~P and the model creation method programs M□~M will be explained with reference to FIG. This is for analyzing streamline-parallel flow that is uniformly received from 360 degrees around the circumference, and can be applied to flow parts where there is a heat source in the circumferential direction. In addition, since this program P uses the difference method, the model created by the modeling method program M2 converts the target shape (the shape of the part where the resin flows) into a circular pipe with the same flow cross-sectional area. Made. Resin flow analysis program P2
receives heat only from the top and bottom directions, which is also called a flat plate flow.
This is for analyzing flow under conditions where heat transfer from the lateral direction can be ignored (flat plate flow with small height), and because it was created using the finite element method, the direction of the streamlines ( flow direction) is automatically analyzed, unlike the finite difference method.
There is no need to specify anything in particular, and the modeling method program M2 for this is created by dividing a model of the flow part shape into triangular meshes. The resin flow analysis program P can be used under conditions in which the flow progresses radially from a single outlet, and its modeling method program M is based on a differential method.

In addition, the shape feature registration section 242 and the shape division rule registration section 243 of the know-how registration section 24 contain items (and calculation formulas) for calculating the shape features of the mold cavity 41 shown in FIG. 9 and cavities corresponding to the features. The shape division method is now registered. Here, we will explain the items for calculating the cavity shape characteristics and the details of the shape division method.The main factors that govern the resin curing reaction are One example is heat transfer. Therefore, it is necessary to divide the shape at points where the heat transfer state inside the mold cavity 41 is different, that is, at points where the resin flow is separated by the internal structure 42, to obtain portions where the heat transfer state is constant. The basis for the shape division is the shape and position of the chip 42, which is an internal structure. Here, considering the ratio of the width 2 of the cavity 41 to the width of the chip 42, the width k of the chip 42 is
of@? If it is almost the same, the flow can be considered to be divided into two parts, above and below the chip 42. On the other hand, is the width of the chip 42 the same as the width of the cavity 41? If the ratio is smaller than a certain ratio α, it is considered that the flow is divided into four parts: top, bottom, left and right of the chip 42. Therefore, the shape feature registration unit 242 includes the chip 4
2 width and the width of cavity 41? The ratio α (=person/)) is also stored in the shape division rule registration unit 243 as to whether to divide the division into two parts above and below the chip 42 or into four parts above and below the chip 42 and left and right. The judgment description is IF-
If it is registered in THEN format, the shape of the cavity 41 can be divided according to the shape characteristics. In addition to this, the shape characteristics that serve as a criterion for division include the ratio CI-/Q) of the length b of the chip 42 and the length 2 of the cavity 41.

In addition to the feature items that serve as the basis for dividing the cavity shape, the shape feature registration unit 242 also registers feature values that serve as the basis for assigning an analysis program to the divided portions.

The items that determine p,, p, include the direction of the streamline (determining factor for programs P□, P,) and shape (height of the flow, which is a determining factor for program P2, and uniformity of the heat source, which determines program P8). , that is, the aspect ratio of the shape). Among these, the characteristics that serve as standards for determining the direction of the streamline are @1 of the cavity 41 and l1l (diameter) of the gate 43.
The ratio of t (1/l) is considered. Width of cavity 41?
When the ratio of the radius 6 of the gate 43 and the radius 6 of the gate 43)/1 is large, the flow can be considered to be flowing from a narrow place to a wide place, and the streamline can be determined to be radial. On the other hand, if the ratio t/L is small, it can be determined that the streamlines are parallel. In this way, feature items and their formulas that serve as criteria for determining an analysis program are registered in the shape feature fM unit 241, and features that serve as criteria for determining an analysis program are registered in the shape feature determination rule registration unit 246. Feature criteria (α, β, etc.) for calculation results are registered in the IF-THFN~ format. Now, to explain the processing procedure shown in FIG. The mold cavity name X is taken into the system by the section 22 (100), and the mold cavity name The mold cavity shape specifications are now searchable.

In addition, if the flow analysis model cannot be synthesized only from the information registered in the know-how registration section 24, the control section 21 causes the model interaction synthesis section 26 of the model synthesis section 26 to
7 is activated and functions to allow the analyst to input necessary information via the output unit 23 to complete the flow analysis model. At this time, the log information being processed is stored in the log holding unit 26.
8, and after the processing by the model dialog synthesis unit 267 is completed, the know-how generation unit 269 converts it into a format that can be registered in the know-how registration unit 24, and the know-how editing unit 270 resolves conflicts with existing know-how. ,
Register in the know-how registration section 24.

