JP6539862B2 - Heat dissipation design method and program of electronic component mounting substrate - Google Patents

Description

  The present invention relates to a heat dissipation design method and program of an electronic component mounting board.

Electronic devices are being rapidly miniaturized for the purpose of improving space utilization efficiency and functionality. With the miniaturization of equipment, the board mounting density of the internal electronic components increases. From the electronic components, energy that has not been converted to work is released as heat. In particular, as in power electronics, when the operating efficiency is high but the input energy is very large or when the operating efficiency is low such as a high-brightness LED (Light Emittinng Diode), the amount of heat generated from the electronic component However, the temperature rise of the device, which in turn, tends to shorten the life and lower the operating efficiency.
In particular, when the substrate mounting density of the electronic component is increased to realize the miniaturization of the device, the electronic component thermally interferes with each other, and the temperature at the approximate center of the region where the electronic component is mounted is The temperature distribution in the substrate becomes non-uniform as it is higher than the temperature of the part.

  In general, the operating efficiency and the lifetime of the electronic component depend on the temperature, and the higher the temperature, the lower the operating efficiency and the shorter the lifetime. Therefore, when the temperature in the substrate becomes nonuniform, the operating efficiency and the life become nonuniform. In such a state, in order to ensure desired reliability even in a place where the temperature is the highest in the substrate, it is necessary to suppress the temperature of the relevant part to a specified value or less. Due to the non-uniformity of the internal temperature, the temperature is suppressed to an excessively low level compared to the specified value, so that the heat dissipation system including the heat sink etc. becomes excessive performance, and as a result, the heat dissipation system and consequently the equipment Inhibit the overall miniaturization.

In addition, if non-uniformity (temperature non-uniformity) of temperature distribution occurs in the LED light source substrate, it leads to non-uniformity of operation efficiency of the LED, that is, light emission efficiency, resulting in luminance non-uniformity in the light source substrate The quality of the light source is reduced. Furthermore, when the substrate mounting density of the LED chips is increased, the chips thermally interfere with each other, the temperature of the substantially central portion of the substrate becomes higher than that of the peripheral portion, and the light emission amount of the central portion called "donutization phenomenon" Problems of decreasing the
Therefore, in order to reduce the size and improve the operation efficiency and quality, it is important to make the temperature distribution in the substrate uniform (suppress the temperature unevenness within a certain range).

Conventionally, for smoothing the substrate temperature, a metal core substrate in which a metal having high thermal conductivity such as copper is included in the substrate, or a flat heat pipe such as a vapor chamber which is a latent heat transport type heat conducting member has been used.
For example, in patent document 1, it has an arrangement | positioning part for providing a to-be-cooled apparatus in any one of an upper plate and a lower plate, and provides one or several middle plates between the said upper plate and the said lower plate. A cooling unit main body, and inside the cooling unit main body, a vapor diffusion passage for transmitting heat generated by the cooled device as refrigerant to the peripheral portion of the cooling unit main body; And a capillary channel configured to return the refrigerant condensed in the peripheral portion to the side of the placement portion, and the placement portion is formed to be thinner than the other regions, A heat pipe provided with a recess for mounting the device to be cooled has been proposed.

Simulation is often used for heat dissipation design. However, in the case of trial-and-error design by simulation, it is necessary to study many factors, and it takes a huge computational cost, which may make optimization difficult under realistic computational time and cost. There were many.
Therefore, in Patent Document 2, the heat conduction from the heating end to the cooling end is modeled as one-dimensional heat conduction to determine the amount of heat release for each region, and the heat conduction state from the heating end to the cooling end is analyzed. An analysis method of a plate type looped capillary tube heat pipe has been proposed.

Patent No. 4119944 Japanese Patent Application Laid-Open No. 10-222492

In the heat dissipation design of the electronic component mounting substrate, design parameters such as substrate size, heat input, heat source size (mounting area), allowable temperature nonuniformity, thermal conductivity of substrate, substrate cooling condition, peak temperature, etc. There is a need. However, in order to optimize these design parameters, complicated calculations have to be repeated, which requires a great deal of effort.
While there are unknown parameters, there is a need to quickly carry out a heat dissipation design to suppress substrate temperature unevenness within a certain range, but no design method has been proposed to solve this need.

  Then, this invention aims at providing the thermal radiation design method and program of an electronic component mounting board which can solve the problem of the temperature nonuniformity of an electronic component mounting board.

A first invention is a method of designing an electronic component mounting substrate including a substrate and a large number of semiconductor elements integrated and mounted on the substrate, wherein design parameters include a substrate temperature distribution, a substrate surface thermal conductivity, and a substrate thickness direction. Enter values for design parameters excluding one unknown parameter selected from among the design parameters, including thermal conductivity, substrate size, substrate thickness, cooled body size, heat input, and overall heat transfer coefficient, The bionumber Bi _{r of the} substrate, the calculation parameter a, and the dimensionless coordinate r ^* calculated by the following equation 1 are calculated, and the following equation 2 is used without the calculation parameter a and the dimensionless coordinate r ^*: The dimensional temperature 算出 is calculated, and the temperature T _R at the dimensionless coordinate r ^* = 1 is calculated by the following equation 3 including S _mod calculated from known data, and the temperature distribution of the substrate by the following equation 4 A heat dissipation design method of an electronic component mounting board characterized by calculating T [K] .
[Equation 1]
[Formula 2]
Here, each coefficient is as shown below.
h: Overall heat transfer coefficient [W / (m ² · K)]
k _r : Thermal conductivity in the substrate surface direction [W / (m · K)]
R: Size of cooled object [m]
d: Substrate thickness [m]
I ₀ : Modified Bessel function of the first kind
[Equation 3]
T _R : Temperature at end of object to be cooled [K]
k _z : thermal conductivity in the substrate thickness direction [W / (m · K)]
A: Cooled body area [m ² ]
h: Substrate backside cooling strength [W / (m ² · K)]
S _mod : shape factor [-]
[Equation 4]
T _:: Environmental temperature [K]
Q _in : Heat input [W]
A: Cooled body area [m ² ]
h: Substrate backside cooling strength [W / (m ² · K)]

