REFINED THE ACCURACY OF ELECTRONIC COMPONENT TEMPERATURE MODELING IN COMPUTER-AIDED PROCESS PLANNING

ABSTRACT

The main goal of this work is to obtain the most accurate method for calculating the detailed model in terms of thermal effects.

As well as identifying the advantages of each method. To achieve this goal, it is necessary to study each method, perform calculations under the same conditions, and finally compare the results of all models.

We took a model of a power diode with a printed circuit board and the calculations were performed under the same boundary conditions and the effects of constant thermal loads on the model are used to determine the temperatures, heat flow rates and heat fluxes in a part. Steady-state analysis is often a precursor to transient thermal analysis to determine initial conditions.

АННОТАЦИЯ

Основная цель этой работы - получить наиболее точный метод расчета детальной модели с точки зрения тепловых эффектов.

А также определить преимущества каждого метода. Для достижения этой цели необходимо изучить каждый метод, выполнить расчеты в одних и тех же условиях и, наконец, сравнить результаты всех моделей.

Мы взяли модель силового диода с печатной платой, и расчеты были выполнены при тех же граничных условиях, а влияние постоянных тепловых нагрузок на модель используется для определения температур, скоростей тепловых потоков и теплопритоков в детали. Стационарный анализ часто предшествует переходному термическому анализу для определения начальных условий.

Keywords: Convection; product design; automation system for project poaching ; thermal analysis.

Ключевые слова: Конвекция; дизайн продукта; система автоматизации проектирования ; термический анализ.

 Introduction

A detailed model is a model that represents or reconstructs the physical geometry of a component to the extent possible. Thus, the physical model will always look similar to the actual component model [1]. A correctly constructed detailed model, by definition, does not depend on boundary conditions, i.e. The model will accurately predict the temperature of the various elements inside the package (including connections, housing and leads) regardless of the computing environment in which it is located. The industry standard thermal design Creo Ansys Simulation allows you to simulate thermal conditions

A steady state thermal analysis calculates effects of constant thermal loads on a model and is used to determine temperatures, heat flow rates, and the heat fluxes in a part. A steady state analysis is often a precursor to a transient thermal analysis to determine the initial conditions[10] .

As an example, we use a model of a power diode with a printed circuit board section for calculations. The power diode (Schottky barrier rectifier) is a good example of an electronic component that is first cooled through the board, which is ensured by the design of his body:

(The Double Decawatt Package, D2PAK, SOT404 or DDPAK, standardized as TO-263, is a semiconductor package type intended for surface mounting on circuit board) D2PAK or TO263.  In correct assessment of diode heating can lead to underestimation of its heating and resulting in an unstable device, This leads to overestimation and the need to build a redundant cooling system, i.e. to an increase in the device's complexity, its weight, gas, development time and cost.

Research Methodology

We built a detailed model based on experiments on natural convection in two stages.

First, a printed circuit board is built, and then control is carried out at the level of the body of the component being used. The proven detailed model is then used to create compact thermal models. In this study, we compared the results of all other modeling approaches with this detailed model [2]. Modeling using 2R models is also an approach for solving the problem of computing systems used in industry.The 2R model requires only two cells to allow heat transfer within the component. The resistor values depend on the air flow and board conductivity. It is also necessary to develop a single empirical model covering all boundary conditions [3]. The advantage of the DELHPI model is that it accurately records heat flows and temperatures using nine nod (grid cells).

The DELPHI-based model was developed to show that a complex component with multiple heat sources can be represented by networked models, if its surface satisfies the condition [4].To compare the listed models with each other, the following boundary conditions were used (ambient temperature To = 20 °C):

– natural convection with radiators and 6 of them;

– forced convection at different air flow velocities.

In this study, body temperature is the only quantity that is analyzed for various models. The temperature of the casing is measured at the top part specified by the manufacturer. Heat flows through the sides are not analyzed, because they do not present problems until the temperature of the case matches the results of the detailed model.We show the increase in the component's temperature body depending on the ambient temperature in the table. 1 for the case of natural convection without a radiator.Here, the 2R model was simulated with Rjc = 50 °C/W and Rjb = 0.1 °C/W. As a result, we find out that the model based on DELPHI provides the best results, with errors of less than 2%.The body temperature predicted by the lumped element model gives results of less than 10%.For preliminary analysis, a model with lumped data works well, while a DELHPI-based model can be used when a high level of accuracy is required.

Table 1.

 Comparison of component housing temperatures for natural convection without radiator

We show the increase in the component's temperature body depending on the ambient temperature in the table. 2 for the case of natural convection with the radiator.

Here, a typical radiator is used. Using a radiator under conditions of natural convection is a common method of temperature control when the body temperature exceeds specified limits. The results show that the DELPHI-based model and the lumped element model produce results with errors of less than 5%.

