Thermal Simulation Analysis of High-Power DC-DC Converters for Spacecraft

introduction

With the rapid development of electronic technology, the power density of electronic equipment is constantly increasing. The high temperature caused by high power density will have a serious impact on most electronic components, which will cause the failure of electronic components and then cause the failure of the entire equipment. Therefore, the thermal design of electronic equipment plays an increasingly important role in the design of the entire product, and traditional thermal design methods have been difficult to adapt to the needs of development. In order to reduce design costs, improve the first-time success rate of products, and improve the performance of electronic products, thermal simulation technology is increasingly commonly used in the thermal analysis process of electronic equipment. With the help of thermal simulation, designers can reduce the cost of design, production, redesign and reproduction, simulate the boundary conditions in special working environments, and shorten the development cycle of high-performance and high-reliability electronic equipment.

1 Thermal design requirements for spacecraft high-power DC-DC converters

The DC-DC converter is an important device that converts the primary bus voltage of solar or nuclear energy into the secondary bus voltage or the voltage required by various electronic equipment in the spacecraft during ground testing and on-orbit operation, and stably and reliably supplies the corresponding working current of various electrical equipment and payloads in the spacecraft. With the rapid development of my country's space industry, especially the development of high-orbit, large-capacity, long-life satellites, manned spacecraft and space station related technologies, the power supply required for spacecraft has gradually increased. High-power DC/DC power supplies will play an increasingly important role, and their thermal design is directly related to the reliable operation of the entire system. The high-power DC-DC converter of spacecraft has the characteristics of poor heat dissipation conditions and high heat consumption, concentrated heat generation, and uneven heat consumption distribution. Due to the particularity of heat dissipation of space electronic products, there are more special requirements for the heat dissipation method of the power supply.

The power MOSFET tube, diode, and high-frequency transformer in the spacecraft high-power DC-DC converter are the main heat-generating devices. Excessive temperature will deteriorate the characteristics of power electronic devices, make them unstable, or even damage them. When the temperature exceeds the Curie temperature, the magnetic state of the magnetic core changes from ferromagnetism to paramagnetism, damaging the high-frequency transformer and thus the DC-DC converter. The purpose of thermal design of spacecraft high-power DC-DC converters is to control the temperature of all electronic components inside electronic equipment in a space environment without convection heat transfer, so that they do not exceed the specified maximum allowable temperature under the working environment conditions of the equipment.

2 Several ways to obtain temperature parameters

The primary issue in thermal design of electronic equipment is the acquisition of temperature parameters. The acquisition of temperature parameters can be divided into two categories: contact and non-contact.

2.1 Contact-type temperature parameter acquisition method

The principle of contact temperature parameter acquisition is simple and the measurement accuracy is high; however, because the temperature measuring element and the measured medium need a certain amount of time to fully exchange heat and reach the thermal balance required for the test, there is a delay in temperature measurement. At the same time, the temperature measuring element will inevitably absorb some heat from the device. When the temperature measuring element is powered on for measurement, it will generate some heat itself, which will have a certain impact on the test results.

The following temperature measuring elements are commonly used in the contact-type temperature parameter acquisition method:

a. Thermistor: NTC thermistor has the characteristics of small size, high test accuracy, fast response speed, stability and reliability.

b. Thermocouple: Thermocouple is one of the most commonly used temperature detection elements in industry. Its advantages are: wide measurement range. Commonly used thermocouples can continuously measure from -50 to +1600℃; simple structure and easy to use.

c. Thermal resistor: Thermal resistor is the most commonly used temperature detector in the medium and low temperature range. Its main features are high measurement accuracy and stable performance. Among them, platinum resistor has the highest measurement accuracy. It is not widely used in industrial temperature measurement, but is made into various standard thermometers (covering national and world benchmark temperatures) for measurement and calibration.

The commonly used platinum resistance temperature sensor has a zero resistance of 100Ω and a resistance change rate of 0.3851Ω/℃. Platinum resistance sensors have good long-term stability, and typical experimental data are: at 400℃ for 300 hours, the maximum temperature drift at 0℃ is 0.02℃.