A second invention is a method of designing an electronic component mounting substrate including a substrate and a large number of semiconductor elements integrated and mounted on the substrate, wherein the substrate effective thermal conductivity k _e [W / (m · K)] is constant. The first model in which the overall heat transfer coefficient h [W / (m ² · K)] is changed, or the overall heat transfer coefficient h [W / (m ² · K)] remains constant, and the substrate effective thermal conductivity k _e For the second model in which [W / (m · K)] is varied, the range of effective thermal conductivity k _e [W / (m · K)] or the overall heat transfer coefficient h [W / (m ^{2) of the case} A method of calculating the range of K), wherein an approximate curve of the temperature unevenness of the substrate with respect to the bionumber is calculated based on the first model or the second model, and a target range of the temperature curve of the approximation curve and the substrate Yes from calculating the range of the Biot number, by the following equation 5 based on the range calculated number of Biot substrate The thermal conductivity _{k e [W / (m ·} K)] range or overall heat transfer coefficient h of the housing ^{[W / (m 2 · K} )] range of the electronic component mounting board, characterized in that to calculate the in Heat dissipation design method.
[ Equation 5 ]
Here, each coefficient is as shown below.
h: Overall heat transfer coefficient [W / (m ² · K)]
k _e : Effective thermal conductivity of substrate [W / (m · K)]
R: Size of cooled object [m]

A third invention is the heat dissipation design method for an electronic component mounting board according to the first invention, characterized in that the design parameters further include a substrate maximum temperature, a substrate temperature unevenness, an environmental temperature, and an area to be cooled. is there. The reason why the substrate effective thermal conductivity is used in Equation 3 is that it is difficult to evaluate the anisotropy of the substrate thermal conductivity, and it is for dealing with the case where only the effective thermal conductivity is disclosed.
According to a fourth aspect of the present invention, the electronic component mounting substrate has a plurality of LED chips integratedly mounted on the substrate, a translucent resin portion covering the LED chips, and a wiring layer formed on the substrate. A method for heat dissipation design of an electronic component mounting board according to any one of the first to third inventions, which is an LED light emitting module including an LED light emitting module and a housing on which the LED light emitting module is mounted.
A fifth invention is a computer program that causes a computer to execute the method for designing heat dissipation of an electronic component mounting board according to any of the first to fourth inventions.

  According to the present invention, it is possible to quickly perform heat dissipation design for suppressing temperature unevenness of the electronic component mounting board within a certain range.

It is a block diagram in the case of thermally evaluating a LED lighting module. It is a figure which shows the input-output screen of a thermal design program. It is a flowchart of processing of a thermal design program. It is a detailed flow chart of calculation processing of a thermal design program. It is a top view of the LED lighting module concerning Example 1 and 2. FIG. It is sectional drawing of the LED lighting module which concerns on Example 1 and 2. FIG. It is a figure showing the mounting pattern of the LED chip of the LED lighting module concerning Example 1 and 2. FIG. It is a figure which shows the result of having measured temperature distribution of the board | substrate of the LED lighting module concerning Example 1. FIG. It is the graph which compared the analysis solution of the temperature distribution of the board | substrate of the LED lighting module which concerns on Example 1 and 2, and a measurement result. It is an (a) perspective view of the underwater high-intensity LED illumination concerning Example 3, (b) Top view, and (c) Cross section. (A) top view of the heat-receiving part and board | substrate of underwater high-intensity LED illumination which concern on Example 3, (b) Side view, (c) It is a bottom view. It is a simulation result of the temperature nonuniformity of the underwater high-intensity LED illumination which concerns on Example 3. FIG.

  SUMMARY OF THE INVENTION The present invention provides a heat dissipation system design method for smoothing the temperature distribution of an electronic component mounting substrate and making the operating conditions of the mounted semiconductor devices uniform. This heat dissipation system design method is realized by a dedicated program executed on a computer, and the thermal conductivity and the substrate required of the substrate can be obtained by giving the maximum temperature and temperature unevenness allowed in the substrate. It is possible to quantify the heat dissipation performance required for the casing in which the Although the present invention can be applied to electronic devices in general such as a system using power electronics such as an LED light source system and an inverter, a thermal design example of the LED lighting module will be described below.

[Thermal design program]
The thermal design program according to the embodiment is a program executed on a personal computer, and can execute the thermal design method according to the embodiment. As design parameters, four dimensions and seven thermal conditions can be input. Among the eleven design parameters, by inputting ten parameters, it has a function of automatically calculating one unknown parameter X based on the algorithm of the present invention. The unknown parameter X can select any parameter out of 11 design parameters. Details of the parameter input screen of the thermal design program, each parameter, and the algorithm will be described later.

[LED lighting module]
The (a) top view of this LED lighting module 1 is shown in FIG. 1, and the (b) side view is shown. The LED lighting module 1 has a mounting area 10 and a substrate 20, in which a large number of LED chips are disposed, as main components, and the thermal characteristics are evaluated using a liquid-cooled cooler (30 to 33) controlled to a constant temperature. Be done.
The mounting area 10 is an area where a large number of LED chips are mounted at high density, and is approximated as one heating element.
The mounting area 10 is mounted on the substrate 20, and the substrate 20 plays a role of transferring the heat generated by the mounting area 10 to the metal block 30 of the liquid-cooled cooler.
The metal block 30 of the liquid-cooled cooler is used to keep the temperature of the back surface 20B (surface opposite to the surface on which the mounting area 10 is mounted) of the substrate 20 constant, for example, the thermal conductivity of aluminum or copper The main component is a block of good metal. A circulation channel 31 through which a refrigerant such as antifreeze liquid passes is provided in the metal block 30 of the liquid cooling cooler, and the inlet 32 and the outlet 33 of the circulation channel 31 are connected by pipes to an external circulator not shown. ing. The refrigerant is pumped by a pump in the circulator and circulates between the circulation channel 31 and the circulator. The circulator controls the temperature of the refrigerant while monitoring the temperature measured by a temperature sensor (not shown) inserted in the circulator in order to keep the temperature of the metal block 30 of the liquid-cooled cooler constant. Thermal grease (not shown) is applied to the surface 30A of the liquid cooling cooler metal block 30 to lower the thermal resistance, and the substrate 20 is fixed to the liquid cooling cooler metal block 30 via the thermal grease. ing.

