Validation methodology to analyze the temperature-dependent heat path of a 4-chip LED module using a finite volume simulation
Introduction
The analysis of the thermal properties of an LED module is accompanied by the investigation of the underlying heat transport within the device. The heat produced by the junction of the LED propagates through the module's layers and ends up at the heat sink. In this work this heat transport was investigated with the help of the structure function [1], [2], [3]. The structure function maps this heat flow path in terms of cumulative thermal capacitances (CthΣ) with respect to thermal resistances (RthΣ), starting from heat source to the ambient (see Fig. 1). All essential structural changes in the heat flow path through the LED are reflected by the structure function [4].
One aim of this paper was to understand the measured structure function of a four-chip-LED module. Structure functions were obtained by thermal transient experiments and by finite volume simulations. The correlation between simulated and measured structure functions opened the possibility to assign regions of the function with regions in the module via isothermals. In addition, a simulation in accordance with the measurement, enabled an alternative approach of the electrical transient correction. A further target was the investigation of the thermal behavior of the four-chip module operated by different conditions. Therefore, a set of different operating conditions was systematically varied by means of DOE and its structure functions were evaluated. Although the operating condition should cause no differences in the structure functions under idealistic assumptions [5], a temperature dependence of the experimental structure function was observed. Hence, a finite volume simulation was performed to understand the nature of this temperature-dependence.
Section snippets
Theoretical background
The RthCth-model, which is provided by the structure function, is assumed to be linear, that means all RthCth-elements are considered to be temperature-independent. As shown in [6] the assumption of linearity of the RthCth-model should be treated with care, if the device is driven in a broad temperature range. Due to the fact, that the thermal resistance (Rth) is not only the change of temperature (ΔT) divided by the heating power (PH), but also inverse proportional to the thermal conductivity
Test setup
The investigated device was a blue LED-module consisting of four blue LED flip chips, connected in serial. Therefore, the resulted structure function had to be considered as result of all four diodes.
The device was investigated thermally according to the JEDEC JESD 51-1 standard [8] with the T3ster measurement setup, and optically within an integrating sphere complying the CIE 127-2007 [9]. The LED was measured electrically by the thermal test setup (T3Ster from Mentor Graphics), whose output
Finite volume simulation
Further information about the thermal behavior of an LED module was gained by a thermal path analysis, performed by a finite volume simulation (FloTherm®) [12]. Comparing the experimental and simulated structure functions allowed the verification of thermal models.
Design of experiment
In this study the design of experiment (DOE) approach was used to investigate effects of the device's behavior to a set of variations in its application conditions [21]. The varied parameters of the DOE (see Fig. 3) comprised the temperature of the heat sink (THS), the forward current IF, two different thermal interface materials (TIM) attaching the LED-module to the heat sink, the correction of the electrical transient (Trans-Correction) and the time range of heating and cooling (tmeas) steps.
Conclusion
The combination between measurement and simulation is a powerful tool to get a deeper physical understanding of the structure function. The way of iterative simulation's adjustment was shown in detail to get a validated simulation model. The validated model offers the visualization of the heat path by means of isothermal surfaces, allowing the correlation of physical regions with regions of the structure function. The validation opened an alternative approach for the electrical transient
Acknowledgements
Financial support by the Austrian Federal Government (in particular from Bundesministerium für Verkehr, Innovation und Technologie) represented by Österreichische Forschungsförderungsgesellschaft mbH within the framework of the “7. Ausschreibung Produktion der Zukunft nat. Projekte” Programme (project number: 848574 project name: FlipTheLED) is gratefully acknowledged.
References (24)
- et al.
Development of a standard for transient measurement of junction-to-case thermal resistance
Microelectron. Reliab.
(2012) - et al.
Thermal investigation of high power optical devices by transient testing
IEEE Trans. Components Packag. Technol.
(2005) - et al.
Comparison of JEDEC dynamic and static test methods for the thermal characterization of power LEDs
- et al.
Thermal Management for LED Applications
(2014) - et al.
On the standardization of thermal characterization of LEDs
- et al.
Thermal Conductivity, 28.02.2001
- Integrated Circuits Thermal Measurement Method - Electrical Test Method (Single Semiconductor Device), JEDEC Standard,...
Measurement of Leds CIE127:2007
(2007)
Increasing the accuracy of thermal transient measurements
Creating Multi-port Thermal Network Models of LED Luminaires for Application in System Level Multi-domain Simulation Using Spice-like Solvers
Cited by (10)
A Heat Pipe Cooler for High Power LED's Cooling in Harsh Conditions
2019, Procedia ManufacturingParameter driven monitoring for a flip-chip LED module under power cycling condition
2018, Microelectronics ReliabilityCitation Excerpt :Each Cth,i/Rth,i section can be interpreted as different elements of the LED module in the heat path down to the heat sink. A detailed information about the structure function of this four chip module was given via validated simulation and was already discussed by Mitterhuber et al. in [5, 20]. In this paper the focus was to investigate the changes in Rth of LED modules in terms of reliability issues.
Thermal transient measurement and modelling of a power cycled flip-chip LED module
2018, Microelectronics ReliabilityCitation Excerpt :The validation process included the adjustment of the material properties of the simulation model to the measurement. How to validate the simulation to the measurement was explained in detail in [36]. For the validation of the simulation model, the same heating and cooling time as in the measurements had to be taken into account.
Reliability and failure analysis of solder joints in flip chip LEDs via thermal impedance characterisation
2017, Microelectronics ReliabilityCitation Excerpt :Analysis of the structure function as intake control of each chip showed differences in the total Rth (Fig. 9). The four structure functions varied in the total Rth between 20 and 25 K/W, assuming this variation came from the interconnection technology [13]. Calculations of the solder joint area under the LED chip were carried out with the software FG Studio Max of the radioscopic intake control of the LED-modules.
Design of power cycling for reliability testing of LED systems: Numerical and analytical approach
2023, 2023 29th International Workshop on Thermal Investigations of ICs and Systems, THERMINIC 2023Trigger specific failure in LED system by power and duty cycle patterns lifetime testing
2023, 2023 24th International Conference on Thermal, Mechanical and Multi-Physics Simulation and Experiments in Microelectronics and Microsystems, EuroSimE 2023