According to the IEC751 international standard, the temperature coefficient TCR = 0.003851, Pt100 (R0 = 100Ω), Pt1000 (R0 = 1000Ω) are uniformly designed platinum resistors. The test current of conventional products is 1mA for Pt100 and 0.5mA for Pt1000. In actual application, the test current should not exceed the allowable value. Temperature coefficient TCR = (R100-R0) / (R0 × 100), where:

Temperature/resistance characteristics:

2.2 Non-contact method for obtaining temperature parameters

The non-contact temperature parameter acquisition methods mainly include numerical calculation method, infrared camera method, etc. Among them, the numerical calculation method mainly relies on the classic junction temperature formula: Tj=TA+PDθJA (that is, the device junction temperature Tj is equal to the ambient temperature TA plus the product of the device power consumption PD and the device thermal resistance θJA) to calculate the device junction temperature; or use the junction voltage to change with temperature by about -1mV/℃ to -2mV/℃ after applying a constant current source to the PN junction to estimate the device junction temperature.

Infrared photography is a method of using an infrared camera to take infrared photos of an object (it can be a photo of a certain moment or a continuous image over a period of time), and analyze the photos to convert the infrared radiation emitted by each part of the target into an optical signal visible to the naked eye, thereby obtaining a non-contact temperature parameter acquisition method for the surface temperature distribution of the object. The temperature is measured based on the principle of thermal radiation. The measuring element does not need to contact the measured medium, will not destroy the temperature field of the measured object, and the reaction speed is generally faster; however, it is greatly affected by external factors such as the surface emissivity of the object, the measurement distance, and the space environment.

Figure 2.2.1 shows the thermal imaging data of a ground test circuit board of a spacecraft high-power DC-DC converter in the atmosphere, collected by an IR913A infrared thermal imager with a temperature measurement range of -20℃ to +400℃.

The infrared camera method can better obtain the temperature of the photographic part of the electronic device. The imaged temperature cloud map is more intuitive than the results of other temperature parameter acquisition methods. However, its disadvantage is that the test results are greatly affected by the working environment conditions, and it is difficult to collect temperature parameters in places that cannot be captured by the infrared lens.

Since various temperature acquisition methods have their own advantages and disadvantages, various methods are usually used alternately or simultaneously throughout the entire development process of electronic products to achieve the purpose of obtaining detailed temperature parameters for feedback design.

From the perspective of economy and development cycle, the application of only measured temperature parameter feedback design is increasingly unable to meet the development and production needs of products. At present, the general mode of thermal design of electronic products has changed to the introduction of thermal analysis software for simulation analysis in the early stage of electronic product development to assist in design, and the use of actual measurement and simulation to verify during the development cycle, so as to feedback the design more quickly and effectively and improve the thermal design of the product.

3 Thermal Simulation Analysis of Spacecraft High-Power DC-DC Converter

Since high-power DC-DC converters for spacecraft are high-power density electronic products working in the space environment, their thermal design is particularly important in the overall product reliability design. The simulation of the space thermal environment requires a lot of money and a long period of time, so the application can reduce the test costs, simulate the boundary conditions in the special working environment, and shorten the development cycle. The participation rate of thermal simulation in the thermal design process has been greatly increased. This article will introduce the use of professional electronic product thermal analysis software to perform thermal simulation analysis on high-power DC-DC converters for spacecraft used in the space environment, so as to obtain the simulation data of the vacuum thermal stress environment and the heat dissipation and temperature distribution of the product under vacuum thermal stress.

3.1 Thermal simulation software

At present, many foreign companies have developed a wide variety of electronic equipment heat dissipation design auxiliary analysis software based on computational heat transfer technology (NTS) and computational fluid dynamics technology (CFD), including Flotherm, Ice-pack, I-deas based on the finite volume method, and Ansys based on the finite element method. Among them, Flotherm and Ice-pack occupy most of the market share.

Icepak software from Fluent, an American company, is a powerful CAE electronic equipment heat dissipation professional analysis software tool jointly developed by Fluent and ICEM-CFD. It can simulate the heat transfer, flow, and radiation of electronic products, thereby performing simulation analysis and feedback design to improve product quality. Icepak uses the Fluent computational fluid dynamics (CFD) solver engine. The solver can complete flexible meshing and can use unstructured meshes to solve complex geometric problems. The multi-point discrete solution algorithm can speed up the solution time. It can help designers monitor data at locations that cannot be measured. The entire software uses a unified integrated environment interface. Users can apply the software to actual design analysis in a relatively short time.