  Design parameters in the present embodiment will be described below with reference to FIG.

[L _hs : board size]
The substrate size L _hs is defined as half the maximum length of the substrate 20. Therefore, the shape of the substrate is not limited to the illustrated shape, and may be, for example, a polygon or an L-shape, and in this case, the substrate size L _hs is set based on the maximum length. The substrate size L _hs affects the cooling capacity, and the larger the size is, the better the cooling capacity is, but it hinders miniaturization, so the design is made in consideration of the balance between the size and the cooling capacity.
[D: substrate thickness]
The substrate thickness d is defined as the thickness of the substrate 20. The larger the substrate thickness d, the lower the thermal resistance of the substrate 20 in the surface direction of the substrate 20 decreases, and the temperature distribution of the mounting region 10 is easily made uniform. However, the thermal resistance of the substrate in the thickness direction increases. Design in consideration of the balance of thermal resistance in the direction.
[R: Size of object to be cooled]
The cooled object size R is defined as a half of the size of the mounting area 10. The larger the amount of heat generation, the larger the amount of heat generation, the easier it is to be cooled, but in consideration of specifications such as the amount of luminous flux, the mounting density of the LED chip, and the like.
[A: area to be cooled]
The cooled body area A is the area occupied by the mounting area 10. The concept of design is the same as the size of the object to be cooled R.

[Q _in : Heat input]
The amount of heat input Q _in is the total amount of heat transferred from the mounting area 10 to the substrate 20. Of the energy input to the mounting area 10, a fixed percentage of energy not converted to light is generated as heat. The generated heat is transferred to the air around the mounting area 10 and the substrate 20, but in the case of mounting in a normal electronic device, about 90% moves to the substrate side, and in such a case, Q _in is a calorific value Set as about 90%. In the case of using a liquid-cooled cooler, high-performance heat sink or the like to cool the back of the substrate as in the case of the LED lighting module 1, the heat transfer to the air is small and ignored. Assume heat transfer. The amount of heat input Q _in is determined by the light emission efficiency, power consumption, the number of mounted LEDs, and the like.
[T _:: Environmental temperature]
The environmental temperature T _は is the temperature of the metal block 30 of the liquid-cooled cooler. In design, consider using the operating environment temperature.
[K _r : thermal conductivity in the substrate surface direction, k _z : thermal conductivity in the substrate thickness direction, k _e : substrate effective thermal conductivity]
The in-plane thermal conductivity k _r is the in-plane thermal conductivity of the substrate 20, and the in-plane thermal conductivity k _z is the in-thickness thermal conductivity of the substrate 20. When using a high performance heat pipe type heat spreader (Fine Grid Heat Pipe: FGHP) or a graphite product as the substrate for the substrate 20, the thermal conductivity has anisotropy in the surface direction and thickness direction due to structural factors. , Prepared as separate parameters. The substrate effective thermal conductivity k _e is the thermal conductivity when it is assumed that the thermal conductivity in the plane direction and the thickness direction is substantially equivalent.

[H: Substrate backside cooling strength]
The substrate back surface cooling strength h is a coefficient representing the heat transfer efficiency from the substrate 20 to the metal block 30 of the liquid-cooled cooler, and is equivalent to the overall heat transfer coefficient in the LED lighting module 1. It is a constant determined by the shape and surface state of the substrate 20 and the fluid (eg, air) with which the substrate 20 contacts.

[T: Substrate temperature distribution]
The substrate temperature distribution T is a distribution of the temperature of the substrate 20. Since the maximum substrate temperature T _max , the minimum substrate temperature T _min , and the temperature unevenness T _max −T _min to be described later are determined automatically as a function of the substrate temperature distribution T, these are together with the substrate temperature distribution T as one design parameter deal with.
[T _max : Maximum board temperature]
The substrate maximum temperature T _max is the maximum value of the temperature of the substrate 20. The temperature of the substrate 20 affects the deterioration of the LED chip, and the larger the value, the more easily the LED chip is deteriorated and the life is reduced. Therefore, the substrate maximum temperature T _max is designed not to be too high.

[T _min : board minimum temperature]
The substrate minimum temperature T _min : is the minimum value of the temperature of the substrate 20. In the LED lighting module, although there is no problem on the low temperature side, consideration is required to calculate temperature unevenness as a difference from the substrate maximum temperature T _max .

[T _max −T _min : temperature unevenness]
The temperature non-uniformity T _max −T _min causes a difference in the light emission amount of the LED chip, which causes the deterioration of the illumination quality. Furthermore, temperature unevenness in the substrate also causes variation in the lifetime of the LED chip, the phosphor, the resin and the like. Therefore, it is designed so that temperature nonuniformity _Tmax - _Tmin may be held within the allowable range.
Similar to the substrate maximum temperature _Tmax , the substrate minimum temperature _Tmin is defined as the minimum value of the temperature of the substrate 20, and the temperature unevenness _Tmax - _Tmin is determined.

[Parameter input screen]
FIG. 2 shows an input / output screen 40 of a thermal design program (hereinafter sometimes simply referred to as a "program"). The input / output screen 40 mainly includes a design parameter name column 41, a design parameter input / output column 42, a setting column 43, a calculation result display column 44, and a calculation execution button 45. It corresponds to the above-mentioned design parameters, and in the figure, symbols representing each design parameter are attached.
The design parameter name column 41 describes each of the design parameters described above, and the symbols in the drawing are symbols representing the corresponding design parameters.
The design parameter input / output section 42 is a section for inputting a fixed value of the design parameter or outputting a solution when the program calculates the design parameter.
The setting column 43 is a list box, and if you want to fix the design parameter value with the value input in the design parameter input / output column 42, let the program calculate the design parameter “Use left value”. If you want, you can select "Find ...". When the latter is selected, the corresponding design parameter input / output column 42 is grayed out and can not be input thereafter.
The calculation result display column 44 displays the calculation result of the temperature distribution, with the coordinates and the temperature in parallel. If the check box “Refresh calculation window every time” in the lower part of the calculation result display column 44 is checked, the calculation results under execution will continue to be updated until the solution converges when the program is made to calculate design parameters.
When the calculation execution button 45 is clicked, the program executes the calculation.