In this example, the mold cavity shape specifications for “X′” are not registered, so the control unit 21 activates the model dialogue synthesis unit 267 and requests the input of the shape specifications via the output unit 23. , based on this, the commentator is “X”
If the mold cavity shape specifications are inputted for
1, that is, the shape registration section 2
41, the log holding unit 268 is also referenced (
110). After that, the control unit 21 controls the shape feature calculation unit 2.
60, an instruction is given to calculate the features of the input shape. Based on this instruction, the shape feature calculation section 260 receives the shape feature items and their calculation formulas from the disk device 18 via the shape feature registration section 242, and further receives the shape of 11 X IT from the disk device 18 via the shape registration section 241. After reading the specifications, feature values corresponding to each feature item are calculated, and the calculation results are registered in the shape feature registration section 242 (120).In this way, the features are calculated. After that, based on the mold cavity shape division instruction from the control unit 21, the shape division unit 264 calculates each feature value registered in the shape feature registration unit 242 and the shape division rule registered in the shape division rule registration unit 243. It is read and it is determined how to divide the shape of the mold cavity. As a result, for example, the 10th
As shown in the figure, the mold cavity is now divided into sections A-D (130). Furthermore, based on this division result, the shape division section 264 calculates the shape specifications of each divided portion A-D from the mold cavity shape specifications registered in the shape registration section 241, and
The part name and its shape specification data are sent to (140).

In response to this, the partial shape registration unit 245 registers the partial shape specifications in the disk device 18 with one part name as a heading. The shape dividing unit 264 further calculates the positional relationship between the divided parts A and D, and registers it in the divided part positional relationship registration unit 244 (150). As shown in FIG. 10 as an example, the divided parts B and C are placed next to the divided part A in parallel, and the divided parts are placed next to the divided parts B and C. A
-D's positional relationship is registered.

Following the shape division, the control unit 21 instructs the shape feature calculation unit 260 to calculate features that serve as standards for allocating analysis programs for each divided portion A to D. Based on this, the shape feature calculation unit 260 The partial shapes of divided portions A-D registered in the partial shape registration unit 245,
The feature items and their calculation formulas registered in the shape feature R14 unit 242 are read, and after calculating feature values according to the calculation formulas, they are registered in the shape feature registration unit 242 (16Q).Subsequently, The control unit 21 instructs the partial shape feature determination unit 265 to allocate an analysis program for the divided portions A to D to perform the above feature determination. In the partial shape feature determination section 265, the shape identification determination rule registration section 24
6, the IF~THEN~ format rule for estimating the factors that determine the analysis program from the feature values is read, and furthermore, the feature values of the divided parts A to D are read from the shape feature registration unit 242, and the division is performed based on these. The above-mentioned reference factors are estimated for the analysis program allocation for portions A to D, and the estimated results are registered in the partial shape registration unit 245 (170). For example, for divided part A, based on the ratio of the width of game 1-, which is the population, and the flow width, the streamline pattern is radial ``'' and the flow height is ``height'', and for 1 divided part B and C, the flow height is The height of the flow is determined to be "low", and the flow height of the divided portion is determined to be "parallel" and the flow height is "high".

As described above, when the feature determination for all divided portions A-D is completed (180), the control unit 21 assigns an analysis program to each divided portion A-D. You will be instructed to attach it. The analysis program allocation section 261 reads the characteristics of the divided portions A to D registered in the partial shape registration section 245 and registers them in the program registration section 25.

Among the flow analysis programs, one that matches the characteristics of each divided portion A to D and the program usage conditions, and a unique model creation program name are selected, and the partial shape 31.
It is now registered in the green section 245 (190),
As a result, analysis program P and model creation program M3 are applied to divided part A, analysis program P2 and model creation program H are applied to divided parts B and C, and analysis program master and model creation program are applied to divided parts. Program M is assigned to each program. Subsequently, from the control unit 21, the model creation unit 262 reads the model creation program for the divided parts A-D registered in the partial shape registration unit 245 and their shape specifications, and reads the model creation program for the divided parts A-D and their shape specifications.
- A model is created for each D (200
), the created model is sent to the disk device 18 via the model H8 unit 240 using the part name as a heading.

