Heat Dissipation Capability Research Articles

Effective heat dissipation and higher light extraction efficiency of GaN vertical light emitting diodes for solid state lighting applications

AbstractVertical sapphire‐free GaN based LEDs (VLEDs) on metal have shown much better performance and prove to deliver an efficiency much higher than most conventional light source. In this paper we show that in addition to many advantages of VLEDs over conventional LEDs in terms of better current spreading, low operation voltage, higher drive current density, faster heat dissipation, VLEDs also enables top surface engineering for efficient light extraction to further light output. High power white LED having efficiency of 110 lumen/watt was achieved through a combination of surface engineering and optimization of the n‐GaN layer thickness underneath the corrugated pyramids surface. The low operation junction temperature of VLEDs demonstrates excellent heat dissipation capabilities. (© 2009 WILEY‐VCH Verlag GmbH & Co. KGaA, Weinheim)

Effect of Plated Metal Film on the Performance of Helix TWT Slow-Wave Structure

The effect of the plated metal film on the heat dissipation capability and cold-test characteristics of the helix slow-wave structure (SWS) have been studied in several experimental tests and computer simulations. The thermal conduction and the power losses of the helix SWS were improved by using a plated helix. A brazed SWS with end-plated rods exhibits excellent heat dissipation capability. The side-plated rods affect the cold-test characteristics of the SWS.

Compact and 10-Gb/s $\times$ 12-Channel Optical Transmitter for Optical Interconnection With High Heat-Dissipation Capability Using Alumina Substrates

A compact 10-Gb/s times 12-channel optical transmitter was developed for very short-reach optical interconnections. The 10-Gb/s/ch operations were achieved without employing the heat pipes or heat sinks typically required to keep vertical-cavity surface-emitting lasers at a low temperature under 85degC. Alumina ceramic substrates with high heat dissipation capability are incorporated in order to maintain the low temperature operation. The optical transmitter is designed to be as compact as a standard mechanically transferable (MT) optical connector. To enhance usability, the transmitter can be directly connected with the MT-type optical connector with guide pins assembled in the alumina substrate. The compactness and usability of the optical module is effective in intrabox and interbox optical interconnections.

Study on the thermal interface resistance of the helix slow-wave structure

The thermal interface resistance of the helix slow-wave structure （SWS） has been studied theoretically and experimentally. The effect of the thermal resistance at two interfaces on heat dissipation capability of the SWS is analyzed. A novel method to calculate the thermal interface resistance has been developed, and its accuracy and feasibility were validated by experiments. With this method, the thermal interface resistances at two interfaces can be calculated separately, considering the variation of the material performance with the temperature changes. This novel method is used to study the effect of the rod material and the assembling method on the thermal interface resistance.

High-Performance BOI FinFETs Based on Bulk-Silicon Substrate

A new body-on-insulator (BOI) FinFET device structure based on bulk-Si substrate has been proposed and experimentally demonstrated in this paper. In comparison with other bulk FinFETs, the BOI FinFET features the localized insulator below the Si-Fin body, which can achieve both low source/drain (S/D) parasitic resistance and effective suppression of the S/D leakage beneath the Si-Fin channel, as well as good heat dissipation capability. The device fabrication process is basically compatible with conventional CMOS technology. High drive current, low subthreshold swing, and excellent short-channel behavior are observed in the fabricated BOI FinFETs.

Thermal Management for Boron Nitride Filled Metal Core Printed Circuit Board

This study aims at investigating novel thermal material based on filled epoxy for metal-core printed circuit boards (PCBs). The next generation PCB materials are expected to possess high heat dissipation capability in addition to low coefficient of thermal expansion (CTE) as the accumulated heat from high performance electronic devices should be removed for proper operation. In this study, boron nitride (BN) filler with different size (from micro to nano size) and content were employed to prepare thermally conductive polymer composites. Various percentage of coupling agent was used for the surface treatment of fillers in order to produce homogenous composites with lower CTE. The effect of filler and coupling agent to thermal conductivity of composite are discussed. It indicated that the experimental results fit well with the Bruggeman model with variety filler size. The use of 1% of couple agent was found to be more effective in increasing thermal conductivity of the composite. Despite the maximum content of the BN allowed to be added into the epoxy is about 30% due to the comparatively high viscosity of the varnish, as low as 13% of the micron-sized BN-filled dielectric is enough to fulfil the requirement of thermal conductivity larger than 1 W/m K and balance other critical properties used for PCB application. It is feasible to be a cost-effective advanced thermal material.