Processing flow
FIG. 3 shows a flowchart of the process of the thermal design program. This program optimizes unknown design parameters such that one of the design parameters is unknown and the target value of the given temperature non-uniformity T _max −T _min is obtained. Moreover, when temperature nonuniformity _Tmax - _Tmin is unknown, optimization is not performed but the calculated value of temperature nonuniformity _Tmax - _Tmin is returned and it complete | finishes.
Although the bisection method was used as an optimization algorithm, it is not limited to this.
Description will be made with reference to FIG. 2 along the step numbers (S1 to S9) assigned to the respective processes in FIG.
(S1)
Among the design parameters, one unknown design parameter X is selected, and "Find ..." is selected in the corresponding setting field 43. In the design parameter input / output section 42 of the design parameter to be fixed, a fixed value of each design parameter is input.
(S2)
It is determined whether an unknown parameter is temperature nonuniformity _Tmax - _Tmin . If the unknown parameter is temperature unevenness T _max −T _min , the process proceeds to S12.

(S3)
The section upper limit X ₁ and the section lower limit X ₂ of the unknown parameter X are set. Interval upper limit X ₁ and interval lower limit X ₂ is given in advance as a database for each design parameters as in Table 1, it is set by referring to the database.
(S4)
The midpoint X _M between the section upper limit X ₁ and the section lower limit X ₂ is calculated. The calculation formula of X _M is given by Equation 1.
X _M : midpoint X ₁ : section upper limit X ₂ : section lower limit

(S5)
Segment limit _{X 1,} section lower _{X 2,} and the substrate temperature distribution _T 1 at an intermediate point _{X M, T 2,} and calculates the _{T M.} The detailed calculation method will be separately described as S51 to S56.
(S6)
Temperature unevenness Y ₁ , Y ₂ , and Y _M are calculated from substrate temperature distributions T ₁ , T ₂ , and T _M. The calculation formula of the temperature non-uniformity Y ₁ , Y ₂ and Y _M is given by Equation 2.
Y _i : Temperature unevenness [K]
T _{i_max} : Maximum value of substrate temperature [K]
T _{i_min} : Minimum value of substrate temperature [K]

(S7)
The residue [Delta] Y ₁ is a difference value between the temperature variation _Y 1, _{Y 2,} and _{Y M} and the design parameters input column 42 set value _{Y 0} of the temperature unevenness Y input in calculates the [Delta] Y _2, and [Delta] Y _M. The residue [Delta] Y _1, formula for [Delta] Y _2, and [Delta] Y _M is given by Equation 3.
ΔY _i : Residue [K]
Y ₀ : Set value of temperature unevenness [K]

(S8)
The residue [Delta] Y _M is determined whether the following target value epsilon. The target value ε is determined by the accuracy, and the smaller the value is, the closer to the true solution but the number of calculations increases. The target value ε is given, for example, a value of 0.01K. If residue [Delta] Y _M is less than the target value epsilon, the accuracy is secured, the X _M and a solution of unknown parameter X, the process proceeds to S11. Since the residue [Delta] Y _M is larger than the target value epsilon, accuracy will be not satisfied, it calculates again proceeds to S9.
(S9)
It is determined whether the positive or negative signs of the residues ΔY ₁ and ΔY ₂ are different from each other. When the positive and negative signs of the residues ΔY ₁ and ΔY ₂ are different, a solution of X with Y = Y ₀ is present between the sections X _{1 and} X ₂ . If the positive and negative signs of the residues ΔY ₁ and ΔY ₂ are the same, there is no solution of X for Y = Y _{0 in the} interval X _{1 to} X ₂ , so an error is returned and the program is ended.

(S10)
The residue [Delta] Y ₁ or of either of [Delta] Y _2, which determines whether equal positive or negative sign is the sign of the residue [Delta] Y _M. If sign of positive or negative sign and the residue [Delta] Y _M residues [Delta] Y ₁ are equal, changes the value of the section limit X ₁ to X _M, a value of the section upper limit X ₂ is maintained as it is. If sign of residue [Delta] Y ₂ of the positive or negative sign and a residue [Delta] Y _M are equal, it changes the value of the section limit X ₂ in X _M, the value of the interval lower limit X ₁ is maintained. And it returns to S4.
(S11)
The solution X _M is displayed in the design parameter input / output field 42 of the unknown parameter X, and the program ends.
(S12)
The substrate temperature distribution T is calculated. The calculation method is the same as that of S51-56.

(S13)
From the substrate temperature distribution T, the temperature unevenness Y is determined. The calculation formula of the temperature unevenness Y is given by Equation 4.
Y: Temperature unevenness [K]
T _max : Maximum substrate temperature T _min : Minimum substrate temperature

(S14)
The temperature unevenness Y is displayed in the design parameter input / output column 42 of the temperature unevenness, and the program ends.

The detailed calculation steps S51 to S56 of S5 and S12 will be described with reference to FIG.
(S51)
Request Biot number Bi _r by the number 5 of the formula.
Bi _r : Bionumber h: Substrate backside cooling strength [W / (m ² · K)]
R: Size of cooled object [m]
k _r : Thermal conductivity in the substrate surface direction [W / (m · K)]

(S52)
The calculation parameter a is determined by the equation (6).
a: Calculation parameter (related to bio number)
R: Size of cooled object [m]
d: Substrate thickness [m]

(S53)
The coordinate r is defined as array data in which each element is a value obtained by dividing a value between 0 and a cooled object size R at a constant number or at regular intervals. By the equation 7, the dimensionless coordinates r ^* are calculated in which the coordinates r are normalized by the cooled object size R.
r ^* : dimensionless coordinates r: coordinates [m]
(S54)
The dimensionless temperature Θ is determined by equation (8).
Θ: dimensionless temperature I ₀ : first-order modified Bessel function

(S55)
The temperature T _R at the end position of the object to be cooled (r ^* = 1) is determined by the equation (9).
T _R : Temperature at end of object to be cooled [K]
k _z : thermal conductivity in the substrate thickness direction [W / (m · K)]
A: Cooled body area [m ² ]
h: Substrate backside cooling strength [W / (m ² · K)]
S _mod : shape factor [-]
Here, S _mod is a shape factor, which can be calculated from data when the substrate thermal conductivity and cooling conditions are known.