Once the model creation for all divided parts A-D is completed (
210), the control unit 21 instructs the input/output item adjustment unit 263 to adjust the input/output items required for successively analyzing each divided portion A to D. The input/output item adjustment unit 263 reads the positional relationship of divided parts A-D from the divided part positional relationship registration part 244, and further reads the analysis program name corresponding to divided parts A-D from the partial shape registration part 245, and inputs analysis data. Connected section of the analysis program that requires output item adjustment, that is, A on the divided part →
The program registration unit 25 searches for a data conversion program for adjusting the input/output data of B, A-+C, B-+D, and C-)D. As a result, A-
+B has the data conversion program H3z, and similarly A-+
Data conversion program H12 is in C, B-+D, C→
Since data conversion programs H and □ correspond to D, the names of these conversion programs are registered in the divided portion position relationship registration unit 244 (220).
.

After that, from the control unit 21 to the analysis model integration unit 266,
Analysis program and model corresponding to each divided part A-D,
Furthermore, input/output data conversion programs between analysis programs are integrated, and an instruction to create one overall model is issued. The analysis model integration unit 266 selects the divided portion A- from the partial shape registration unit 245.
The analysis program name corresponding to D is read out from the model registration unit 240, the modeling result corresponding to divided portions A-D is read out, and the data conversion program name corresponding to adjacent portions is read out from the divided portion positional relationship registration unit 244. , which is edited as an analysis execution procedure.

Based on this analysis execution procedure, the program execution unit 28 creates an overall flow analysis model within the mold cavity.
Analysis program P 11 P! #P3 is automatically generated in a mixed form.

As described above, according to this embodiment, multiple existing analysis programs can be combined when performing an analysis for which an analysis method has not been established, such as flow analysis in a mold cavity associated with semiconductor plastic package design. As a result, an analytical model that enables flow analysis as a whole is automatically generated.

Further, an example in which the flow analysis system according to the present invention is applied to material evaluation for evaluating the viscosity of a resin in a flow state in a mold will be described below.

First, an overview of the resin characteristic equation parameter estimating section 27, which is a component of this system, will be explained.

The semiconductor plastic packages targeted by this system are sealed with a thermosetting resin that undergoes a hardening reaction when exposed to heat. The basic formula expressing the viscosity characteristics of resin ((1
) are six resin-specific parameters α# bl
It has dl el f+ 2.

In the past, decisions were made by trial and error while drawing graphs manually. For this reason, it took more than a month to determine sufficiently satisfactory parameters, and the accuracy of the viscosity formula varied depending on the person making the determination. The quality of the viscosity formula parameters is determined by the determined parameters (1
), it depends on whether the apparent average viscosity obtained by simulation matches the apparent average viscosity obtained from the measuring device. Therefore, we ask as follows.

From the flow pressure loss ΔP of the measurement data, it is shown in the following formula.

70=9so・8P/(β・Q) ・・・・・・(
21) Here, ro: apparent average viscosity, P: pressure drop,
β: shape resistance, Q: flow rate, apparent average viscosity f0 calculated using Hagen-Boiseuille's equation, and using equation (1), under the same conditions as the measurement conditions,
Is it the apparent average viscosity obtained as a simulation result?
Evaluate the error. The evaluation formula is as follows: Vectorized.

F: Area (time, tube diameter, mold temperature) 1.1 (*.

``.):rr* and ``.Which distance (3
α = α which minimizes (α). It is to seek. That is, 0 (a a) = minG (a) =-
・(23) α becomes a field. It is to seek. Furthermore, if F(α) is defined as δ1, that is, D, (a),"', fs(a))T, then equation (23) becomes F Cca) = 0 -------・(2
5) α=α. This corresponds to finding . F(α)
When we perform Taylor expansion around α, we get F=(a+Δa
)=F(a)+F(a)・Δa+ +++ +・+
(26), and if we ignore the quadratic or higher terms of Δj, then Δ
m=-[F(s] ・80 ・・・・・・(2
7), and the iterative expression of the nonlinear equation (25) can be defined as follows. By giving an appropriate initial value of a, the viscosity equation parameters can be determined by repeating the iterative equation shown in equation (28) until convergence.

d, :a+Δa.