The case temperature predicted by the 2R model with Rjc = 50 °C/W and Rjb = 0.1 °C/W has a high level of error, showing that determination of these resistances is necessary, let's check it out.

Table 2.

Comparison of component housing temperatures for natural convection with radiator

The model should only be used when the temperature difference across the body is less than 1 °C. The increase in the temperature of the component body depending on the ambient temperature is shown in the table. 3 for forced convection of air flow at a speed of 2 m/s.  The 2R model used in this case has resistance values Rjc = 59.2 °C/W and Rjb = 9.77 °C/W.

Table 3.

 Comparison of component body temperatures for forced convection of air flow at a speed of 2 m/s

It can be seen that the DELPHI model still has the smallest error value, which is less than 4%. Although the lumped data model produces good results, removing the mesh used to resolve gradients may prohibit its use when the system is present There are a large number of components. But the error is highest in the 2R model, which still predicts the case temperature with approximately 95% accuracy [8].

Result and discussion

Detailed modeling consumes a huge amount of computing resources and time. To avoid this, simplified models usually represent components. We can present these simplified models as a lumped data model, a 2R model, or a DELHPI model.

1. We analyzed these three models in relation to each other and also in relation to the validated detailed model. The model with lumped data has errors of less than 10% both when modeling natural convection and when modeling forced convection. Although this result is good, we should use it with caution. Only if the component has a gradient of less than 1 °C on its surface should this model be used. Otherwise. We recommend it to use more complex models, such as the 2R model or the DELHPI model.  When the system has many components, the use of a model with concentrated data is prohibited.At this stage, it is advisable to use models based on 2R or DELPHI.

2. 2R models offer good mesh reduction, but in their current state they do not meet the boundary conditions .If the appropriate resistance values based on the air flow pattern are not used, then high heat will occur when the case temperature is predicted schnoct. The advantage of the 2R model is its flexibility. A range of resistances can be used to represent more than one type of component, and the power can vary.

3. The DELHPI model was developed as truly independent of boundary conditions. Although it overcomes the problems of independence from boundary conditions, the level of accuracy can be even higher.

The DELPHI model is not very flexible. It can only be used for the specific component for which it was created. The choice of model ultimately depends on the specific situation. If preliminary analysis is required, we can carry it out using simple models with lumped data. If a high level of accuracy is required and the boundary conditions are known, then the 2R model can be used. If a high level of accuracy is required, and the boundary conditions are unknown, then the DELHPI model can be used.

Reference:

1. G. Eason, B. Noble, and I. N. Sneddon, “On certain integrals of Lipschitz-Hankel type involving products of Bessel functions,” Phil. Trans. Roy. Soc. London, vol. A247, pp. 529–551, April 1955. (references)
2. J. Clerk Maxwell, A Treatise on Electricity and Magnetism, 3rd ed., vol. 2. Oxford: Clarendon, 1892, pp.68–73.
3. I. S. Jacobs and C. P. Bean, “Fine particles, thin films and exchange anisotropy,” in Magnetism, vol. III, G. T. Rado and H. Suhl, Eds. New York: Academic, 1963, pp. 271–350.
4. K. Elissa, “Title of paper if known,” unpublished.
5. R. Nicole, “Title of paper with only first word capitalized,” J. Name Stand. Abbrev., in press.
6. Y. Yorozu, M. Hirano, K. Oka, and Y. Tagawa, “Electron spectroscopy studies on magneto-optical media and plastic substrate interface,” IEEE Transl. J. Magn. Japan, vol. 2, pp. 740–741, August 1987 [Digests 9th Annual Conf. Magnetics Japan, p. 301, 1982].
7. M. Young, The Technical Writer’s Handbook. Mill Valley, CA: University Science, 1989.
8. Khedher N. Ben, Biswas N., Togun H., Mohammed H.I., Mahdi J.M., Ibrahem R.K., Talebizadehsardari P. Geometry modification of a vertical shell-and-tube latent heat thermal energy storage system using a framed structure with different undulated shapes for the phase change material container during the melting process. Journal of Energy Storage. 72 : 108365. 2023.
9. Z. H. Al-Araji, N. A. Swaikat, A. Muratov and A. V. Turetsky, "Modeling and Experimental Research of Vibration N Properties of A Multi-Layer Printed Circuit Board," 2019 4th Scientific International Conference Najaf (SICN), Al-Najef, Iraq, 2019, pp. 43-47, doi: 10.1109/SICN47020.2019.9019364.
10. Al-Obaidi, Abdulkareem Sh Mahdi, et al. "Embedded design and implementation of mobile robot for surveillance applications." Indonesian Journal of Science and Technology 6.2 (2021): 427-440.‏