(S56)
The substrate temperature distribution T is obtained by equation 10.
T: Substrate temperature distribution [K]
T _:: Environmental temperature [K]
Q _in : Heat input [W]
A: Cooled body area [m ² ]
h: Substrate backside cooling strength [W / (m ² · K)]

According to the heat dissipation design program of the embodiment described above, one unknown parameter X can be calculated based on the algorithm of the present invention by inputting 10 parameters out of 11 design parameters including temperature unevenness. It becomes possible to calculate automatically.
The number of design parameters is not limited to 11 in the example, and the minimum required parameters among them are substrate temperature distribution, substrate thermal conductivity, substrate size, substrate thickness, substrate size to be cooled, heat input, and It is preferable to include parameters of the environmental temperature and the maximum substrate temperature from the viewpoint of durability, which is the back surface cooling strength of the substrate.

  Hereinafter, the present invention will be described in detail by way of examples, but the technical idea of the present invention is not limited to the examples.

Example 1 shows an analysis result in the case where the substrate 20 of the LED lighting module 1 is made of a copper plate having a thickness of 2 mm.
[Constitution]
FIG. 5 is a top view of the LED lighting module 1. 6 is a cross-sectional view of the LED lighting module 1 taken along line B-B in FIG.
Hereinafter, the configuration of the LED lighting module 1 will be described using FIGS. 5 and 6.

As shown in FIG. 5, the LED lighting module 1 mainly includes a mounting area 10, wires 12 (12a and 12b), external connection terminals 13 (13a and 13b), a mounting substrate 20, and a dam material 22.
The mounting area 10 is an area surrounding the outermost periphery of 2210 LED chips 11 mounted at a high density so-called COB (Chip On Board), and is an area which is a light emission source and a heat generation source. The mounting area 10 has different power supplies supplied at the boundary of the line A-A in FIG. 5, and is divided into a mounting area 10a on the left side and a mounting area 10a on the right side. In the left mounting area 10a and the right mounting area 10a, 1105 LED chips 11 are mounted.
The LED chip 11 is, for example, an InGaN-based blue LED bare chip. The LED chips 11 are electrically connected to the LED chips 11 or to the wirings 12 (12a, 12b, 12c, 12d) by the wires 14. The wire 14 is made of, for example, a metal material such as gold or aluminum, and is formed and bonded by wire bonding. The back surface (lower surface) of the LED chip 11 is fixed on the front surface of the substrate 20 by a high thermal conductive adhesive or the like. The arrangement of the LED chips 11 will be described in the section of [High-density mounting] described later.

  The wiring 12 and the external electrode terminal 13 are disposed on the surface of the substrate 20 so as to surround the mounting area 10 and supply power to the LED chip 11. The wires 12 and the external electrode terminals 13 are electrically connected to each other with the same reference numerals (a, b, c, d), and a constant distance is taken with the line AA in FIG. 5 as an axis of symmetry. It is open and forms a line symmetrical pattern. The external electrode terminal 13 is electrically connected to an external power supply (not shown), and a voltage is applied between the external electrode terminals 13a and 13b and between the external electrode terminals 13c and 13d. A fixed distance is provided between the wires 12a and 12b and between the wires 12c and 12d, and a protection diode device 15 for preventing breakdown due to an overvoltage is connected. Further, on the surface of the wiring 12, an inorganic white insulating layer 24 which is made of the same material as the inorganic white insulating layer 21 and which also functions as a solder resist is formed.

  The substrate 20 is made of a copper plate with a diameter of 120 mm and a thickness of 2 mm, and is electrically insulated by the inorganic white insulating layer 21 formed on the surface and conducts heat of the mounting region 10 with low thermal resistance. Heat is released.

  The inorganic white insulating layer 21 is formed on the surface of the substrate 20 by coating, and serves as a reflection layer of light emitted from the insulating layer and the mounting area 10. The inorganic white insulating layer 21 contains white inorganic powder (white inorganic pigment) and silicon dioxide (SiO 2) as main components, and has a withstand voltage of about 1 KV / 10 μm and an average reflectance of 70% or more in the visible light wavelength range. It has physical properties. The thickness of the inorganic white insulating layer 21 is, for example, 10 to 150 μm, and for example, an inkjet method, a dispenser method, a spray coating method, or a screen printing method is used for coating formation.

  The dam material 22 is formed in a substantially square shape surrounding the mounting area 10 on the surface of the substrate 20 and functions as a bank for filling the translucent resin 23. In addition, light reflectivity is imparted to the surface of the dam member 22, and the light emitted from the mounting area 10 is reflected.

  The translucent resin 23 is filled in the area surrounded by the dam material 22 and seals the LED chip 11 and the wire 14. This protects the LED chip 11, the wires 14 and their junctions, and adjusts the light emitted from the LED chip 11 to a desired color. In the translucent resin 23, for example, a phosphor for obtaining white is mixed. As a system for obtaining white light, a system in which a phosphor of three primary colors is excited by an ultraviolet LED, a system in which a two-color phosphor of green and red is excited by a blue LED, and a system in which a yellow phosphor is excited by a blue LED Wherein, the phosphors selected according to the type of the LED chip 11 are mixed. The color temperature is set, for example, in the range of 2700 to 6500K.

[High density mounting]
The arrangement and electrical connection of the LED chips 11 mounted in the mounting area 10 will be described using FIG. 7.
FIG. 7 is a simplified view of the LED chip 11 mounted in the first mounting area 10a of FIG. 5 and the wirings 12a and 12b. The LED chips 11 mounted in the second reflective area 10b are also arranged in the same manner.