3.2 Heat consumption calculation and heat consumption distribution

The heat loss of the power conversion circuit of the spacecraft high-power DC-DC converter is mainly borne by the power MOSFET tube and the transformer, the heat loss of the control circuit is mainly generated by the chip, and the heat loss of the output rectifier circuit is mainly borne by the output rectifier diode.

Theoretically, the heat dissipation of electronic components can be obtained by measuring the current and voltage, but it is difficult to operate in practice, especially when measuring the current value in a complex circuit. The usual solution is to simulate the power consumption through some circuit simulation software, such as Pspice or saber, but the power consumption is a function of temperature. At present, most circuit simulation software still does not take temperature into consideration sufficiently, and not all power consumption is converted into heat consumption. Magnetic loss and electromagnetic radiation loss cannot be ignored in heat consumption calculation. The heat consumption calculation value and heat consumption distribution obtained through analysis and simulation by designers largely determine the credibility of thermal simulation analysis data.

3.3 Determination of boundary conditions and selection of thermal parameters

There are three ways of heat transfer: radiation, convection and conduction. In space applications, there is basically no such form of heat transfer as convection, and only heat conduction and radiation are examined. The bottom plate of the spacecraft high-power DC-DC converter product is closely attached to a heat sink with a temperature of 50°C, and the temperature is constant at 50°C. The power consumption of the heating element is added to the element model or the heat source used to simulate the chip, and the surrounding environment is a vacuum.

The selection of thermal parameters used in thermal simulation analysis mainly refers to the selection of thermal conductivity λ used to calculate thermal resistance.

The selection of material thermal conductivity for thermal simulation analysis of spacecraft high-power DC-DC converter products is shown in Table 3.3.

When doing thermal simulation, the equivalent thermal conductivity λeq is used to represent the thermal conductivity of the PCB board and components.

The equivalent thermal conductivity λeq of the PCB board is calculated based on the mass fraction and volume fraction of each part of the PCB board. PCB boards are generally made of insulators (such as FR4) and copper through heating and pressurization. The function of copper is to conduct electricity and heat. The thermal conductivity of FR4 is generally 0.35W/(m?K), and the thermal conductivity of copper is 385.1W/(m?K), so the copper content is an important factor affecting thermal conductivity. The cross-sectional structure of a multi-layer PCB board is shown in Figure 3.3.

The selection of thermal parameters of equivalent thermal conductivity is calculated according to formula (1):

The conductor residual rate of layer i is: for the copper foil layer, it is the residual rate of the copper foil; for the insulation layer, its residual rate is approximately 1.

The equivalent thermal conductivity λeq of the component is composed of the thermal conductivity of the packaging material, the pin material, the mounting material, etc., and is calculated using the equivalent thermal resistance formula. The heat transfer from the component node to the printed circuit board is regarded as one-dimensional heat conduction. According to the different installation methods of the component, different types of electrothermal simulation thermal circuit diagrams can be established, and the equivalent thermal conductivity can be calculated using the equivalent thermal resistance formula (2).

Among them: δ is the equivalent thickness along the heat conduction direction; S is the equivalent heat conduction area perpendicular to the heat transfer path; Rtot is the total thermal resistance of the component's electrothermal simulation thermal circuit diagram.

3.4 Thermal simulation modeling

Establishing a reasonable thermal simulation model is a prerequisite for ensuring the accuracy of thermal simulation analysis results.

For the main heat dissipation devices, power MOSFET tubes and rectifier tubes, they are installed on the power aluminum substrate, and SMD-1 packaging is used. The packaging form is shown in Figure 3.4.1. The installation method adopted is to weld the power MOSFET tube on the aluminum substrate and closely contact the bottom surface of the product aluminum shell through thermal conductive silicone grease. The bottom surface of the aluminum shell is in close contact with the temperature control heat sink to achieve conduction heat dissipation. The structure is shown in Figure 3.4.2.