=a--c [F(a--t)] ・F(a,-t
)-(28) The above algorithm is realized by the material characteristic equation parameter estimation unit 27 (see FIG. 2). When an analyst estimates viscosity formula parameters using this system, the experimental data registration unit 271 of the material property formula parameter estimating unit 27 is activated by the control unit 21 in accordance with the analyst's input from the keyboard 17. , the flow pressure loss ΔP obtained from the measuring device is registered in the disk device 18 via the communication line. Then, when the experimental data registration unit 271 registers the flow pressure loss in the disk device 18.

Using equation (21), the data is converted into an apparent average viscosity and the same data is also registered in the disk device 18 (500).Furthermore, when the analyst inputs a command from the keyboard 17, the controller 2
1, the characteristic expression definition section 272 of the material characteristic expression parameter estimating section 27 is activated.

The characteristic expression definition unit 272 provides the analyzer with the display device 16.
It is required to define the viscosity equation (Equation (1)) through When the analyst inputs the viscosity formula (formula (1)) into the keyboard 17 in accordance with the message on the display device 16, the viscosity formula is imported into the system (510).Next, the control unit 21 inputs the difference evaluation formula definition. 273.

The difference evaluation formula definition unit 273 is configured through the display device 16.

Requests input of the difference evaluation formula (formula (22)). When the analyst inputs the dissimilarity evaluation formula (formula (22)) from the keyboard 17 in accordance with the message on the display device 16, the dissimilarity evaluation formula is imported into the system (520);
Then, the control unit 21 starts the difference evaluation unit 274.

The dissimilarity evaluation unit 274 uses the dissimilarity evaluation formula to calculate the dissimilarity as follows. First, the measurement conditions stored together with the measurement data in the disk device 16 are taken out and sent to the model synthesis section 26 via the control section 21 to generate a flow analysis model that can perform flow analysis under the same conditions as the measurement conditions. Then, the program execution unit 28 executes the analysis program according to the flow analysis model. At this time,
The equation input in the characteristic equation definition section 272 is used as the viscosity equation. Using the apparent average viscosity for each part and time obtained by the execution by the program execution unit 28, the degree of dissimilarity is calculated using the degree of difference evaluation formula defined by the degree of difference evaluation formula definition unit 273 (560). If the degree of dissimilarity is less than acceptable.

The difference evaluation unit 274 ends. However, if the degree of difference is greater than the allowable value, the control unit 21 is requested to start the parameter correction amount calculation unit 275 (570).

The control section 21 starts the parameter correction amount calculation section 27'5. Then, the parameter correction calculation unit 275 calculates (2
The correction amount of the viscosity formula parameter is determined by the iterative formula shown in equation 8), and the viscosity formula parameter is corrected (580).

After the correction, the parameter correction amount calculation unit 275 requests the control unit 21 to start the difference evaluation unit 274. Then, the control unit 21 activates the difference evaluation unit 274. Then, the dissimilarity evaluation unit 274 repeats the above process to make the dissimilarity less than or equal to the allowable value.

From the above, the viscosity equation parameters have been determined.

As described above, according to this embodiment, the flow characteristics of the encapsulant for semiconductor plastic packages in the mold flow path can be evaluated from a fixed perspective using the viscosity equation parameter, and the selection of the encapsulant can be made easier. It can be done easily.

〔Effect of the invention〕

Since the present invention is configured as described above, it produces the effects described below.

(1) Even if there are internal structures such as chips, lead frames, gold wires, etc. in the mold cavity, such as in semiconductor molds, a flow analysis model can be synthesized according to them, so the mold Analysis and evaluation can be performed according to the flow path shape, flow control conditions, and materials.

(2) For the flow analysis model synthesized using the present invention, flow analysis can be performed for changes in mold flow path shape, flow control conditions, and materials based on changes in the design plan resulting from molding quality. Since the flow analysis model can be resynthesized, flow analysis can be performed in accordance with trial-and-error analysis.

(3) Since the viscosity prediction formula of the thermosetting resin flowing in the mold can be automatically determined from experimental data, it is possible to accurately analyze the mold for semiconductor plastic packages.

(4) Since the flow analysis results can be displayed on the same screen so that they can be compared and evaluated from multiple viewpoints, the validity of the analysis results can be easily evaluated.

(5) To be able to gradually expand the analysis target and analysis range,
Since know-how and analysis programs can be added, it is possible to flexibly respond to changes in the analysis target and the sophistication of the analysis content.