  The LED chips 11 are arranged in a zigzag at substantially equal pitches in the vertical and horizontal directions in the first mounting area 10a. The LED chips 11 are arranged in 32 odd-numbered rows and 33 in even-numbered rows, forming a total of 34 rows. The leftmost LED chips 11 in the odd-numbered rows have the same position in the left-right direction. Further, the leftmost LED chips 11 in the even-numbered rows also have the same position in the left-right direction. Furthermore, the leftmost LED chips 11 in the odd-numbered rows are offset to the right substantially by a half pitch with respect to the leftmost LED chips 11 in the even-numbered rows.

Here, the position of the LED chip 11 is defined. As shown in FIG. 7, an X direction is defined to the right and a Y direction is defined to the lower, with reference to the upper left LED chip 11. Then, the index (X, Y) is assigned to the LED chip 11 that belongs to the Y-th row from the reference and is the X-th from the left end.
For example, the LED chips 11 belonging to the first row are assigned indexes (1, 1), (2, 1), (3, 1),..., (32, 1) from the left end and Indexes (1, 34), (2, 34), (3, 34),..., (33, 34) are assigned from the left end to the LED chips 11 to which they belong.
Hereinafter, the LED chip 11 to which the index (X, Y) is assigned is referred to as “LED chip 11 (X, Y)”.

The LED chips 11 (X, Y) are electrically connected in series to the wirings 12 a, 12 b by wire bonding as described below.
When Y = 1 or 2, a pair of electrodes (not shown) at the upper end of the LED chip 11 (X, Y) is wire-bonded to the wiring 12 a.
In the case of 3 ≦ Y ≦ 32, the pair of electrodes (not shown) at the lower end of the LED chip 11 (X, Y-2) is the pair of electrodes (upper end) of the LED chip 11 (X, Y) Wire bonded to (not shown). Also, a pair of electrodes (not shown) at the lower end of the LED chip 11 (X, Y) is wire-bonded to a pair of electrodes (not shown) at the lower end of the LED chip 11 (X, Y + 2) Ru.
When Y = 33 or 34, a pair of electrodes (not shown) at the lower end of the LED chip 11 (X, Y) is wire-bonded to the wiring 12 b.

As described above, since the LED chips 11 are electrically connected to each other at every other line in the vertical direction, 17 pieces of the LED chips 11 are electrically connected in series to the wirings 12a and 12b. When the LED chips 11 electrically connected in series are defined as a “unit LED chip group”, the unit LED chip groups are the LED chips 11 to which the same index X is assigned in the odd row group or the even row group The LED chip 11 is assigned the same index X. Since the value of index X is X = 1 to 32 in the odd rows and X = 1 to 33 in the even rows, 32 unit LED chip groups are formed in the odd row group and 33 in the even row group It is formed. A total of unit LED chip groups formed in the first mounting area 10 a is 32 + 33 = 65. Since each unit LED chip group is electrically connected in parallel to the wirings 12a and 12b, 1105 LED chips 11 are mounted in a 17 series × 65 parallel wiring pattern in the first mounting area 10a. It will be.
Since the second mounting area 10b has the same configuration as the first mounting area 10a, 2210 LED chips are mounted in 17 series × 130 parallel in the mounting area 10 in which both areas are combined.

[Analysis result]
Table 2 shows each design parameter of the present embodiment input to the thermal design program. Since the unknown parameter is the substrate temperature distribution T, the dependent substrate maximum temperature T _max and the temperature unevenness T _max −T _min are not inputted into the program as the unknown parameter.

FIG. 8 shows the result of measurement of the temperature distribution of the substrate 20 in the steady state in the LED lighting module 1 of the present example by thermography. FIG. 8 (a) is a diagram in which the temperature distribution of the substrate 20 is displayed on a contour, and FIG. 8 (b) is a plot of the temperature distribution of two paths A and B in FIG. : Path B). As can be seen from FIG. 8B, there is no difference in temperature distribution between the paths A and B passing through the center of the substrate 20, and the path used for design study may be either path A or B. Therefore, for convenience, the path B is used as a path used for design study, and the substrate maximum temperature T _max is defined as the maximum value of the temperature of the substrate 20 in the path B.
The result (dotted line) which asked for substrate temperature distribution T of LED lighting module 1 of this example by a thermal analysis program is shown in Drawing 9 with the measurement result (solid line) of a real machine. In the vicinity of coordinates 0 ≦ r ^* <0.1, although the actual machine has a constant interval in the vicinity of the line A-A in FIG. 5, since the analysis program approximates uniform heat generation, the temperature distribution of the actual machine is reproduced Not done. However, the average error of the analysis result with respect to the measurement result of the actual machine is 2.6%, and it can be seen that the reproduction is accurately performed in other regions.

  Example 2 shows a simulation result in the case where the substrate 20 of the LED lighting module 1 is configured by FGHP. In the following, differences from the first embodiment will be mainly described, and description of the common configuration may be omitted.

[Construction]
The substrate 20 is the FGHP of Patent Document 1 with a diameter of 120 mm and a thickness of 2 mm. Since a heat pipe is provided inside the FGHP and high thermal conductivity can be obtained, heat diffusion is apt to occur, aiming to reduce temperature unevenness.

[Analysis result]
Table 3 shows each design parameter of the present embodiment input to the thermal design program. Since the unknown parameter is the substrate temperature distribution T, the dependent substrate maximum temperature T _max and the temperature unevenness T _max −T _min are not inputted into the program as the unknown parameter.
The result (dotted line) which asked for substrate temperature distribution T of LED lighting module 1 of this example by a thermal analysis program is shown in Drawing 9 with the measurement result (solid line) of a real machine. Although the real machine has constant intervals in the vicinity of the line A-A in FIG. 5 at coordinates 0 ≦ r ^* <0.12, the analysis program approximates uniform heat generation, so the temperature distribution of the real machine can be reproduced. Not. However, the average error of the analysis result with respect to the measurement result of the actual machine is 6.3%, and it can be seen that the reproduction is accurately performed in other regions.