Parametric Studies on Dynamic Performance of Hybrid Airfoil Bearing

Airfoil bearings offer many advantages over oil-lubricated bearings, but they have reliability issues during start∕stops (wear) and limited heat dissipation capability. To address these issues, a hybrid airfoil bearing (HAFB) combining hydrodynamic airfoil bearing with hydrostatic lift was introduced previously by one of the authors of this paper. Their studies show that HAFB has superior performance compared to its hydrodynamic counterpart in load capacity and cooling performance. In this article, the bearing stiffness and damping coefficients of HAFB are calculated using a linear perturbation method developed for HAFB. Simulations showed that feed parameter and supply pressure affect the dynamic characteristics of HAFB. With an increase in either the supply pressure or the feed parameter, the rotor centers itself and hence one sees a decrease in direct stiffness. Simulations showed that the cross-coupled stiffness could be reduced by increasing either the supply pressure or the feed parameter. Direct damping showed increasing trend with the supply pressure and the feed parameter. Frequency-domain analysis of the bearing coefficients was also performed. The direct damping showed marginal changes with supply pressure but showed rapid increase with increasing excitation frequencies. The damping converged to null values for all the pressures for supersynchronous excitations. The loss in damping with high stiffness values for high frequency excitation is a typical hardening effect of gas bearings. In almost all the cases, there are rapid decreases in cross-coupled stiffness and damping and the values show converging trends in supersynchronous regime.

Improvement of Heat Dissipation Capability of Slow-Wave Structure Using Two Assembling Methods

Two assembling methods, i.e., the sputter brazing method and the diffusion brazing method, have been developed for improving the heat dissipation capability of the helix traveling-wave tube slow-wave structure (SWS). The magnetron sputter plating technology and the pressure diffusion technology are employed in this letter to complete the SWS assembly. The effect of these two methods on the thermal conduction of the SWS was verified by several experimental tests. These two methods being pursued enable better heat dissipation capability of the helix SWS, compared to the conventional nonbrazing methods.

Thermal Analysis of a Helix TWT Slow-Wave Structure

A novel and effective analytical method using ANSYS has been developed for studying the heat-dissipation capability of a helix traveling-wave-tube slow-wave structure (SWS). This method, which is based on calibrating theoretical calculations with experimental data, is able to precisely predict the SWS heat dissipation, thereby reducing material costs and saving time. The consistency and feasibility of this method have been verified by experimental tests on SWSs using copper-plated helices and both BeO and BN support rods.

Effect of plated helix on heat dissipation capability of the slow-wave circuit

The effect of plated helix on heat dissipation capability of the slow-wave circuit is studied in this paper. With the plated helix, the improvement of the heat dissipation capability of the slow-wave circuit is estimated and evaluated. The influence of helices coated with copper film and gold film is analyzed and verified by computer simulation and experimental tests. The tests show good agreement with the theoretical analysis. The conclusions will make good reference to the design and improvement of the helix TWT.

Efficient flip-chip InGaN micro-pixellated light-emitting diode arrays: promising candidates for micro-displays and colour conversion

Flip-chip InGaN micro-pixellated LED arrays with high pixel density and improved device performance are presented. The devices, with 64 × 64 elements, each of which have a 20 µm emission aperture on a 50 µm pitch, are fabricated with a matrix-addressable scheme at blue (470 nm) and UV (370 nm) wavelengths, respectively. These devices are then flip-chip bonded onto silicon mounts. Good emission uniformity across the LED array is demonstrated, which can be attributed to the introduced n-metal tracks adjacent to each n-GaN mesa and the p-contact lines running across parallel columns. More importantly, with a flip-chip configuration, the optical power output and the current-handling capability of these new devices are substantially enhanced, due to the improved heat dissipation capability and the increased light extraction efficiency. For instance, each pixel in the flip-chip blue (respectively UV) LED arrays can provide a maximum power density 43 W cm−2 (respectively 6.5 W cm−2) at an extremely high current density up to 4000 A cm−2 before breakdown. These flip-chip devices are then combined with a computer-programmable driver circuit interface to produce high-quality micro-scale displays. Other promising applications of these LEDs, such as colour conversion with quantum dots, are also demonstrated.