(6) With registered know-how (relationship between registered shape features and analysis models), a flow analysis model can be synthesized by performing interactive processing once even for objects for which a flow analysis model cannot be synthesized.

(7) By operating the flow analysis system according to the present invention on the work stage 1, a mold designer can evaluate mold specifications, etc., as necessary.

(8) By running the flow analysis system according to the present invention on the workstation host computer device,
It becomes easy to input the flow path shape, material property values, and flow control conditions, and large-scale calculations can be performed at high speed. It also becomes possible to use a database (related technical information) that is often connected to a host computer.

(9) Not only flow analysis but also complex analysis based on information obtained from flow analysis can be applied, and the analysis target can be comprehensively evaluated.

(lO) By applying the present invention to the development of molds for semiconductor plastic packages, it is possible to easily evaluate molds and molding conditions, resulting in high moldability and high plastic material usage efficiency. Can make molds.

(11) By applying the present invention to the development of plastic materials, it is possible to evaluate the viscosity characteristics during flow within a mold, so it is possible to develop materials by determining the component composition from the viewpoint of moldability.

[Brief explanation of drawings]

Figures 1 and 2 are diagrams showing a schematic flow of processing procedures in an example of the flow analysis system according to the present invention, and Figures 3 and 4 are
Figures (α) to (, j) are diagrams showing the hardware configuration and software configuration, respectively, of an example of the flow analysis system, and Figures 5 and 6 are graphs showing the viscosity characteristics of thermosetting resin. Figures 7 (α) to (f) are diagrams showing the table contents of each of the main registration sections shown in Figure 4, and Figures 8, 9, and 10 are for semiconductor plastic packages. FIG. 1...Workstation, 2...Host computer. 11α, b... Multipath, 12α, b... Bus control device, 13Q, 6... Central processing unit, 1.4a, b
...Disk control device, 15α, 6... Main storage device, 16... Display device, 17... Keyboard, 18α, b... Disk device. 19α, b... Communication control device, 20α, b... Modem, 21... Control unit, 22... Input unit, 23... Output unit, 24... Know-how registration unit, 25... Program registration unit, 26...Model synthesis unit, 27...Material property equation parameter estimation unit, 28...Program execution unit,
240...Model registration section, 241...Shape registration section,
242...Shape feature registration unit, 243...Shape division rule registration unit, 244...Divided part positional relationship registration unit, 24
5... Partial shape registration section, 246... Shape feature determination rule registration section, 260... Shape feature calculation section, 261. Analysis program allocation section, 262... Model creation section, 2
63... Input/output item adjustment unit, 264... Shape division unit, 265... Partial shape feature determination unit, 266... Analysis model integration unit, 267... Model dialogue synthesis unit, 268
...Log holding unit, 269...Know-how generation unit, 27
0... Know-how editorial department, 271... Experimental data registration section, 272... Characteristic expression definition section, 273... Dissimilarity evaluation formula definition section, 274... Dissimilarity evaluation section, 275...
Parameter correction amount calculation unit, 41... mold cavity,
42...chip, 43...gate. 20th 4th (Phantom 84 Sen Shouko 4 En (C) Ko 4 En C〆) Departure 50 2nd 2nd 1st μ 3rd e Awo 70 - each

Claims (1)