When the analysis result of the temperature non-uniformity is compared between Example 1 and Example 2, it was 2 K in Example 2 with respect to 16 K in Example 1. It can be seen that the temperature unevenness is reduced to 1/8 that of Example 1 compared to Example 1. Since the substantial difference between the design parameters of the first embodiment and the second embodiment is the thermal conductivity k _r in the substrate surface direction, it is presumed that the temperature unevenness is affected by the thermal conductivity k _r in the substrate surface direction. .
Therefore, in the third embodiment, the substrate effective thermal conductivity k _e is used instead of the substrate surface direction thermal conductivity k _r and the substrate thickness direction thermal conductivity k _z for simplification, and heat radiation is similarly affected. Using the overall heat transfer coefficient (substrate backside cooling strength) h, these design parameters were changed to investigate the influence on temperature unevenness. Specifically, using the underwater high-brightness LED lighting device 60 shown in FIG. 10A, thermal analysis simulation was performed while changing the shape and the physical property values to analyze temperature unevenness.

[Construction]
FIG. 10 shows (b) a plan view and (c) a cross-sectional view of the underwater high-intensity LED lighting device 60. The underwater high-brightness LED lighting device 60 mainly includes a housing 61, a substrate 62, a lens 63, and two O-rings 64.
The housing 61 is a substantially cylindrical shape having a diameter of 111 mm and a height of 71 mm made of brass, and is configured to include a heat receiving portion 61a, a side surface portion 61b, a plurality of fins 61c, a cap portion 61d, and a socket portion 61e. The substrate 62 is fixed to the surface side of the heat receiving portion 61 a, and receives heat from the substrate 62. The side surface portion 61b is provided with a plurality of fins 61c on the outer periphery thereof, and conducts the heat received by the heat receiving portion 61a to the fins 61c, and the fins 61c dissipate the heat into water. The cap 61 d fixes the lens 63 to the housing 61. The socket 61e guides a power cable (not shown) into the housing 61.
The substrate 62 mounts a large number of high density mounted LED chips (not shown). The number of LED chips mounted on the substrate 62 is, for example, 172. The substrate 62 conducts the heat generated from the large number of LED chips to the heat receiving portion 62a.
The lens 63 condenses and transmits light emitted from the LED chip mounted on the substrate 62.
The O-ring 64 is sandwiched between the side surface 61 b and the lens 63 and between the side surface 61 b and the socket 61 e to prevent water from entering.

The detailed structure of the heat receiving part 61a and the board | substrate 62 is demonstrated using FIG. FIG. 11 shows (a) top view, (b) side view, and (c) bottom view of the heat receiving portion 61 a and the substrate 62.
The heat receiving portion 61a has a disk shape having a radius D _o and a thickness t. The open hole 65 is formed, and the contact area with the substrate 62 fixed to the surface of the heat receiving portion 61a is changed by the presence or absence of the open hole 65 and the change of the hole diameter (radius) D _i. It is possible to change. Further, by changing the thickness t, the connection cross-sectional area with the side surface portion 61b changes, and it becomes possible to change the overall heat transfer coefficient h. That is, the radius D _o of the heat receiving portion 61 a, the hole diameter (radius) D _i , and the thickness t can be used as a substitute for the overall heat transfer coefficient h. Incidentally, as the hole diameter (radius) D _i increases, the contact area between the heat receiving portion 61 a and the substrate 62 decreases, and the overall heat transfer coefficient h decreases. Further, as the thickness t increases, the cross-sectional area of connection with the side surface portion 61b increases, and the overall heat transfer coefficient h increases.

The substrate 62 is a disk-shaped copper plate or FGHP of radius L _hs and thickness d. On the surface of the substrate 62, a plurality of LED chips are mounted at a high density, and a mounting area 66 which can be approximated to one which generates heat uniformly is formed. The mounting area 66 has a circular shape of diameter R. Heat mounting region 66 is emitted is incident heat in the contact portion between the mounting region 66 of the substrate 62 as a heat input Q _in. Assuming that the power consumption per chip is 3.2 V × 0.32 A = 1.12 W, for example, when the number of LED chips forming the mounting area 66 is 172, the heat input Q _in is Assuming that 80% is heat, it is 1.12W × 172 × 0.8 = 154W.

Hereinafter will be described the results of heat were carried out by varying the overall heat transfer coefficient h and substrate effective thermal conductivity k _e analysis simulation. Among the design parameters appearing in the configuration of the present embodiment, the design parameters not referred to are as described above, and the description will be omitted.
Design parameters of the analysis model are shown in Tables 4 to 6.
As shown in Table 4, in Examples 3-1 to 3-7, the substrate 62 as a copper plate, changing the overall heat transfer coefficient h by changing the hole diameter (radius) _{D i} and the thickness t of the heat receiving portion 61a It is a model. That is, this is a model in which the heat transfer coefficient h is changed while the substrate effective thermal conductivity k _e is constant (387.6 W / (m · k)). The thickness t was set so that the volume of the heat receiving portion 61 a was constant in order to make the heat capacity of the heat receiving portion 61 a constant.

As shown in Table 5, in Example 3-8～3-14, the substrate 62 as FGHP, changing the overall heat transfer coefficient h by changing the hole diameter (radius) _{D i} and the thickness t of the heat receiving portion 61a It is a model. That is, this is a model in which the heat transfer coefficient h is changed while the substrate effective thermal conductivity k _e is constant (1800 W / (m · k)). The hole diameter (radius) D _i and thickness t of the heat receiving portion 61 a were set to be the same as in Examples 3-1 to 3-7.

As shown in Table 6, in Examples 3-15 to 3-18, the hole diameter (radius) D _i and thickness t of the heat receiving portion 61a are the same as in Examples 3-7 and 3-14. It is a model in which the substrate effective thermal conductivity k _e is changed. That is, it is a model in which the substrate effective thermal conductivity k _e is changed while keeping the heat transfer coefficient h constant.