A computational physics model is established for high-power DC-DC converter products of spacecraft. Considering the feasibility of computational mesh division and thermal conduction and thermal radiation analysis and calculation, the model is simplified to a certain extent. The thermal conductivity of the printed circuit board (PCB board) is calculated according to the equivalent thermal conductivity; the wires with little thermal influence are ignored; the surfaces of each structure are gray bodies, the emissivity and absorptivity are independent of the wavelength, and the emissivity (ε) = absorptivity (α); the surfaces of each structure are diffuse reflection surfaces, and the reflectivity is independent of the direction of injection/emission; the surfaces of each structure are opaque to thermal radiation, and the transmittance can be ignored.

The thermal simulation model of the spacecraft high-power DC-DC converter product is composed of a plate (PLATE), a cylinder (PRISM, etc.), a printed circuit board (PCB), a face (FACE), a cabinet (CABINET), a block (BLOCK), a source (SOURCE), etc. It is mainly a plate structure (PLATE) and a block (BLOCK) structure.

The simplified computational physics model is shown in Figures 3.4.3, 3.4.4 and 3.4.5.

3.5 Thermal simulation calculation method

Icepak is a professional electronic equipment thermal analysis software that can solve thermal analysis problems at the system level, component level, and package level. It uses unstructured grids and can generate three-dimensional tetrahedral and hexahedral unstructured grids for complex geometric shapes. The solution uses the finite volume method and the Fluent solver to ensure the calculation accuracy of engineering problems. Icepak software solves three control equations: the mass conservation equation, the momentum conservation equation, and the energy conservation equation. Since the heat transfer mode in the space environment is mainly heat conduction and heat radiation, and the convection mode is not considered, only the temperature field is calculated instead of the flow field, and only the convergence of the energy equation is examined.

In the phenomenon of heat conduction, the amount of heat passing through a given cross section per unit time is proportional to the temperature change rate and cross-sectional area in the direction perpendicular to the cross section, while the direction of heat transfer is opposite to the direction of temperature increase. This is the basic law of heat conduction, and its mathematical expression is:

Where: φ refers to the heat transferred through a unit area per unit time, and x is the coordinate axis perpendicular to area A.

?t/?x is the rate of change of the object's temperature along the x direction, where the negative sign indicates that the direction of heat transfer is in the direction of temperature decrease.

In a vacuum, the radiation capacity of an object is determined by the material properties of the object, surface conditions (such as color, roughness, etc.), surface area, and surface temperature. The darker the color of the surface, the rougher it is, the higher the temperature, and the stronger the radiation capacity. The radiation studied in Icepak is face-to-face radiation. The radiation heat transfer from surface 1 (temperature T1) to surface 2 (temperature T2) is given by the following formula:

3.6 Thermal simulation calculation

The grid type of the spacecraft high-power DC-DC converter is unstructured hexahedral grid. The grid of the calculation physical model of the spacecraft high-power DC-DC converter is shown in Figure 3.6.1.1 and Figure 3.6.1.2.

The residual error of iterative solution of energy equation by Icepak software is shown in Figure 3.6.1.3. The temperature cloud diagram of thermal simulation is shown in Figure 3.6.1.4, Figure 3.6.1.5, Figure 3.6.1.6 and Figure 3.6.1.7.

Based on the results of thermal simulation, the simulation data of the maximum values ​​of the junction temperature, shell temperature or hot spot temperature of the main heat-generating components can be obtained. Among them, the temperature of low-power components is approximately the maximum value of the board temperature near the component.

4 Summary of Thermal Simulation Process of Spacecraft High-Power DC-DC Converter

The powerful thermal analysis function of Icepak software can greatly improve the thermal design of electronic products. The results of thermal simulation need to be verified and compared with the measured temperature obtained in the simulated space environment to improve the true approximation of the product's heat dissipation, feedback design, and improve product reliability. The effective application of thermal simulation technology in thermal analysis avoids the repeated production of expensive actual prototypes due to multiple possible design changes, saves the cost of simulated thermal tests, compresses the design process, and advances the product delivery date.

But it is worth noting that any advanced simulation software can never replace people. Software is just one of the tools used by thermal designers. The accuracy of simulation software results depends largely on the experience and theoretical level of the designers.