[Claims] 1. In a system that calculates the flow state of a material and evaluates the flow path shape, flow control conditions, material, etc., an input means for inputting the flow path shape, material physical property values, flow control conditions; , extracts shape features from the flow path shape input by the input means, determines compatibility with a pre-registered analysis model from the extracted shape features, and synthesizes a flow analysis model that can analyze the entire flow path. A model synthesizing means to perform the process and a program corresponding to each analytical model of the synthesized flow analysis model are retrieved from an analysis library registered in advance, and the material property values and flow control conditions inputted by the input means are passed to the retrieved program. , and program execution means for sequentially executing the program, the model synthesis type flow analysis system is characterized in that it can perform analysis and evaluation in accordance with flow path shape, flow control conditions, materials, etc. 2. In the system according to claim 1, by adding model reconstruction means for reconstructing the flow analysis model in response to changes in some of the input data, changes in the flow path shape, flow control conditions, material, etc. This model synthesis type flow analysis system is characterized by its ability to flexibly respond to various situations and perform analysis and evaluation by reconfiguring the flow analysis model. 3. The system according to claim 1, further comprising: an experiment data registration means for registering experiment data; a characteristic expression definition means for defining a characteristic expression representing material properties; and an experiment data registered by the experiment data registration means and the characteristics. A dissimilarity evaluation formula definition means for defining an evaluation formula for numerically evaluating the degree of dissimilarity from the value of the characteristic formula defined by the formula definition means, and a difference between the experimental data and the value of the characteristic formula using the evaluation formula. and a parameter correction amount calculation means that calculates a correction amount of a material-specific parameter that constitutes a characteristic equation and is determined from experimental data from the dissimilarity calculated by the dissimilarity evaluation means. A model synthesis type flow analysis system is characterized in that by adding a material property equation parameter estimation means consisting of , it is possible to perform flow analysis on a material whose properties change depending on the history. 4. In the system according to claim 1, there is an analysis result registration means for registering the analysis results obtained by the system, and the analysis results registered in the analysis result registration means and the current analysis results are simultaneously stored. A model synthesis type flow analysis system characterized in that the validity of the analysis results can be evaluated by adding a result display means for displaying the results for easy comparison. 5. In the system according to claim 1, a know-how registration means for registering a relationship between a shape feature, which is know-how for constructing an analytical model, and the analytical model, and a program group capable of executing the analytical model on a computer are registered. A model synthesis type flow analysis system characterized by adding a program registration means and making it possible to successively expand analysis targets and analysis ranges. 6. The system according to claim 1, further comprising a model interaction synthesis means for dividing a shape through interactive processing, defining feature values interactively, and assigning an analytical model to the partial shape, and retaining input information (log) during the interactive processing. a log holding means; a know-how generation means for generating know-how effective for model synthesis from the know-how registered in the know-how registration means according to claim 5 from the log information; and know-how already registered in the know-how registration means and the know-how generation. By adding a know-how editing means that merges the know-how generated from the means and prevents conflicts between the know-how, the model synthesis means according to claim 1 can perform the analysis with the registered shape features. A model synthesis type flow analysis system is characterized in that it is possible to synthesize a flow analysis model by once interacting with an object for which a flow analysis model cannot be synthesized due to the relationship (know-how) with the model. 7. A bus that serves as a signal transmission path between connected devices, a bus control device that controls the bus, a central processing unit connected to the bus, a disk control device, and a main unit connected to the central processing unit. A workstation comprising a storage device, a display device, a keyboard, and a disk connected to the disk device, wherein the model synthesis means and program execution means according to claim 1 are stored on the main storage device. and storing the compatibility relationship between the shape features and the analysis model and the analysis library on the disk, so that the central processing unit operates the model synthesis means and program execution means stored in the main storage device. A model synthesis type flow analysis system. 8. A communication control device connected to a bus and configured to transmit data sent from the bus to another device and receive data from the other device is added to the workstation according to claim 7, and the workstation is connected to the workstation according to claim 7. A bus that performs data communication via a modem, a bus control device that controls the bus, a central processing unit connected to the bus, a disk control device, a communication control device, and a main memory connected to the central processing unit. A workstation/host computer device connected to a host computer comprising a disk device and a disk connected to the disk device, wherein the model synthesis means and program execution means according to claim 1 are also stored on a storage device on the side of the host computer. and storing it in an analysis library on the disk, executing the input means and model synthesis means according to claim 1 on the workstation side, and transmitting the synthesized flow analysis model through the communication control device, A model synthesis type flow analysis system in which the program execution means is executed at high speed on the host computer side, calculation results are transferred to the workstation using the communication control device, and the results can be confirmed on a display on the workstation side. . 9. The system according to claim 5, wherein a relationship between an analytical model and the output result that enables more advanced analysis is registered in the know-how registration means using output data obtained from the result of the flow analysis, and the relationship between the analytical model and the output result is registered in the know-how registration means. A flow analysis system characterized in that by registering a program in the program registration means, it is possible to perform not only flow analysis but also a combination of analyzes related to flow analysis. 10. Flow analysis characterized in that by using the system according to claim 1 for mold design, it is possible to quickly evaluate changes in specifications of mold flow path shape, molding conditions, resin, etc. system. 11. A flow analysis system in which the system according to claim 3 is used for material design to evaluate the viscosity of a thermosetting resin in a fluid state in a mold and to determine the components of the resin.