The analysis results of Examples 3-1 to 18 are shown in FIG. The vertical axis of the graph indicates temperature unevenness, and the horizontal axis indicates biot number Bi. The Biot number Bi is determined by the thermal resistance R _{th (all)} and the heat transfer coefficient h calculated using the center temperature T ₆₆ of the mounting region 66 determined by the thermal analysis simulation according to Equations 11-13. Here, the cooling body area A, the contact area of the substrate 42 and the heat receiving portion 41a is changed by diameter (radius) D _i of the heat receiving portion 41a, it is calculated using the heat radiation area of the common fins 43.
R _{th (all)} : Thermal resistance [K / W]
T ₆₆ : Center temperature of mounting area 66 obtained by thermal analysis simulation [K]
T _:: Environmental temperature [K]
Q _in : Heat input [W]
h: General heat transfer coefficient (fin 61c heat radiation area standard) [W / (m ² · K)]
A: Area to be cooled (heat radiation area of fins 61c) [m ² ]
Bi: Biot number R: Size of cooled object [m]
k _e : Effective thermal conductivity of substrate [W / (m · K)]

As can be seen from FIG. 12, in each of the examples 3-1 to 3-18, a tendency is observed along one approximate curve shown by a dotted line in FIG. That is, it is possible to uniquely determine the temperature unevenness by the bionumber Bi. Therefore, if the specification of the temperature unevenness is determined, it is possible to determine the Bi of the heat dissipation system by following the approximate curve as shown by the arrows (1) to (2) in FIG.
The Biot number is a value determined by the overall heat transfer coefficient, the size, and the substrate effective thermal conductivity, as in Equation 13. According to the above concept, when the configuration of the casing is determined first in the design of the heat dissipation system, the overall heat transfer coefficient is first determined by the analytical solution or numerical solution of the steady heat conduction equation for the casing, and as a result thereof The effective thermal conductivity required of the substrate can be determined. Conversely, if the configuration of the substrate is determined first, then the effective thermal conductivity of the substrate can be determined first, and as a result, the overall heat transfer coefficient required for the housing can be determined.

  According to the third embodiment described above, the design parameters of the heat dissipation system can be determined quickly.

DESCRIPTION OF SYMBOLS 1 LED illumination module 10 Mounting area 11 LED chip 12, 12a, 12b Wiring 13, 13a, 13b External electrode terminal 14 Wire 15 Protection diode 20 Substrate 20B Back surface 21 Inorganic white insulating layer 22 Dam material 23 Translucent resin 24 Inorganic system White insulating layer 30 to 33 Liquid cooler 30 Metal block 30A Surface 31 Circulation flow path 32 Inlet 33 Exit 40 Input / output screen 41 Design parameter name column 42 Design parameter input / output column 43 Setting column 44 Calculation result display column 45 Calculation execution button 60 Underwater high-brightness LED lighting device 61 housing 61a heat receiving portion 61b side surface portion 61c fin 61d cap portion 61e socket portion 62 substrate 63 lens 64 O ring 65 opening hole 66 mounting area

Claims (5)

In a method of designing an electronic component mounting substrate including a substrate and a large number of semiconductor elements integrated and mounted on the substrate,
Design parameters include substrate temperature distribution, substrate surface thermal conductivity, substrate thickness thermal conductivity, substrate size, substrate thickness, cooled body size, heat input, and overall heat transfer coefficient,
Input values to design parameters excluding one unknown parameter selected from the design parameters,
The bionumber Bi _{r of the} substrate, the calculation parameter a, and the dimensionless coordinate r ^* calculated by the following equation 1 are calculated, and the following equation 2 is used without the calculation parameter a and the dimensionless coordinate r ^*: The dimensional temperature 算出 is calculated, and the temperature T _R at the dimensionless coordinate r ^* = 1 is calculated by the following equation 3 including S _mod calculated from known data, and the temperature distribution T [K] of the substrate by the following equation 4 A heat dissipation design method of an electronic component mounting board characterized by calculating.
[Equation 1]
[Formula 2]
Here, each coefficient is as shown below.
h: Overall heat transfer coefficient [W / (m ² · K)]
k _r : Thermal conductivity in the substrate surface direction [W / (m · K)]
R: Size of cooled object [m]
d: Substrate thickness [m]
I ₀ : Modified Bessel function of the first kind
[Equation 3]
T _R : Temperature at end of object to be cooled [K]
k _z : thermal conductivity in the substrate thickness direction [W / (m · K)]
A: Cooled body area [m ² ]
h: Substrate backside cooling strength [W / (m ² · K)]
S _mod : shape factor [-]
[Equation 4]
T _:: Environmental temperature [K]
Q _in : Heat input [W]
A: Cooled body area [m ² ]
h: Substrate backside cooling strength [W / (m ² · K)] In a method of designing an electronic component mounting substrate including a substrate and a large number of semiconductor elements integrated and mounted on the substrate,
The first model in which the overall heat transfer coefficient h [W / (m ² · K)] is changed while the substrate effective thermal conductivity k _e [W / (m · K)] remains constant , or the overall heat transfer coefficient h ^{W / (m 2 · K)} ] for the second model changed remains constant substrate effective thermal conductivity _{k e [W / (m ·} K)], effective thermal conductivity _{k e [W / (m ·} K ) Or the range of the overall heat transfer coefficient h [W / (m ² · K)] of the case,
An approximate curve of temperature unevenness of the substrate with respect to the bionumber is calculated based on the first model or the second model ,
The range of the bionumber is calculated from the aim curve of the approximate curve and the temperature unevenness of the substrate,
Calculated effective thermal conductivity of the substrate by the following equation 5 based on the range of Biot number _{k e [W / (m ·} K)] of the range or the housing of the overall heat transfer coefficient ^{h [W / (m 2 ·} K)] A method of heat dissipation design of an electronic component mounting board, characterized in that the range of
[ Equation 5 ]
Here, each coefficient is as shown below.
h: Overall heat transfer coefficient [W / (m ² · K)]
k _e : Effective thermal conductivity of substrate [W / (m · K)]
R: Size of cooled object [m]   The method for designing heat dissipation of an electronic component mounting board according to claim 1, wherein the design parameter further includes a maximum substrate temperature, a substrate temperature unevenness, an environmental temperature, and an area to be cooled.   An LED light-emitting module, wherein the electronic component mounting substrate comprises dozens or more of LED chips integratedly mounted on the substrate, a translucent resin portion covering the LED chips, and a wiring layer formed on the substrate The method for designing heat dissipation of an electronic component mounting board according to any one of claims 1 to 3, wherein the method is an LED light emitting module including a housing on which the light emitting module is mounted.   A computer program that causes a computer to execute the method for designing heat dissipation of an electronic component mounting board according to any one of claims 1 to 4.