A Study on the Thermal Fatigue Behavior of Solder Joints Under Power Cycling Conditions

Failure mechanisms exposed by environmental accelerating testing methods such as thermal cycling or thermal shock test, may differ from those at service operating conditions. While the device is heated up or cooled down evenly on its external surface during environmental testing, real operating powered devices experience temperature gradients caused by internal local heating, components' different heat dissipation capability, and ambient temperature variation, etc. In this study, a power cycling technique is introduced to better approximate the field operating conditions so as to activate the field failure modes. Power cycling thermal fatigue test is performed with different ball grid array solder joints, that is, lead contained [Sn/37 Pb (SP)] and lead free [Sn/4.0Ag/0.5 Cu (SAC)], and the result is compared. In order to account for the thermal fatigue life behavior discrepancy for different solder joint composition, real time Moire interferometry is applied to measure the global/local thermo-mechanical behavior during power cycling excursion. Effective damage parameter, the total average shear strain, is extracted from the experiment and applied to account for the difference in fatigue life result of two different solders. In addition, amount of experimentally measured total average shear strain is mutually verified with finite element method analysis. It is clear that total average shear strain of a solder joint can be an effective damage parameter to predict thermo-mechanical fatigue life. A physical mechanism in terms of thermal material property of solder joints' is proposed to offer some thoughts to abnormal shear strain behavior that leads to discrepancies in fatigue life of two solders. An importance of power cycling testing method is emphasized for certain package designs.

Improved local thermal management of AlGaN-based deep-UV light emitting diodes

Local thermal management has a strong impact on the optical output power of AlGaN-based multiple-quantum-well deep-UV flip-chip light emitting diodes (LEDs). During the study, an improved methodology was developed by incorporating a metallic barrier layer of WSiN and a thermal interfacial layer of graphite/copper composite. The performance of these devices with improved thermal management and that of conventional flip-chip LEDs were compared. It was found that with the use of exotic material layers and modified Au–Sn bumps, the device heat dissipation capabilities were significantly enhanced. As a result, 70% enhancement in the optical output power of 250 µm disk LEDs (λ ∼ 287 nm) was observed with a better device thermal stability.

An Evaluation of Heat Dissipation Capability of Slow-Wave Structures

A new experimental evaluation method is presented for evaluating and comparing the heat dissipation capability of the slow-wave structures (SWSs) of helix traveling-wave tube (TWT). The practicality and simplicity of the evaluation method have been validated through experimental tests in some components with different assembling and processing methods, such as the cold stuffing, the heat shrink, the hot insertion, the molybdenum wrapping methods, and so on. Experimental results demonstrate that our evaluation method can be used effectively and conveniently to estimate and compare the thermal dissipation capability of the SWSs of the helix TWT

High brightness GaN vertical light emitting diodes on metal alloyed substrate for general lighting application

The characteristics of the GaN based Vertical Light Emitting Diodes on Metal alloyed Substrate (VLEDMS) were investigated. The VLEDMS exhibits very good current-voltage behaviour with low average dynamic resistance of 0.7 Ω and low operated voltage of 3.2 V at 350 mA. High current operation up to 2000 mA in continuous mode was demonstrated without any performance deterioration. The high thermal conductivity of metal alloyed substrate demonstrated excellent heat dissipation capability. Chip scaling without efficiency loss shows a unique property of VLEDMS and enabled to enlarge up to 2 mm × 2 mm chip size. A light output efficiency of 70 lm/W or better was achieved in single chip or multiple chips package. Coupled with mass production ability, VLEDMS is very suitable for general lighting application. (© 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim)

High brightness GaN vertical light emitting diodes on metal alloyed substrate for general lighting application

The characteristics of the GaN-based vertical light emitting diodes on metal alloyed substrate (VLEDMS) were investigated. The VLEDMS exhibits very good current–voltage behaviour with low serial dynamic resistance of 0.7 Ω and low operated voltage of 3.2 V at 350 mA. High current operation up to 3 A in continuous mode was demonstrated without any performance deterioration. The high thermal conductivity of metal alloyed substrate exhibits excellent heat dissipation capability. Chip scaling without efficiency loss shows a unique property of VLEDMS. A light output efficiency of 70 lumens/W or better was achieved in single chip or multiple chips package. Coupled with good reliability and mass production ability, VLEDMS is very suitable for general lighting application.

Optimization of a space radiator with energy storage

A simple model for heat transfer from a space radiator with latent heat thermal energy storage has been developed. For a given heat storage and dissipation capability, analytical results have been obtained for the optimum geometry of a radiator panel/fin based on a minimum mass criterion. Numerical results for a typical configuration show that mass reductions of 20–25% or more can be achieved for pulsed heating loads of durations of the order of 1 h or less. At the same time, the radiator sizes can be reduced by a factor of 4 or more. The results also suggest that the benefits of energy storage will be higher for operating conditions with lower heat dissipation rates.

Experimental study of the heat sink assembly with oblique straight fins

Vertical planar fin array is used in conventional heat sinks for CPU cooling to create extended surfaces for the enhancement of heat transfer. The cross-sectional areas within this parallel fin array always remain unchanged. However, as the number of fins increases to meet the challenging task generated from a high-speed CPU, the area of the flow channel decreases and the flow resistance thus grows considerably in spite of the increase in the heat transfer areas. In this work, to reduce this undesired blockage on the incoming airflow, oblique planar fins were utilized in heat sink assemblies to improve the overall performance of the heat sink assembly. In addition, a high-pressure, axial-flow fan (70 × 70 × 15 mm 3) was designed to incorporate with the vertical and oblique fins to form complete heat sink assembly units. The prototypes were fabricated using a CNC machine. Besides, a CPU simulator for an Intel Pentium IV processor was constructed and utilized to execute the thermal-performance measurement for comparison with the reference fan. The amount of power consumption the CPU simulator used was 82 W. Experimental results indicate that, due to the larger surface area and accelerating flow between the fins, the heat sink assembly with oblique planar fins shows better performance than that with typical vertical fins. For the case of high-pressure fan operating at 2000 rpm, the extra cooling effect could result in a reduction of 6 °C on the CPU case temperature by introducing oblique fins. Furthermore, even though the flow-rate of the designed fan operating at 4000 rpm was only 31.4 CFM, which was 5.7 CFM lower than the reference fan, its static pressure and static efficiency were, respectively, 1.3 mm Aq and 3.8% higher than those of the reference fan. At low flow rates, the designed fan yields a better heat-dissipation capability than the reference fan. Obviously, the static pressure of cooling fan is the dominant factor for a high-resistance heat sink assembly.

Thermal Transport in High Porosity Cellular Metal Foams

The heat dissipation capability of highly porous cellular metal foams with open cells subject to forced air convection is studied using a combined experimental and analytical approach. The cellular morphologies of six FeCrAlY (an iron-based alloy) foams and six copper alloy foams with a range of pore sizes and porosities are quantified with the scanning electronic microscope and image analysis. Experimental measurements on pressure drop and heat transfer for copper foams are carried out. A numerical model for forced convection across opencelled metal foams is subsequently developed, and the predictions are compared with those measured. Reasonably good agreement with test data is obtained, given the complexity of the cellular foam morphology and the associated momentum/energy transport. The results show that cell size has a more significant effect on the overall heat transfer than porosity. An optimal porosity is obtained based on the balance between pressure drop and overall heat transfer, which decreases as the Reynolds number is increased. Nomenclature ˜ a = specific surface area per unit volume C f = heat capacity of fluid C I = inertial coefficient d f = diameter of cell edge ligament d p = pore size FI = inertial variable f = friction factor H = channel height h sf = interfacial heat transfer coefficient

Thermal design modifications to improve firing frequency of back shooting inkjet printhead

This paper presents design modifications of a backshooter type inkjet printhead, DomeJet, to enhance the firing frequency. DomeJet is a thermally driven monolithic inkjet printhead that consists of dome-shaped ink chambers and omega-shaped heaters over the chambers. Since it is monolithic, the manufacturing is very cost effective. As the firing frequency increases, the residual heat accumulated in heater layer may cause unstable droplet ejection. Thus, an efficient heat passage is necessary to increase the firing frequency. Contrary to a roofshooter architecture printhead, a backshooter type printhead has disadvantage in heat release since the heater is confined to a thin film which has low thermal conductivity. In this study, four models are proposed and studied. Compared to the basic model of DomeJet, all the proposed models are found to improve heat dissipation capability. The best performing model is found to operate at a frequency of 40 kHz, a much higher frequency than 15 kHz of the basic model.